Telecom Tensions: Internet Service Providers and Public Policy in Canada 9780228007920

Citation preview

Acknowledgments

T ELECOM T ENSIONS

i

preface

TELECOM TENSIONS Internet Service Providers and Public Policy in Canada

MIKE ZAJKO

McGill-Queen’s University Press Montreal & Kingston • London • Chicago

iii

iv

preface

© McGill-Queen’s University Press 2021 isbn 978-0-2280-0588-9 (cloth) isbn 978-0-2280-0589-6 (paper) isbn 978-0-2280-0792-0 (epdf) isbn 978-0-2280-0793-7 (epub) Legal deposit second quarter 2021 Bibliothèque nationale du Québec Printed in Canada on acid-free paper that is 100% ancient forest free (100% post-consumer recycled), processed chlorine free This book has been published with the help of a grant from the Canadian Federation for the Humanities and Social Sciences, through the Awards to Scholarly Publications Program, using funds provided by the Social Sciences and Humanities Research Council of Canada.

We acknowledge the support of the Canada Council for the Arts. Nous remercions le Conseil des arts du Canada de son soutien.

Library and Archives Canada Cataloguing in Publication Title: Telecom tensions : internet service providers and public policy in Canada / Mike Zajko. Names: Zajko, Mike, 1983– author. Description: Includes bibliographical references and index. Identifiers: Canadiana (print) 20200397966 | Canadiana (ebook) 20200398059 | isbn 9780228005889 (cloth) | isbn 9780228005896 (paper) | isbn 9780228007920 (epdf) | isbn 9780228007937 (epub) Subjects: lcsh: Internet service providers—Canada. | lcsh: Internet service providers—Government policy—Canada. Classification: lcc he7583.c3 z35 2021 | ddc 384.3/30971—dc23

This book was typeset by True to Type in 10.5/13 Sabon

preface

v

Contents

Figures vii Acknowledgments ix Abbreviations

xi

Introduction 3 1 Scenes from Canada’s Internet 26 2 The History and Political Economy of Canadian Intermediaries 41 3 Connectivity for the Public Good 58 4 Chasing Competition 81 5 Common Carriage, Net Neutrality, and Bad Traffic 104 6 Internet Service Providers as Privacy Custodians 123 7 Internet Service Providers as Security Partners 145 Conclusion 165 Notes 177 References 193 Index 227

vi

Contents

Contents

Figures

1.1 Interconnecting isps 32 3.1 Open-Access Municipal and Middle-Mile Networks 76 4.1 Competitive Dependency 93

vii

viii

François Crépeau

Foreword

ix

Acknowledgments

This book was enabled by the support of many sources. First and foremost, my family has kept me going over the long stretch. Thank you for your patience, and for making the journey a joyful one. Most of the research involved in the production of this work was carried out while I was at the University of Alberta, where this manuscript benefited from the guidance of faculty including Kevin Haggerty, Michael McNally, and Geoffrey Rockwell, and where I learned a great deal from other graduate students. This work draws on research supported by the Social Sciences and Humanities Research Council; additional funding was also provided through the University of Alberta and the University of British Columbia. Chapter 6, “Internet Service Providers as Privacy Custodians,” was adapted from an article with the same name published in the Canadian Journal of Law & Society in 2018, where it benefited from the comments of anonymous reviewers. I want to thank everyone involved in helping me through the publishing process at mqup, including Jonathan Crago and the anonymous readers, for their detailed attention. I appreciate the help of public servants carrying out the thankless work of processing access-to-information requests, without which some of these topics would have been impossible to write about. I would also like to thank all of the people who have taught me about telecom and public policy in Canada. This includes the telecom reporters cited in these pages, whose journalism has been indispensable in documenting industry and regulatory developments. The work of past and current Canadian communications scholars and telecom policy analysts has provided an essential foundation for this book. Thank you to everyone who enabled me to

x

Acknowledgments

attend various telecom-related events, and who took the time to share their thoughts and experiences. I learned many invaluable lessons through discussions in conference halls and informal telephone calls, but am especially grateful to those who invited me into their offices and data centres, and who participated in interviews by entertaining my questions. Finally, this book was made possible by the internet – thank you to the people who helped build it, and who quietly keep it working.

Acknowledgments

xi

Abbreviations

brc catv

Board of Railway Commissioners cable television (originally Community Antenna Television) ccaice Canadian Coalition Against Internet Child Exploitation cccs Canadian Centre for Cyber Security cdn Content Delivery Network cnoc Canadian Network Operators Consortium crtc Canadian Radio-television and Telecommunications Commission cse/csec Communications Security Establishment csis Canadian Security Intelligence Service cstac Canadian Security Telecommunications Advisory Committee ctc Canadian Transport Commission ctcp Canadian Telecommunications Cyber Protection Working Group dpi Deep Packet Inspection icann Internet Corporation for Assigned Names and Numbers ietf Internet Engineering Task Force iisp (In)dependent Internet Service Provider isp Internet Service Provider ised Innovation, Science and Economic Development Canada (formerly Industry Canada) itmp Internet Traffic Management Practice ixp Internet Exchange Point nanog the North American Network Operators’ Group

xii

osp pipeda ren

Abbreviations

Online Service Provider Personal Information Protection and Electronic Documents Act Research and Education Network

The Backstory

T ELECOM T ENSIONS

1

2

On the House

Introduction

Before carrying out the research for this book, my mental image of the internet was a box with blinking lights that needed an occasional reset. Then, in the summer of 2013, I made my first visit as a researcher to an isp’s office and its modest data centre. I was there to meet Graham Fletcher, the company’s owner and an Alberta internet pioneer, who responded to my questions by sharing stories of his many years of experience dealing with governments, suppliers, and competitors. He showed me the cooled racks of servers hosting his clients’ data, and the building’s safeguards against any human or natural events that might threaten it. Then we went outside, where the building connects to the rest of the internet. One of these connections was a black cable hanging below all the other wires on the utility poles. I recognized the distinct bulges (“splice enclosures”) along the cable, which I had seen countless times around my neighbourhood, but which were one of those inscrutable infrastructural components that never registered in my consciousness. With a bit of explanation, what was formerly just a common strand became legible as one intermediary’s property – a link in a network connecting networks to other networks. This hardware of communication, I would learn time and again, is also the product of human relationships, organizational dynamics, and regulatory regimes. While the internet has become our society’s connective tissue, there is a social sphere of remarkable people and practices holding it all together. In addition to being a technical marvel, the internet has a political economy,1 a contentious history, and an uncertain future.

4

Telecom Tensions

INTERMEDIAR IES AND INTERNET GOVERNANCE

Societies develop through human communication, but humans are not limited to a particular communicative repertoire. From storytelling and stone carvings, to radio and internet, we express, share, inform, and govern through the messages we communicate to each other. Each technique or medium of communication enables distinct social forms to develop, and technologies cannot be studied independently of their uses. When linked together as socio-technical systems, communications technologies enable ideas to spread far beyond their point of origin, permit governments to organize their affairs, and allow producers and consumers to come together in global trade networks. Writing a message preserves it across time, while a messenger can carry it to a distant point, where it can be received and acted upon. Messenger systems are an important foundation for this analysis, because they insert intermediaries into the communication process. The Romans and Inca utilized sophisticated messenger systems to coordinate their empires, ensuring that a message from the central government would pass faithfully from hand to hand (or mouth to ear) over thousands of kilometers.2 Today, we rely on digital messengers to coordinate our lives. A message can have a sender and a recipient, with multiple intermediaries in between. Intermediaries allow us to communicate across time and vast distances, from one to a multitude of points, from the multitude to the centre, or through a decentralized tangle of relationships. The simplest intermediaries are relays, which transport a message without altering it. The nightmare intermediary is represented by the “man-in-the-middle attack,” feared by information security practitioners, where the intermediary is hidden from both ends of a communication. A malicious intermediary can read messages, exploit the information they contain, and modify communications. Between the faithful relay and the attacker-in-the-middle is a range of ways intermediaries can affect communications, making it worthwhile to consider who these entities are and what they do. These considerations are more crucial than ever before, because we depend on intermediaries for all sorts of vital and mundane tasks. The trend towards increasingly powerful intermediaries has been developing over the past two centuries, as telecommunications networks have come into widespread use and branched across the globe. Therefore, the starting point of this book is the simple observation that

Introduction

5

more and more of our interactions with each other and the world around us occur through an intermediary. It is true that the world has grown more connected, but those connections must be built, maintained, and modulated. Consider how many people and organizations you interacted with directly today. How many people did you speak with face-to-face, how many purchases did you make inside a store, how many services did you access by transporting your body and waiting in line at some office? Now ask the same of your indirect interactions. In other words, how many times did an intermediary help you communicate across some distance? How many times did you text, call, or video chat? How many online purchases did you make? How many online services did you access? I would guess that your number of indirect interactions far exceeds your direct interactions, but whatever the ratio, it has been growing steadily since the public adoption of the internet in the 1990s, and has recently been accelerated by the covid-19 pandemic. Activities that used to require the “co-presence” of individuals can now be carried out remotely. Many government services can only be accessed online, and digital platforms give producers new ways to connect with consumers. These may feel like direct relationships, and they are often described as such: “direct marketing,” “direct-to-fan,” “peer-to-peer” communications. But these terms exist because our messenger systems operate in the background. Like all infrastructure, when the internet works we do not give it a second thought, freeing us to focus on the activities it supports. And so this book analyzes governance through the most fundamental and invisible intermediary of our times – the internet service provider, or isp. The opacity of the world behind our screens, and its growing impact on our daily lives, makes it critically important as a domain of study. We need to understand how this digital world is governed in order to understand how it governs us. Intermediaries are the central concern of this analysis, and the term refers to whatever (or whoever) sits between the parties involved in a communication. But intermediaries also govern our digital flows – they mediate communications. This why Jonathan Zittrain (2003) characterized intermediaries as “points of control,” through which one can affect communications, identify individuals, and block access. isps may only exercise a limited range of these powers, but other actors attempt to enroll or compel them into new governmental strategies. Intermediaries are vital to contemporary society not just because they provide a connective layer

6

Telecom Tensions

for social interaction, but also because they exercise control over how human communication takes place. The understanding of “governance” that this work is based on requires some explanation, given the different meanings with which the concept has become associated (Bevir 2011). Most importantly, social and political theorists have used it to move away from a statecentric idea of “government,” and towards a more pluralistic, heterogeneous understanding of “who governs” in society. If we look at the sorts of organizations or collective actors that govern our lives, or which carry out policies that affect the public, we will find that these encompass a variety of state and non-state actors, often working in relation to one another. For instance, commercial isps make and revise policies that apply to our experience of the internet, governing how different kinds of data are treated and what behaviours are forbidden online, as well as the consequences of violating these policies. This kind of internet governance is certainly based on the private interests of isps to operate as profitable businesses, but your isp’s “terms of service” also refer to compliance with laws, law enforcement agencies, and courts. The governing practices of state and non-state actors often depend on one another. Scholars of governance have frequently claimed that there has been a historical shift, particularly since the late twentieth century, away from state-centric forms of rule and towards more networked, decentralized forms of power (Stoker 1998). This may or may not be true in some given context, but I do not base my understanding of governance on the political transformations of the 1980s. Various kinds of societies have been governed in past centuries through “a multiplicity of sites and forces beyond the direct control of formal political authorities” (Dean 2007, 52). Certainly, there have been historical changes in modes of governance, but the conceptual shift away from state-centric forms of power denotes a change in perspective rather than empirical reality. In Canada, telecom networks have historically been governed through complex relationships of state and non-state actors. Since the 1990s, state agencies have stepped back from some kinds of control (such as setting prices), but have become more involved in a wide range of others, as covered in the following chapters. Internet governance scholars focus on the role of non-state actors and private organizations in the internet’s administration, given the historical roles that these have played (Brousseau, Marzouki, and Méadel 2012; DeNardis 2009; Mueller 2004). They also recognize how

Introduction

7

state agencies shaped the internet’s development, and how states assert their sovereignty to shape internet policy (Goldsmith and Wu 2006; Mueller 2010). While internet governance scholarship has typically treated the internet as an object of governance, this book expands the topic of internet governance beyond the traditional realm of networks, standards, and protocols, to society itself. Rather than discussing governance of the internet, I will analyze governance through the internet. Specifically, how have the most fundamental digital intermediaries – internet service providers (isps) – become instruments and agents of public policy, and to what effect? What political desires can these organizations help to achieve? There is a long and growing list of answers to this question, but I will address those that have proven particularly relevant to Canadian telecom politics over the past twenty years. While my focus will remain largely on Canada, the processes and tensions covered in this book have also played out in many other countries, given that much of the world transitioned from a monopoly-based telecom industry and into a liberalized political economy at roughly the same time, followed closely by the proliferation of internet access. Canadian telecom policy is unique in many ways, such as the details of its mandated access regime (chapter 4) or its development of a network neutrality framework (chapter 5), but the principles underpinning Canadian telecom policy are often shared internationally. For instance, common carriage is based in English common law (Noam 1994), facilities-based and service-based competition are often pursued together in other countries (oecd 2012), and privacy governance is based on internationally developed principles (oecd 1980). Finally, the trend of governing through intermediaries is a broad and international one; intermediaries around the world have been gaining in power, capacities, and political value as instruments of governance (DeNardis 2014; oecd 2011). My focus in this book is on the instrumental use of isps to advance certain forms of public policy, as well as the agency that isps exercise in becoming public policy actors serving some notion of the public good. I am interested in how the organizations that build telecom infrastructure and allow us to connect at home have become instruments of order, guardians of morality, enablers of self-determination and economic prosperity. What rationalities, techniques, and regulatory regimes determine internet governance in Canada? Asking these questions draws us into contact with the broad field of governmentality studies, inspired by the work of Michel Foucault. This book is an

8

Telecom Tensions

account of a particular “milieu,” as “the medium of an action and the element in which it circulates” (Foucault 2007, 21). This also makes the milieu a target of intervention, for what Foucault would describe as techniques of governmentality. The relevant milieu for this analysis is the medium through which data packets circulate, and the interventions that target these flows and their conduits. In his governmentality lectures, Foucault (2007) discussed interventions directed at the population, beginning with the circulation of goods, people, and diseases in eighteenth-century towns and cities. As early urban planners shaped these flows in response to specific problems, such as scarcity, disease, and the exercise of sovereignty, today’s agents of internet governance shape digital flows and configure infrastructures to address problems of connectivity, competition, neutrality, privacy, and security. Rather than cohering into an unfolding governmental logic, these problems become objects of struggle and the clash of competing desires. Governmentality scholars sometimes analyze the world as an anonymous landscape of processes, technologies, discourses, strategies, and identities. However, this book is populated with actors that make decisions and pursue objectives. Also contrary to popular readings of Foucault as a theorist who “decentered” the state in the study of government (Dean and Villadsen 2016), sovereignty maintains a distinctive status in this analysis, hierarchizing telecom governance and demarcating loci of supreme power. Certainly, the internet has challenged some forms of sovereign control through its “debordering perforation” (Bratton 2015, 6), but sovereignty has also been reinscribed through territorial claims on the networks constituting the internet (Bratton 2015; Goldsmith and Wu 2006; Mueller 2017). isps, as operators of material infrastructure rooted in particular territories, are particularly susceptible to claims of sovereignty from state regulators. While websites are relatively portable, able to relocate to wherever servers are available, isps are grounded in fixed infrastructure, operating with the approval of territorial governments where their assets are based. Infrastructures for transporting data packets have much in common with the utilities that transport electricity, gas, and water, over which state agencies have the final say and exercise ultimate authority. In the rare instance that an isp challenges this sovereign authority, it confronts the most naked expression of state power: physical force, the seizure of property, even imprisonment.3 Sovereign claims over internet infrastructure are one important reason why intermediaries are expanding and accumulating roles in soci-

Introduction

9

ety, as new responsibilities are imposed by state agencies. But Canadian telecom policy tends to be the outcome of a contentious process involving competing actors, sometimes calling the sovereign jurisdiction of particular bodies (namely the crtc, as well as provincial and municipal governments) into question. One significant trend in these struggles has been a growing public involvement in previously obscure and esoteric issues (see Geist 2016a), elevating internet policy to the highest level of political visibility. I will consider some of the major political issues in internet governance that have been contested in Canada since the mid-1990s, surfacing tensions around the proper role of intermediaries. These include conflicts over isps’ roles and responsibilities in aiding policing and surveillance, cyber security, and copyright enforcement, and the sorts of interventions isps can make while remaining “common carriers” or practising “net neutrality.” There are also continuous regulatory struggles about the responsibilities that different classes of isps have to one another, efforts to mold the Canadian government’s vision of a “competitive” telecom industry into existence, and the persistent challenge of extending broadband connectivity beyond Canada’s urban centres. INTERMEDIATION AND ROLE CONFLICT

Internet technologies have often been characterized as a disintermediating force, “disrupting” and circumventing traditional gatekeepers. But alongside this trend of disintermediation, we see the simultaneous ascendance of new intermediaries and their growing significance in our lives. As digital flows and identifiers become increasingly consequential, so does control over them. Intermediaries become privileged sites of governance and recurring targets of intervention. This process is called “intermediation,” referring to the growing concentration of power and capacity in the internet’s “middle,” where isps reside. Google and Facebook are the most visible manifestations of intermediation in North America, but isps occupy an essential position in the internet’s topography and are subject to more localized forms of regulatory control. As a result, isps have become ideal and idealized instruments of governance for the twenty-first century. They are seen as the means through which all of the various dreams associated with connectivity can be achieved. Since isps operate gateways to the rest of the internet, they can be employed as gatekeepers. If certain kinds of traffic need to be blocked

10

Telecom Tensions

or monitored, this can be done by enlisting the isp. If an individual internet user needs to be identified, an isp is in a position to link their online activities with a physical location and a subscriber’s account. If there is not enough internet access, then isps can be mandated to open their gates wider, build more infrastructure, or offer certain services at an affordable rate. All of this has made isps attractive as instruments of public policy, particularly since governing the internet by other means (such as through its “multistakeholder” organizations, or through direct control over users’ computers) is so unwieldy. Rather than dealing with the jurisdictional headaches of internet governance, trying to bring distant actors in line with domestic policies, or bringing the internet under direct state control, governments can use isps to achieve their political aspirations. As state agencies and other actors come to see them as instruments for governing society, intermediaries themselves can choose to govern users and data packets in new ways. An isp’s core business remains moving data across its network, but these organizations now also provide a larger list of services, including video programming, cyber security, safety education, home automation, and advertising. isps are moving further and further away from the ideal of a pure intermediary, one that simply moves data packets from one edge of the network to the other, becoming increasingly capable and vital stewards of society’s data flows. As their centrality to our daily lives intensifies, and intermediaries’ social roles accumulate, they take on more of the responsibilities that go along with these positions – sometimes willingly, and sometimes as an imposition. Growing expectations on an isp’s conduct eventually overlap and contradict one another. “Role conflict” results when an intermediary occupies multiple roles, and is therefore subject to contradictory expectations. When this happens, one role may become predominant, elevating some responsibilities at the expense of others, or the organization may attempt a more balanced distribution of compromises. In sociology, role theory developed to address the circumstances of individuals, such as the strain experienced when women are subjected to contradictory work and family expectations (Stryker and Macke 1978). While organizations such as isps do not experience strain and conflict in an embodied sense, they share the human characteristic of being decisionmaking actors,4 and their decisions are shaped by the expectations of others. Intermediaries occupy social roles that cannot be reduced to their human employees, managers, or executives. The roles of inter-

Introduction

11

mediaries in society (as economic facilitators, public utilities, surveillance agents, privacy custodians, marketers, and mandated wholesalers) are positions in social relationships. When tied to public policy, these roles are solutions to problems of governance, shaping an intermediary’s conduct through attendant responsibilities.5 Some roles are self-professed and chosen, such as when an isp chooses to address economic inequality and work towards more equitable connectivity. Other times, an isp will act in a role mandated by law, such as that of privacy custodian, surveillance agent, or wholesale provider, but may choose to do so in a way that exceeds the organization’s legal responsibilities. As these roles expand or multiply, so does the potential for conflict between them. In these conflicts, intermediaries are more than instruments of policy being pushed and pulled in different directions. They are actors, which, even when enrolled by others, demonstrate varying degrees of autonomy, resistance, and discretion. Intermediaries navigate role conflicts individually or collaboratively, choosing a path through competing expectations. Different organizations facing similar role conflicts have chosen different paths, demonstrating the discretion with which they can resolve these contradictions. Unlike individuals, organizations can be internally divided into different agencies operating at different levels. Contradictory roles can be allocated to these different parts of an organization (Geser 1992), such as when an isp’s wholesale division cooperates with the competitors of the retail division (see chapter 4). Role conflicts may therefore appear at one level of analysis (the organization as a whole), but this can be the result of different parts of an organization playing contradictory roles. This assumption of collective or corporate agency is another way that this book differs from other analyses inspired by governmentality, since Foucault spent much of his career analyzing how subjects are produced through different kinds of practices.6 Foucauldian analyses therefore often problematize actors, showing how subjects (including individuals and collectives) are constituted. But a theory of agency is hard to do without in social theory, and ideas about what counts as an actor are often implicit in contemporary studies of governmentality. My treatment of agency is broadly consistent with the theory of “nodal governance,” with intermediaries acting as “nodes”7 that govern the conduct of users and data traffic, while also being “objects of governance” that can be enrolled or mobilized by others (Wood and Shearing 2007, 27). isps work through alliances to influence competi-

12

Telecom Tensions

tors and regulators, they make policy decisions, and they govern users, employees, and data traffic. isps are also objects of political contests to use these organizations as instruments of public policy, as other actors work to shape intermediaries’ conduct. Holding this dual view can be challenging when our language presumes a distinction between subjects and objects, but sociologists have the related challenge of thinking past a decades-long struggle over the relationship between agency and structure. In sociology, agency has typically been a property of the individual and structure is identified with institutions and social systems. When the chief distinction is between agency and structure, it becomes difficult to conceptualize a collective actor that can also be a social structure. Disciplines that deal more specifically with organizations, such as organizational studies, international relations, and foreign policy analysis, have more thoroughly addressed the topic of collective or corporate agency,8 but the problem of theorizing the relationship of individual actors to organized groups and institutions is a persistent one. Intermediaries are more than organizations that structure the lives of employees, or determine the experiences of users. They have institutionalized means of making collective decisions, and act with purpose to achieve particular ends. Some organizations and their decisions are more “centred” (Domingues 1995), and closely resemble an individual goal-oriented actor. This can happen when a small isp is personally managed by its president or ceo, or when a large corporation makes high-level decisions that hierarchically determine the actions of individual employees. Depending on where one is looking, the same organization can also appear more decentred or heterogeneous, such as when individual employees are given wide discretion in their sales or technical support jobs, or when different divisions of an organization pursue contradictory goals. This book deals primarily with centred forms of decision-making, where intermediaries adopt specific legal positions, implement policies, and orient the organization towards particular ends. Examples include intermediaries developing their collective positions in regulatory conflicts before the Canadian Radio-television and Telecommunications Commission (crtc), or the decisions of executives to expand the organization’s scope to encompass new roles. In such cases it is appropriate to treat a governing node such as an isp as a collective actor, even though in other circumstances the same organization can appear as a diverse and contradictory patchwork of agencies

Introduction

13

(Abdelnour, Hasselbladh, and Kallinikos 2017). With this in mind, I want to return to the central argument of the book and exemplify it with a current case of accumulating roles and role conflict in Canadian telecom. Know Your Role If we are all actors occupying roles in social relationships, then what is your role as a consumer of connectivity, or as a customer of an isp? Most fundamentally, your role is to pay your bill on time; this is what is expected of you. The isp’s role is to provide what you pay for, and as long as the organization meets this expectation they are likely to keep you as a satisfied customer. However, what happens when an isp also plays other roles that become relevant to this relationship? This book will consider a number of such situations, but let us begin with a recent example that could place our role as customers into question. Mobile phones are one of the most intimate communications technologies we have – multi-sensory networked computers that we keep close at hand wherever we go. Many of us are familiar with the phrase “if you’re not paying for it, you’re the product,”9 and so we might assume that paying our monthly mobile bill would inoculate us against other commercial exchanges. However, data about mobile phones and their users is increasingly valuable, and so many us are now playing a dual role: buying connectivity and generating value. Meanwhile, our intermediaries are also playing the dual role of serving two sets of customers: one that purchases connectivity, and one that purchases information about us. In 2018, a news story drew attention to the fact that Canada’s largest mobile phone service providers, through their joint ownership of a company called EnStream, also had another set of customers with a different set of expectations (Braga 2018). EnStream offers its customers information about Canadian mobile phones and their users, helping companies verify the identity of a person on a customer support call, or determine a caller’s location. In theory, the two roles played by these service providers need not be contradictory. Bell, Rogers, and telus do provide mobile customers their connectivity, while also giving EnStream customers what they are interested in. As EnStream and its owners emphasize, these two roles can also be complementary, and a favourite example is that of a phone subscriber who wants to share their location with a roadside assistance company. But

14

Telecom Tensions

it is not difficult to imagine a situation where someone would want to pay for location information with no benefit to the phone’s user, or in a way that would do them harm. This is what has happened in the United States, where location information sold by phone companies through LocationSmart (an EnStream partner) was used by a police officer to spy on colleagues and a political rival (Murphy 2018), and where a black market for location information has existed for years (Cox 2019). Canada has not seen a similar scandal, but Canadian companies have had to confront the same role conflict since choosing to become suppliers of location services. They resolve this conflict by insisting Canadian subscribers give consent anytime EnStream is used to share their personal information. This is, after all, a legal requirement. But laws can be broken, and when intermediaries try to profit from personal information while also protecting privacy, conflicting priorities can result in legally dubious behaviour (chapter 6; see also fcc 2020; Office of the Privacy Commissioner 2015b). Regardless of how they choose to navigate it, a role conflict remains for intermediaries that choose to sell access to information about subscribers. The conflict can be resolved by the intermediaries strictly limiting the information trade in the interests of individual subscribers, or embracing more of the potential profits from their new role as information merchants. Intermediaries will rarely highlight these role conflicts and contradictory pressures, preferring to discuss mutual benefits and adherence to some singular law. This reinforces the need for independent analysis, and to that end this book addresses the most important role conflicts in Canadian telecom at a time when intermediaries’ roles are expanding. A NET WORK OF RELATIONSHIPS AND REGULATORY REGIMES

The following chapters dive into the world of connectivity beyond our “devices,” analyzing the institutional domain on the other side of our routers and modems. This is a world many of us have never wondered about, just as we are rarely curious about where our sewage goes or how the electricity grid is configured. Infrastructures become pressing issues for us when they break down, but otherwise they are unobtrusive. If we decide to follow the wires that lead from our routers and modems out of our homes and down the street to map how data flows through them, we will quickly encounter obstacles to

Introduction

15

our understanding. Not only is internet infrastructure typically hidden from view in underground conduits, locked cabinets, and nondescript windowless buildings, but this physical “layer” provides only the base for higher-level processes, such as the networking protocols that are used to control how data moves. Protocols help to solve very technical problems, such as how to translate information from one medium to another, or how two very different networks can exchange traffic with each other, but a technical education will only go so far in clarifying how the internet works. This is because while internet protocols do “govern” the data flowing into and out of our homes, and while some sort of material infrastructure is required for connectivity, regulations and relationships between organizations shape how protocols operate and how infrastructure is built. The internet has historically been free from some of the regulations that applied to other media, but it has never been ungoverned (Mathew 2016). Furthermore, the common definition of the internet as a “network of networks” highlights the importance of considering how these networks connect to each other – for example, whether one isp sees another as a partner or a competitor. The only extended internet outage I have ever experienced was the result of two competitors who were forced to partner with each other, rather than a technical problem. I moved into a building with a perfectly functional cable connection, but chose to become a customer of an “independent” isp that offered service cheaper than the massive “incumbent” cable company. The problem was that the independent isp does not own the physical infrastructure that extends through my neighbourhood, and must therefore lease access to these wires from their larger incumbent competitor. The incumbent, or infrastructure owner, would be happy to have me as a customer instead, but federal regulations require them to lease access to their network to smaller competitors (see chapter 4). In the end, my home went ten days without internet access because the incumbent isp that owns the network was in no hurry to connect a customer of a competing isp. When the technician eventually arrived, he did his job quickly and effectively, but it was clear that the technical service he was providing was limited to his employer’s requirements under Canada’s convoluted system of competition regulation. During this encounter, the technician was acting as an instrument of public policy, connecting a new customer due to regulatory obligation rather than the pursuit of corporate profits.

16

Telecom Tensions

Now that my home is connected to the internet, data packets flow smoothly in both directions, but this has as much to do with regulations and relationships as it does with properly built and maintained infrastructure. An isp can build an extensive network of wires throughout a city, funnelling all the neighbourhood traffic into larger and higher-capacity cables that eventually lead to a central office. But the network only joins the internet when it connects to all the other networks, which is a process determined by human relationships, business strategies, and regulation. In the end, there does need to be a physical connection between networks on the internet – a cable must ultimately be plugged in between them for the data to flow. This often happens in a building designed for the purpose, known as an internet exchange point (ixp) or “carrier hotel.” The locations of these facilities are not a secret,10 but they also rarely attract attention. There are few windows at a carrier hotel and security is typically tight. After being cleared through the access-control airlock or “mantrap” at the front door, one enters the first of several floors of humming hardware and whirring cooling fans, mounted in racks, arranged in aisles. A network that rents space in the building can extend a connection up from the basement and into their rack-mounted equipment. isps can then snake cables around the building to each other, but this is not simply a matter of finding a long enough cord and keeping the cables tidy. A human relationship must be established first. The human relationships that underpin the internet are often made and maintained through emails and online messages, but meeting in person has also been historically important. One venue where such relationships are formed is nanog (the North American Network Operators’ Group), which (until 2020) was held in a different conference venue every four months. The attendees are network operators from across North America, and while they may be employees of competing businesses, at nanog they deal mostly with issues of common interest. Some attendees are responsible for interconnecting their networks or “peering,” and can meet their peering counterparts at a scheduled session to shake hands, exchange information, and potentially make deals (see Norton 2010). If both parties agree on the terms, they can get to work on the technical issues of how to interconnect, collapsing the distance between their networks and establishing a new, permeable border at the internet’s centre. But a human relationship underpins this newly established network connection.

Introduction

17

This human relationship is crucial when an unexpected problem strikes one or both networks, and interconnection partners are quickly in touch with each other to understand what is happening and how to coordinate a response. However, when certain parts of the internet become “congested” or difficult to access, the underlying reason can be a breakdown of the business relationship between networks, rather than technical problems.11 Like most Canadians, the bulk of my internet traffic is streaming video, and keeping these videos playing smoothly depends on a good relationship between the video streaming service and isp. In the United States during 2014, some internet subscribers had difficulty watching Netflix because the company was refusing to pay isps to handle its traffic.12 Canadian isps avoided demanding these payments and treated connectivity with Netflix as a matter of mutual benefit (smooth streams make for happy customers), but regulations also shape considerations of how networks connect to each other to form the global internet (Meier-Hahn 2016). Some Canadian isps have in recent years offered their own video services, making Netflix their competitor. This can change the strategic calculus of conflict and cooperation, giving an isp a reason to neglect, degrade, or even block its connection with Netflix if it thinks this will drive customers towards the isp’s own offerings (see Higginbotham 2014b). One reason Canadian subscribers have not experienced this kind of disruption is a regulatory regime that prohibits isps from discriminating between competing services. Some wireless providers have tried to discriminate through mobile data pricing, effectively charging subscribers more to watch Netflix than their own video services, but this resulted in regulatory intervention and a strengthening of net neutrality policies, as discussed in chapter 5. Such examples demonstrate that while the internet is indeed a technical marvel, its topography and technical configuration are shaped and supported by human relationships (Mathew 2014; Meier-Hahn 2017a), business relationships (Higginbotham 2014a; Norton 2010), and government regulations (Goldsmith and Wu 2006; Meier-Hahn 2016). STUDYING INTERMEDIARIES

The internet’s inner workings contain many secrets – secret surveillance, interconnection contracts, information filed in confidence to the crtc – but in other ways they remain remarkably open. The Inter-

18

Telecom Tensions

net Engineering Task Force (ietf) is the best-known example of this open attitude (Oever and Moriarty 2012), but North American isps will also share knowledge on the nanog mailing list, and nanog presentations are posted to YouTube. There are a number of other telecom industry events that produce less public documentation, but are open to whoever wants to pay the registration fee. This is the world I immersed myself in to understand the issues playing out in Canadian telecom, by attending conferences,13 reading trade publications and regulatory documents, and interviewing internet professionals or other experts.14 The result was an “explanatory mapping” (Wood 2006) that familiarized me with the actors involved, how they relate to one another, and the kinds of knowledge and rationality underpinning internet governance in Canada. Before I began my research in 2012, I had been following the major conflicts in internet policy that had erupted into the Canadian news media, over copyright, surveillance, and usage-based billing. It was only once I started to attend industry events, read industry publications, follow regulatory proceedings, and speak with individuals involved in Canadian telecom that I came to appreciate more mundane pressures and role conflicts. These included the constant tensions between incumbents and their smaller competitors, the coexistence of intermediaries with very different governing rationalities within a single “market,” and the issues around providing connectivity as a public good. Observing Canadian telecom events allowed me to see which concerns have remained relatively constant throughout the recent period (mandated access to facilities and associated costs), those that have faded in and out of salience in relation to federal government actions (lawful access, copyright), and events that had wide-ranging impact throughout the industry (Snowden’s effect on privacy and surveillance discussions in 2013). Based on the themes that arose from my initial mapping, I have organized the topics of this book into three parts. The first part (chapters 1–2) sets the stage by showing how the political economy of telecom developed in Canada, and visits some of the hidden places occupied by the internet. The second (chapters 3–4) addresses the question of what connectivity means and how best to provide it. The third part (chapters 5–7) deals with how intermediaries govern internet traffic and data, given the problems posed by net neutrality, privacy, copyright, and security. The data used for these chapters varied depending on what was ultimately available to me; I utilized documents, record-

Introduction

19

ings, and transcripts where these were public, and in other cases I sought to obtain my own data. Government agencies have been a rich source of publicly available documents, and documents obtained from federal agencies through access to information (ati) requests provided valuable insight into government policy-making, the positions taken by different actors, and the processes through which governing nodes are coordinated. I reviewed tens of thousands of pages from approximately one hundred relevant ati files. These were most helpful in documenting processes that are normally hidden from public view, namely lawful access and cyber security. However, large areas of Canadian telecom remain poorly documented, including the topography and history of networks. Interviews (and informal discussions) were often necessary simply to establish basic facts about current and historical events. These expert interviews allowed me to fill gaps in documentary sources, and to gain a deeper understanding of my selected topics and themes. I am deeply grateful to all those who took the time to share their knowledge. WHO ARE THE ACTORS ?

Intermediaries I have repeatedly referred to both intermediaries and isps as being central to my analysis. It is important to clarify what I mean with these terms, since others have used different and inconsistent definitions. In this book, an intermediary is any entity that is positioned between two ends of a communication, occupying an intermediate position between users and internet services. This includes providers of internet connectivity such as isps, as well as online service providers (osps) that play a connective role (social networks that mediate communications between users; search engines that connect users to an organized Web). My primary focus is on providers of internet connectivity, but today’s isps often have a broader role that encompasses osp functions, such as providing content, webmail, or hosting. Of the institutional actors described in this section, the category of intermediaries is the broadest (see Winseck 2015a), and the term has been adopted in internet governance discussions largely because distinctions between different classes of service providers (such as those merely providing internet access and those providing additional services) have become blurred. At the same time, many of the diverse institutions categorized as inter-

20

Telecom Tensions

mediaries are subject to similar legal requirements or expectations on conduct (Edwards 2011). What follows is a list of some of the kinds of intermediaries referred to in this book. However, it is important to note that a single actor can play multiple roles, and be characterized in a number of ways. For instance, Shaw is an incumbent cableco isp that also acts as an osp. The choice of which role or characteristic is foregrounded in my analysis depends on which role is of particular interest. •

•

•

•

•

Cableco: A telecom company that developed out of a pre-internet cable television (co-axial cable or catv) company. Contemporary Canadian examples include Rogers, Shaw, and Vidéotron. FreeNet: An isp organized around a model of equitable connectivity: providing free, low-cost, or subsidized internet access to “members.” Incumbent: A telecom company that possessed facilities prior to liberalization. Incumbency is territorially based, so that telus is an incumbent in Alberta and British Columbia, but not in Ontario, where Bell is the incumbent telco. iisp – Independent Internet Service Provider: A non-incumbent isp, or an isp that originated following telecom liberalization. As with the “independent” telephone companies of the twentieth century, the term can be quite misleading. This is because iisps are often heavily dependent on larger isps (usually the local incumbent) for upstream connectivity. Therefore, a more appropriate label might be “(in)dependent” internet service provider, or simply iisp.15 isp – Internet Service Provider: An isp is an organization that provides users, clients, or subscribers with an internet connection. This connectivity can be provided through network facilities owned by the isp, or through leased access to facilities owned by another. isps have sometimes been differentiated into classes based on how central they are in the internet’s topography (the centre being occupied by Tier-1 isps or “backbone” providers). isps can also be differentiated depending on the clients/users that they serve, with long-distance (international) carriers serving territorially bounded isps, who in turn serve residential and business subscribers, or other sorts of users.

Introduction •

•

•

•

21

Municipal Network/Municipal isp: A publicly owned municipal network. This may be a “community network,” created by individuals organizing as a group to improve connectivity in their locale (ONet), or it may be the result of a municipal government finding a better way to meet its networking requirements (QNet; City of Calgary). osp – Online Service Provider: An organization that provides a service that is accessible online, often through a website (including Facebook and Google). ren – Research and Education Network: A network that provides connectivity to research and education institutions, such as universities, but which may also have a broader function in promoting entrepreneurship and innovation (see chapter 3). Typically, rens are publicly funded. In Canada, there are twelve provincial and territorial rens (also known as Optical Regional Advanced Networks, or orans), operated by institutions such as bcnet and Cybera. These are interconnected at the national level through the National Research and Education Network (nren), which is operated by canarie. Telco: A telecom company that developed out of a pre-internet telephone company. Contemporary Canadian examples include Bell, telus, and SaskTel. At the crtc, these companies have often been classified as Incumbent Local Exchange Carriers (ilecs). Government Agencies

Intermediaries govern our data flows, but who governs the intermediaries? There are controls that are internal to the organization (selfgovernance), shareholder or “member” control, internet governance organizations such as icann and Regional Internet Registries (rirs) that play limited but important roles, and industry associations that can influence (but not compel) member organizations. Finally, we must consider the various state and quasi-public agencies to which intermediaries are accountable, and which often exercise sovereign authority over a domain. It is this final category of governing actors which are principally important in a national context and which are the main focus of this book. At the federal level in Canada, the job of governing communications networks in the public interest is largely the responsibility of the

22

Telecom Tensions

crtc, particularly through its administration of the Telecommunications Act. The crtc is a quasi-judicial body whose Commissioners rule on a variety of issues brought to them, primarily by telecom companies taking issue with regulations or the behaviour of competitors. The national public interest in telecommunications is also decided by the elected federal government, with the federal Cabinet deciding budget priorities, and able to exercise its authority over the crtc (by issuing policy directions and orders-in-council).16 Other relevant federal agencies include Innovation, Science and Economic Development Canada (ised, formerly Industry Canada), which among its many responsibilities manages the wireless spectrum in Canada. isps can also be subject to decisions by the Copyright Board, which has the power to require payments to copyright owners, and the Competition Bureau, which regulates anti-competitive conduct. However, there are numerous other governing nodes in Canada that exercise meaningful control over internet intermediaries towards public ends – even if these publics are more circumscribed than that of the nation. A number of provinces have addressed the lack of federal broadband projects with their own provincial initiatives. Some municipalities and First Nations have taken an active role in deploying networks to serve their residents. There are also regional initiatives that have been undertaken when a number of communities have collaborated out of common interest. In Canada, internet governance has effectively become a concern for all levels of government. OUTLINE OF THE BOOK

Having outlined the theoretical and methodological basis for the study in this Introduction, the first part of the book (chapters 1 and 2) establishes the historical and material context for the topics that follow. The second part of the book deals with public policy debates over how to improve connectivity (chapter 3), and what it means to do so through a commitment to competitive market forces (chapter 4). The third part (chapters 5 through 7) addresses how isps constitute points of control over the data flowing through their networks, and the roles these organizations play in traffic management, copyright and law enforcement, national security, and privacy protection. In chapter 1, I consider some of the different ways of looking at what the internet is, including its materiality, its myths and imaginings, and the interpersonal work of internet governance. The chapter

Introduction

23

visits a crtc hearing, a carrier hotel, a telecom industry conference, and the rural Alberta town of Olds. While infrastructure is largely meant to stay out of sight, the internet’s “cloud” exists as hardware, distributed across fortified, temperature-controlled data centres, connected through cables stretching over our heads and under our streets. In contrast, ideas of internet infrastructure are often abstract, or even utopian. Our ideas about the internet are important because decisions are informed by stories and visions of what people would like to achieve. In Canada, important decisions about the internet are made in a variety of settings, including corporate boardrooms, quiet corners of conferences, regulators’ offices, and city council meetings. The topography of Canada’s internet infrastructure often follows the distribution of pre-internet “legacy” infrastructure, such as telephone and cable TV networks. Canada’s dominant telecom firms developed during the monopoly era of regulation, and this incumbency goes a long way to explaining their current success. Finally, much of the regulatory regime for Canadian telecom was developed to address pre-internet problems. Chapter 2 situates the political economy of internet access as the consequence of these historical developments, dealing with the continued legacy of monopoly-era incumbents and regulations meant to open the industry to market forces. Chapter 3 discusses how extending and deepening connectivity is an important public policy goal in itself, which some intermediaries have been given an explicit obligation to pursue. Access to broadband networks is widely understood to be vital to society, and this access is valued to enable other ends, such as economic development and education. Since the liberalization of telecom regulation in the 1990s, Canadian public policies to improve connectivity have been shaped by the federal government’s commitment to market forces and competition among private networks. In other words, improving connectivity is an important public policy priority in Canada, but the pursuit of this priority must align or contend with a governing rationality that privileges economic considerations and competition between private companies for consumers of services. The result is a role conflict when commercial intermediaries provide connectivity as a public good, and when public intermediaries operate in a predominantly commercial industry. Canadian isps have been called upon to provide greater connectivity, to meet demand with supply, and to do so according to market principles. But “the market” often fails to deliver what is expected of

24

Telecom Tensions

it, and chapter 4 deals with the consequences. Relations between isps have been regulated with the aim of steering the industry towards greater competitiveness. It was imagined that once the aim of this regulatory regime was achieved, the hand of government would fall away and relationships within the telecom industry would be determined by market forces. However, market forces have not complied with regulatory desires, and incumbents have been able to maintain their dominance in exchange for acting as the instruments of regulated competition. As a result of decisions made by the crtc, an elaborate regime mandates that dominant isps protect the viability of their smaller competitors, restricting incumbents’ competitive advantages. The role conflict that inevitably results from this regime is due to the fact that some isps are simultaneously expected to compete and cooperate, and that small isps are both customers and competitors of the large incumbents that dominate the internet’s “bottleneck” facilities. The last set of regulatory regimes discussed in this book addresses how intermediaries govern traffic passing through their networks, and the personal information that these organizations hold in their possession. isps’ ability to control digital flows or discriminate among traffic is regulated through the expectations applied to “common carriers” and the net neutrality framework that evolved from this basis. While some describe net neutrality as a foundational ethos of the internet, chapter 5 explains how Canadian net neutrality regulations developed to distinguish between kinds of discrimination that are acceptable and those that are to be prohibited. These prohibitions have been advanced to limit the power of giant, vertically integrated firms as these companies accumulate roles through the process of intermediation. However, net neutrality regulations also permit discrimination in numerous ways, legitimating common practices used by isps to sort and intervene in traffic flows, particularly when targeting various kinds of “bad” traffic. As service providers are entrusted to carry increasing volumes of highly sensitive information, their importance as privacy custodians has been brought into starker relief and explicitly recognized as a core responsibility. Chapter 6 traces how intermediaries’ privacy expectations developed, and argues that some isps are adopting a more positive orientation to this responsibility. This means that some isps are actively taking steps to advance privacy, rather than treating privacy protection as a set of limitations on conduct. However, commitments to privacy stewardship are often neutralized through contradictory legal

Introduction

25

obligations (such as mandated surveillance access) and are recurrently threatened by commercial pressures to monetize personal information. Finally, chapter 7 deals with isps’ security and surveillance responsibilities, first in struggles over lawful access, and more recently in the domain of cyber security. As with their privacy responsibilities, isps have historically exercised significant discretion in governing data towards these ends. However, there have been substantial and repeated efforts by the federal government to standardize isps’ conduct and expectations, thereby limiting their discretion in matters of security, public safety, and law enforcement. These include repeated versions of lawful access legislation, and cyber security collaboration or “partnerships” between industry and state agencies. For isps, securitization involves a significant role adjustment that moves them even further from the traditional neutrality of the “mere conduit.” Cyber security programs are a crucial frontier of telecom governance, transforming intermediaries away from the ideal of “dumb pipes” (that simply carry traffic) and towards that of intelligent networks that act upon data flows and protect against cyber threats. The Conclusion summarizes intermediation as the broader shift of power from the edges and into the middle of the network, even as internet technologies dismantle old hierarchies and enable peer relationships. Google and Facebook command our attention, but isps have become the only truly indispensable intermediaries. In addition to being indispensable, they are “multipotent” as governing nodes, and implicated in a growing range of social relationships. Rather than embracing intermediation head-on, and empowering isps to be our personal data shepherds and collective enablers, we would do well to fight for autonomy at the edges and endpoints. isps might be destined as society’s connective fabric, but they need not define how we make use of that fabric, nor lock us into peripheral dependence.

26

Telecom Tensions

1

Scenes from Canada’s Internet

The internet is made of glass. This glass is flexible and durable – thin strands bundled together and encased in a protective sheath, carrying pulses of light along fibre-optic paths that snake over and under our streets. The pulses are governed by protocols, communicating information between networks that agree to use the same digital language. The internet’s enormous success is based on its ability to connect networks and devices that are dissimilar, but can agree on how to interpret a patterned series of informational “bits.” Protocols determine the structure of light pulses transmitted through glass, and how these pulses are read when they arrive at their destination. Alexander Galloway (2004) famously wrote about the power of protocol, but this is just one way to consider what holds the internet together. Glass is part of the material topography of internet infrastructure (Hu 2015; Starosielski 2015), and these fibre-optics are integrated into vast telecom architectures consisting of company offices, data centres, buried conduit, outdoor cabinets, cell sites, and the wires hidden within the walls of your home. The distribution of these components is determined by landforms and the geography of property rights, regulatory regimes and contractual agreements. We can also consider shared ideas about internet technologies, or our internet myths and visions (Flichy 2007; Streeter 2011). Finally, there are organizations of internet governance, determining standards (including new protocols), internet addressing, and interconnections between networks (see DeNardis 2009; Mueller 2004). The internet is made from all of the elements listed above. It is what science and technology scholars refer to as a socio-technical system, an assemblage, or an actor-network, and it operates on a

Scenes from Canada’s Internet

27

global (even extraterrestrial) scale.1 Because the internet is infrastructure, it also extends through everything it supports: protocols built atop protocols, applications layered onto transport infrastructure, a wide variety of everyday services. The internet exists as an idea, but also as matter in the ground and radiation in the air. It is controlled by technical processes, agents constructed out of code (see McKelvey 2018), people meeting in boardrooms, and administrative tribunals. This chapter describes some of these different dimensions in order to provide a fragmented answer to a surprisingly difficult question: What is the internet? THE INTERNET IS VITAL

As telecom networks have expanded our abilities to communicate, we have come to depend on connectivity. But dependence is not just a consequence that follows from the proliferation of networks – dependence also leads to the proliferation of networks. Once a technology becomes essential for a society to function, it spreads. Property developers increasingly install fibre-optic connections or conduit in new housing developments. In a market, growing dependence manifests as increasing demand, to which the internet’s suppliers eagerly respond by expanding and intensifying networks. Newly essential technologies are also governed accordingly, which is not to say that they become a state responsibility, particularly in societies where essential services are often distributed through corporations and market forces. However, when market forces make it unprofitable (or less profitable) to meet the needs of some populations, public bodies bear some responsibility. In practice this often means that state agencies give private actors a push, either by offering an incentive (funding) or imposing an obligation (law) for them to meet an essential need. When did the internet stop being a luxury – a service that one could choose to pay for – and become essential for participation in society? When did the few remaining Canadians without internet access become a public policy problem? Government agencies have long highlighted the importance of internet infrastructure, but when did we come to presume that everyone should either have a broadband connection, or was badly in need of one? For many Canadians the indispensability of being connected hit home during the covid-19 pandemic in 2020. However, for regulators the moment that best cap-

28

Telecom Tensions

tures this shift occurred on 18 April 2016, during the crtc’s public hearings in Gatineau, Quebec, when the Commission was considering the level of connectivity that Canadians should expect to have available to them. Decisions of the crtc can have enormous societal impact, shaping how data flows in society, how communications infrastructure is configured, who participates in the telecom and broadcasting industries, and what information Canadians can access. However, despite these substantial powers, the crtc is not a court. At best, it is a quasi-court (or “quasi-judicial administrative tribunal”), which must stay within the lanes of its governing statutes, can be overruled by a higher power, and has limited power to investigate or punish wrongdoing. The crtc headquarters lacks the imposing architecture of Canada’s Supreme Court on the opposite bank of the Ottawa River, and its Commissioners wear technocratic attire rather than lordly robes. Hearings on the future of the nation’s infrastructure convene in a nondescript conference room, in an office block shared with other government departments. Commissioners arrange themselves at a long table elevated on a riser at the front of the room, while “presenters” take turns sitting before them. After each presentation, the Commissioners take turns asking questions. While hearings are formal, they are not as encumbered by procedure as a court proceeding – a characteristic designed into the crtc during the 1970s to increase public involvement in the policy process. However, the question of how to best facilitate public participation has lingered throughout the crtc’s history (see Shepherd, Taylor, and Middleton 2014). In April 2016, public voices appeared to be ascendant at the crtc. Jean-Pierre Blais was entering the last year of his term as Chairman, during which he had worked to increase engagement and participation in the Commission’s proceedings, with an emphasis on consumer issues. That month, the crtc was holding hearings on one of the most wide-ranging public policy questions it could possibly address: determining what telecom services were required by Canadians to participate in society. A key question to be answered by the proceedings was whether some level of broadband access would be categorized by the crtc as “basic,” which the Commission could then oblige or incentivize telecom companies to provide to Canadians. For one week, the four Commissioners and their Chairman had sat and heard proposals from various selected groups and individuals who articulated the connectivity needs of Canadians. Then, just prior

Scenes from Canada’s Internet

29

to the lunch break on 18 April, Chairman Blais told the room he would be taking the unusual step of making a statement, given the exceptional nature of the hearing. The timing of his statement was important, he said, as the crtc was now scheduled to hear from the large incumbent isps, who arguably had the greatest amount of capital at stake. The previous week’s presenters had included Indigenous organizations, regional associations, rural Canadians, low-income internet users, the deaf and the speech-impaired. Blais wanted to publicly communicate what he had learned from these speakers, but he was also addressing the corporate executives who were scheduled to present next. First, Blais dispensed with the question of the internet’s importance. He declared broadband to be “vital” and stated that the Commission would “not spend more hearing time on this self-evident truth.” The dictionary definition of “vital” offered by Blais was “essential to life, to the existence of a thing, to the matter at hand, and to success more broadly.” Broadband specifically was “vital to economic, social, democratic and cultural success of individuals and collectivities.” Having thus established broadband’s importance in society as “given,” the Commission could move on to “more important things,” such as how best to eliminate gaps in connectivity (crtc 2016a). Blais’s surprise declaration left some of the scheduled participants scrambling to revise their upcoming presentations, in light of the vital force that had been identified at the heart of Canadian society. However, Blais was merely putting his official seal on what had become internet policy common sense. Canadians’ growing reliance on connectivity had been a key justification for the government’s Cyber Security Strategy in 2010 (Public Safety Canada 2011a), and bridging the “digital divide” had been a public policy concern for even longer (Birdsall 2000). Canada’s telecom companies had extended their networks, reaching more communities, but as connectivity became a widespread assumption, the several million Canadians without broadband faced larger and larger impediments. Twenty years ago, not having internet access meant missing out on an exciting new social realm. Now, it meant lacking access to basic information and government services. What counted as basic connectivity had also shifted, and at the end of 2016’s basic service hearing the crtc set “ambitious” broadband target speeds that were ten times the previous goal (Jackson 2016). Rather than defining basic connectivity as a bare minimum for contemporary citizenship, the crtc looked ahead to a future of digital plenty.

30

Telecom Tensions

In 2021, going without connectivity in Canada is acceptable if one is travelling through the deep bush or “detoxing” from the digital economy. Otherwise, network coverage is a public policy problem that all levels of government address with a mixed bag of programs. The crtc’s universal broadband targets announced in 2016 were a long-term aspiration without sufficient funding or obligations to implement them widely, but these metrics became a standard used to determine funding eligibility,2 and the decision is now remembered as the moment broadband became a basic service. Today, the internet is as vital as ever, and as long as its guiding statutes3 are in place, the crtc remains a centre of Canadian telecom policy. While key to federal jurisdiction over telecom, it is not the only state body that governs connectivity, and the Commission has at times struggled against other organizations that have rejected its authority over aspects of how telecom networks operate. However, anyone who needs to understand how internet governance works in Canada must familiarize themselves with the bureaucratic genre of written submissions to the crtc, how to navigate the Commission’s archaic website, and the public-facing role of its Commissioners. The decisions of this tribunal have national implications for what networks are built, the terms under which these networks can be accessed, and how data packets are treated. But the Commission is often distantly removed from the effects of its decisions. It relies on participants in proceedings to provide information about the disposition of networks, to implement its judgments, and to complain about organizations that break the rules. The work of building, operating, and managing internet infrastructure belongs to other actors. MEETING ROOMS AND MEET - ME - ROOMS

The telecom industry is spread across the globe and difficult to comprehend in its totality. Conveniently for Canadian internet scholars, the industry’s representatives regularly concentrate themselves in a conference hall, where they discuss what they have in common, form and maintain relationships, eat, drink, and make deals. For some attendees, this can be an enjoyable break from their regular jobs: a place to reflect on the bigger picture, to hear news and gossip from beyond their organization’s walls. But conferencing is also work, and important decisions are made or initiated in hotel meet-

Scenes from Canada’s Internet

31

ing rooms, lobbies, hospitality suites, and nearby bars. The crtc, ised, and federal Cabinet govern from their sovereign perches in Ottawa, while the industry sorts itself out through its own hierarchies. Attendees at industry conferences discuss, critique, and forecast their regulators, but the most direct forms of internet governance are effected through the decisions of network operators – the isps. The crtc can set out objectives for rural broadband, interconnection obligations, and tariffed rates for wholesale access, but industry players decide where new infrastructure will be built and negotiate specific arrangements. A telecom conference is a room full of competitors with shared interests, of peers with power differences, and of industry politics personified. The most interesting conference dynamics occur when participants occupy very different positions in the political economy of telecom: incumbents and iisps, regulators and the regulated, along with a few opinionated academics and more reserved public servants. Disagreements sometimes publicly flare during staged regulatory panels, where lawyers spar over definitions of competition, the unfairness of each other’s practices, and regulatory outcomes. But no-one benefits from making enemies in the industry, and Canadian civility generally triumphs. Around the edges of the conference, incumbents and iisps court each other’s business, sometimes disappearing to a private room to seal the deal. There are some smaller and more specialized regional internet conferences and some international meetings, but the biggest regular events happen in Ontario,4 where Canadian business and government concentrate. While Toronto is an important hub for some obvious reasons, its relevance to the internet’s topography is less known to those outside the industry. Many arrivals to the city disembark at Union Station, where thick volumes of train passengers are channelled to and from the downtown core in a daily rhythm. But a short ways down the street from Union Station is a different kind of distribution centre that is no less vital: 151 Front Street West was built to house telecom equipment for the railway, but became the region’s transportation hub for data packets (Morales 2011). This building is a meeting place of networks, sustained by the fibre lines that converge underground. These carry communications from around the globe up through the basement, to the interconnection equipment on higher floors, and then back out to the world.

32

ISP / OSP / CDN

Telecom Tensions Interconnection Facility / IXP

ISP Last-mile Network

Home User

Upstream Downstream

Figure 1.1 Interconnecting isps The isp that covers the “last mile” to your home connects to the rest of the internet “upstream” at one or more interconnection facilities. This might be an ixp such as 151 Front Street West, or a building where an isp with longer reach makes wholesale access available (see chapter 3). At an isp, it is often possible to connect to different kinds of networks, including long-distance global carriers, servers, or “caches” belonging to osps such as Google and Facebook, and cdns that move content closer to “downstream” users.

As is typical of ixps and carrier hotels, the building is unremarkable on the outside. Union Station’s infrastructural importance is reinforced with massive columns, whereas 151 Front Street West resembles a brick with windows. Foot traffic in and out the doors is sparse (all arrivals are screened by the security desk), and the building’s floors have few human occupants. What make the place noteworthy are the greater parts of the structure where sunlight never reaches – where servers, switches, and related telecom equipment operate in a temperature- and humidity-controlled environment. The only hint of this remarkable interior visible from outside is 151 Front Street West’s reinforced rooftop, which one can see by looking down from one of the nearby skyscrapers. Virtually the entire surface of the roof is covered by fans, condensers, chillers, and massive diesel generators. In the

Scenes from Canada’s Internet

33

event of a power outage, these generators are designed to kick in and draw fuel from tenant-owned storage tanks. Tanker trucks are expected to pull up and connect to lines at the back of the building to keep diesel reserves topped up. The central promise of a facility such as 151 Front Street West is that, come what may, its tenants will never experience a service disruption that is beyond their control. 151 Front Street West is not just a data centre, but a place where networks benefit from being in proximity to each other. The “heart of the building” is its secure “meet-me-room,” where interconnection happens between intermediaries and major world-spanning networks (Morales 2011; Rebel Networks n.d.). This space is not accessible to visitors, but outsiders are occasionally granted a tour of another lessimpressive but still-crucial piece of interconnection equipment: the Toronto Internet Exchange (TorIX) switch. Here, multiple networks can exchange traffic over a shared “peering fabric,” which has provided vital opportunities for Central Canada’s iisps. For the past twenty years, TorIX – a volunteer-operated not-for-profit – has made it easier for smaller intermediaries and some content providers to connect to one another. Any network recognized as an “autonomous system”5 is welcome to “peer” at TorIX, but while Canadian incumbents also lease space in 151 Front Street West, they are selective in whom they connect with and see little benefit in participating in an internet exchange such as TorIX (Kyonka 2012). iisps, on the other hand, use TorIX to limit their dependence on incumbent infrastructure and to exchange traffic in mutually beneficial ways. I was able to tour TorIX during a break from a nearby conference, accompanying a small group of peering geeks. Some of those on the tour were setting up much more modest ixps in parts of Canada that had never seen an internet exchange, and were keen to learn what they could from TorIX. Also in our group was William Norton, who had been part of the internet’s development since the 1980s, and translated his long experience with ixps into a career as an international connectivity consultant. Norton is best known for documenting and explaining the art of peering, a process that occurs when two networks share traffic with each other (see Norton 2010). As described in the Introduction, one can see what happens at TorIX or a data centre’s meet-me-room as the materialization of an agreement between parties. The agreement might be sealed with a handshake or with a contract written by lawyers, but this precedes the technical business of pushing packets. Connecting two different pieces of tele-

34

Telecom Tensions

com equipment with a cable is the relatively straightforward outcome of negotiations and discussions, and industry conferences have been an important venue for this dimension of internet governance. For a sociologist of the internet, conferences are a great place to observe industry dynamics and power relationships. They also provide opportunities for serendipitous encounters and unexpected information; you never know who might share your table for a meal, and where the conversation will turn. Later the same day at the conference’s keynote dinner, I sat at a table with individuals representing two very different networks, which occupy separate parts of 151 Front Street West. On one side of the table was the ceo of an Ontario-based iisp, who takes the time to personally answer customer questions on an online message board, and is happy to peer with other networks through TorIX. On the other side were a few representatives of an enormous nation-spanning incumbent, the ceo of which will likely never attend such a conference, and whose company will likely never peer at TorIX. The two organizations actively competed against each other for retail customers, and parts of Ontario were a battleground between their respective networks. The organizations had also fought each other over money and policy before the crtc. And yet, the two isps also had to work together, with the smaller company relying on the bigger network to reach some of its subscribers, and paying for this service as one of the incumbent’s wholesale customers. That night at the telecom conference, the dinner conversation between these potential rivals was perfectly civil. The individuals at the table were human elements in very different socio-technical systems, decision-makers occupying formalized roles in their respective corporations. They were also social beings, sharing a meal in each other’s company, getting along, and doing interpersonal work. The internet that actors such as these are responsible for is composed of light and hardware, linking people and human organizations in ways that often go unnoticed. But when the internet does enter into our consideration, particular ideas, visions, and myths of technology are integral to how we engage with it. UTOPIAS OF CONNECTIVIT Y

Prior to the covid-19 pandemic, Canadians’ view of the internet had been darkened by the “techlash”: a steady series of Silicon Valley scandals, harmful use-cases of connectivity, and disillusionment with the

Scenes from Canada’s Internet

35

promises of a networked society. The “cyber-utopianism” that optimistically accompanied the arrival of these technologies is still very much alive, however, and often remains the default discourse for new infrastructure projects. Utopian visions continue to drive efforts to expand connectivity, and while these must now address a more cynical public, utopias have never been an easy sell. In the case of the failed redesign of Toronto’s waterfront as a “smart city,” fantasies of “Silicon Valley North” were soured by concerns about data collection and corporate control (Barth 2018). Less known are the dozens of other places in Canada where small communities have tried to develop their own limited utopias of connectivity. Attempts to realize these visions often stall when they encounter the challenges of raising capital, gaining political support, achieving technical expertise, and overcoming competition. While no Canadian town has transformed itself into an internet paradise, some have achieved remarkable things in the pursuit of a shared vision. In The Internet Imaginaire, Patrice Flichy examines the utopian “collective vision” that animated the early public internet of the 1990s. Flichy argues that utopias are one of the elements from which reality is built. A utopia can “can lay the foundations for a great project and mobilize actors” (Flichy 2007, 29). A decision to implement a project can also be made on practical political grounds, after which utopias are elaborated to legitimate these efforts. Internet utopias have developed along with its infrastructure, and today’s internet imaginaire is a varied collection of fantasies, with different collectives elaborating different collective visions. For massive intermediaries such as Google, building utopia is a core part of the business model. For some Canadian communities, however, fostering a collective vision can feel more like a matter of survival. Canadian towns are often tied to specific industries; when these industries run their course, Main Street becomes deserted and younger residents relocate to urban areas. Once the historical rationale for a community no longer holds, community leaders will often articulate a new vision of the future. Connectivity, as part of that vision, offers the promise of innovation and redefinition. Unfortunately, the communities most in need of utopia in Canada are also often the least likely to progress towards its realization. Utopias are not cheap to design or build – the initial work can involve a consultant or two, but construction costs quickly escalate into millions of dollars. This cost must typically be recouped, but as a long-

36

Telecom Tensions

term investment that does not provide the kinds of returns that are attractive to private capital (see Crawford 2019). The need for a collective vision remains after public infrastructure is built, because infrastructure must be operated and maintained, and its future depends on the stability of political support. Public policy often proceeds through modelling (Braithwaite and Drahos 2000), where the approach taken by policy-makers in one jurisdiction is used as the template for another. Municipal governments generally do not invent novel solutions to their problems, but look to their peers for models to adapt. In Canada, municipal leaders and administrators frequently share ideas and experiences, and the “Olds model” has been a particularly influential point of reference for those working to reinvent their communities around a vision of bountiful connectivity. Imagining the Future in Alberta No-one driving through Olds would imagine the place to have been a pioneer in municipal broadband, and it looks much like other Prairie towns of similar size. Visitors to Olds typically arrive from nearby Highway 2, the main corridor for intra-provincial traffic, connecting Calgary and Edmonton.6 Once an important stop for the railway, Olds was bypassed when the highway was built, and it lies just out of sight for the tens of thousands of vehicles travelling between the province’s major cities. But highway corridors carry more than just cargo and people; like railways, they often serve as an important channel for communications networks. Buried along the side of Highway 2 are fibre lines that carry internet traffic between Calgary and Edmonton. While semi-trailers haul industrial equipment to the province’s north, data packets zip past them underground, carrying content from the major internet connection points in the United States. Less than an hour’s drive from Calgary, some of these packets are diverted a short distance west to Olds and its 9,000 residents. The town’s proximity to transportation corridors and internet “backhaul” turned out to be a crucial factor when it came time to realize the vision of becoming a connected community. Among its other advantages, Olds hosts a college, expanding residential developments, numerous businesses, and volunteer organizations. While the town has struggled with the challenges of rural eco-

Scenes from Canada’s Internet

37

nomic development, it is greatly advantaged in comparison to most rural communities in Alberta. The story of the Olds fibre project is therefore not one of a community stuck on the wrong side of the digital divide, struggling to remain competitive and attractive, but of a town with the resources and leadership required to innovate and invest in a major piece of public infrastructure. Fibre was built in Olds not because the town was starved for connectivity, but because the town had the social infrastructure in place to plan and deliver on a long-term network infrastructure project. All of this also benefited from the fortunes of internet topography. The geography of backhaul was originally not thought to be a factor in the wiring of Olds, which was initiated by a utopian vision emanating from the provincial government. At the beginning of the 2000s, the new Alberta SuperNet was constructed to overcome geographic barriers to connectivity and to bring rural broadband in line with what was available in urban centres. This would be achieved by extending fibre to all towns in the province, and setting rates for connectivity throughout the network. Conceived at a time of budgetary plenty, the SuperNet was justified by a grand vision of Alberta as a world leader in internet innovation (Collison 2003; Scotton 2000). Because the specifics of this vision were vaguely and inconsistently communicated, rural Albertans were unsure of what to expect from the project, or had expectations that could not be met (Kozak 2010). In Olds, some enterprising community members came to believe that a local isp could wire the town and then plug into the SuperNet. It was only years later, after the province-wide SuperNet was built, that they learned the network would not support their digital aspirations. By then, community leaders had already sketched out a conceptual foundation for the future of Olds, and were proceeding along the path that would make it Canada’s first “gigabit town” (oicrd n.d.). This would mean constructing a comprehensive municipal fibre network, establishing an intermediary (O-Net) that could provide residents with broadband, phone, and television service, and leasing a long-distance connection to Calgary where this isp could connect to the rest of the internet. Utopian connectivity projects can develop without a consistent or coherent vision, or have a vision that is not widely recognized in a community. This is the position that leaders in Olds found themselves in during 2014, as their ambitious fibre build neared completion.

38

Telecom Tensions

They complained about citizens’ misunderstandings of the project, but also acknowledged their own limitations in communicating a vision. I travelled to the town with the Centre for Public Involvement, an Edmonton-based organization that specialized in a particular model of deliberative democracy, working to enhance public participation in decision-making. Our job was to facilitate a “planning circle” consisting of randomly selected citizens of Olds, who would share their hopes and dreams for the gigabit network. Even today, a gigabitspeed internet connection is more than most people know what to do with, and I was tasked with explaining the possibilities of this level of connectivity to the citizen volunteers. Their job was to imagine how their community could be transformed. The citizen planning circle was held in the Bell e-Learning Centre – a facility at the Olds College campus that had opened in 2008 to showcase the capabilities of the SuperNet, and to bring rural Albertans into the knowledge economy (Canada NewsWire 2008). Less than six years later, the planning circle’s participants sat in a classroom that had been wired with expensive teleconferencing equipment (now outdated and idle), discussing the sorts of possibilities that had not been realized through the SuperNet. A professional facilitator helped to coordinate discussions and contributions, and a graphic illustrator translated spoken content into visual form. At the end of the three-day process, the thirteen citizens had identified community values and principles that should inform decision-making about connectivity in Olds, as well as objectives and action proposals for broadband in the community. One participant described broadband as “our pipeline to the future” (Everest 2014) – a future of improved access to medical services, new opportunities for agriculture, education, and business, all while maintaining community safety and individual privacy. Participants saw broadband as a community asset that needed to be developed with broad engagement around a collective vision (Centre for Public Involvement 2014). Internet infrastructure is an open-ended resource for realizing human possibilities. Like the personal computer of the 1980s, broadband is a general-purpose tool that our imaginations can put to work in ways that are difficult to anticipate. A community fibre network can be built without an idea of how it will be used, or its design can form part of a larger vision for the future of a place. A vision can be developed before, while, or after the fibre is buried. Whether the

Scenes from Canada’s Internet

39

future resembles the vision depends on the extent to which visionary ideas are linked to the collective decision-making process, as well as some factors that are outside planners’ control. People do not always play their roles as prescribed: politicians may change their minds or be replaced through elections; citizens may not use new services, or use them in unexpected ways. New residents and businesses may not arrive as promised, despite new forms of connectivity. Consulting with stakeholders can help keep things aligned, but stakeholders raise a variety of issues and proposals, while planners have to be selective and reconcile their budgets. The planning circle in Olds ended up being treated more as an experiment in community engagement than as part of the decisionmaking process. The project had been funded by a provincial government department interested in promoting broadband across rural Alberta, and Olds seemed a good place to start since the town already had a fibre network, which made it easier to imagine future possibilities (Everest 2014). The local isp’s champions had been supportive, but the planning circle was not their project, and they were not accountable to its recommendations. The circle articulated the dreams and aspirations for internet infrastructure in one particular town, as expressed by local volunteers, but in a way that was largely disconnected from the forces that actually determine how infrastructure is built and used. In subsequent years, O-Net would extend public Wi-Fi through the town and promote its internet services to other municipalities, but it has also had to contend with the large debt incurred by the network’s construction and the resulting political struggles with local leaders (Collie 2020). Despite its financial and political challenges, discussions of public connectivity elsewhere in Canada often refer to the town’s network as a prominent example of community broadband. Ideas are the raw materials from which decisions are made, and some version of the Olds “success story” (Hirsh 2020) continues to inform the utopian possibilities for other communities, representing a fulsome approach to public connectivity that challenges traditional public/private distinctions. Across Canada, broadband possibilities are being considered, debated, and acted upon. People who are dissatisfied with their local circumstances are looking beyond for visionary ideas and practical experiences. Some of these plans and imaginings will be actualized in new

40

Telecom Tensions

networks, and many others will be blocked by various obstacles: conceptual, political, material, and financial. Having a broad appreciation of the relevant levels and nodes of decision-making is helpful for negotiating these obstacles. Also, as the next chapter demonstrates, in order to understand present circumstances in Canadian telecom, it is instructive to examine its history.

2

The History and Political Economy of Canadian Intermediaries

To understand the forces governing internet intermediaries in Canada, as well as the current configuration of actors and their power relations, it is instructive to briefly look at the historical development of the country’s telecom networks. Today’s current of intermediation has its roots in the 1990s history of telecom liberalization, early isps, and mobile connectivity, but similar dynamics have played out in different times and places where telecom networks have become vital to society. Before the internet, control over telegraphs and telephones was key to a variety of political projects. While publicly owned networks were sometimes part of local and regional aspirations for autonomy, public and private interests were typically intertwined in the history of Canadian telecom (Armstrong and Nelles 1986; MacDougall 2013). Role conflicts occurred when telecom companies had to decide between competing social demands, such as the tensions involved in being both a capitalist enterprise and a public utility; telephone operators also played many roles with contradictory expectations, as part of the job of mediating communication and interacting with callers (Martin 1991). Finally, the use of private intermediaries as instruments of public policy was an important part of the governing rationality of North American telecom for much of the twentieth century. Liberalization changed the nature of some of the responsibilities associated with connectivity in the 1990s, but did not introduce the idea of governing through private actors. The transitions of the 1990s have often been called “neoliberal,” but I argue that the crucial shift that occurred in the late twentieth century is more accurately described as the embrace of regulatory capitalism, which redefined the public service obligations of telecom

42

Telecom Tensions

companies. This was a shift toward liberalization and market competition, but not deregulation. Instead, regulation became seen as the key to promoting competition, and dealing with its consequences. Canada’s internet topography and the political economy of telecom are grounded in these historical transformations, which also parallel those in many other countries (oecd 2012). Twentieth-century telecom monopolies have become today’s “incumbents,” inheriting a privileged position that governments have tried to neutralize with different kinds of regulation. Rather than constituting a clean break with monopoly telephone regulation, telecom governance is built on the foundations of the past. Today’s telecom infrastructure is also layered upon this history (see Mattern 2015). 151 Front Street West, which houses the internet exchange and carrier hotel described in the previous chapter, was a structure built to house equipment for telex (a descendant of the telegraph) in the 1950s (Morales 2011). Fibre lines are now installed in the same utility corridors through which telephone and telegraph cables once extended. Incumbents’ utility poles and conduits must now be accessible to competitors, but incumbents continue to have the benefit of “priority access” to this infrastructure (crtc 2010b). In addition to today’s networks still making use of material infrastructure that was built during the monopoly era, internet regulations have drawn on many principles and precedents from the regulatory history of telecom and telephony. This chapter traces these developments and their role in shaping the political economy of the present, providing additional context to contemporary actors and policies. The objective is to explain how Canada’s telecom networks are governed following liberalization. As in the days of corporations explicitly chartered to serve the public interest, today’s intermediaries are required to fulfill certain public obligations. Some, such as their responsibilities concerning internet child exploitation and cyber attacks, are a response to internet-era social problems enabled by technological change. Others, such as the mandated access regime, are meant to structure competitive relations following liberalization. However, many of the public policy issues in Canadian telecom have deep roots in a previous era of telephony and monopoly.

History of Canadian Intermediaries

43

TELEPHONY AND MONOPOLY

The early days of telephony in Canada, beginning in the late 1870s, saw a proliferation of competing companies stringing wires above city streets and into the countryside. These freewheeling days of unbridled competition were messy and short-lived, and in the span of a few years the Bell Telephone Company of Canada acquired a telephone monopoly in much of the country, first buying up the local exchanges, and subsequently connecting them with long-distance lines (Rens 2001). The decades that followed were characterized first by regional fragmentation, then by consolidation and monopoly. The Bell company’s rise in Canada was due in large part to its close relationship with the American Bell Telephone Company (later at&t), and its success in petitioning the federal government to grant Bell a charter to develop its telephone network across much of the country.1 The Bell Telephone Company of Canada Act that incorporated Bell in 18802 also had the effect of asserting Ottawa’s regulatory control over telephony. While provinces and municipalities could operate their own telephone networks, they were prohibited from appropriating parts of Bell’s network, or interfering with rights granted to Bell under its charter. These rights included access to public rights-of-way (such as streets and bridges) along which Bell’s network could expand. Rights-of-way were a source of conflict between Bell and municipalities, often resolved through exclusive franchise agreements that Bell was able to negotiate, granting the company dozens of municipal monopolies.3 Bell would refuse to interconnect with many smaller telephone companies, and would buy out competitors in wellpopulated areas.4 However, Bell was largely uninterested in expanding its network to less-profitable rural regions, and was more accommodating with rural companies that it did not see as competitors. Many villages and some municipalities constructed their own telephone systems to meet local communication needs. Prairie farmers depended on communication in areas where low population density made building networks uneconomical, and formed co-operatives to deal with various collective challenges, including communications. The new provinces of Alberta, Saskatchewan, and Manitoba proved particularly independent and resistant to Bell’s westerly expansion, buying the company’s assets in a populist political climate characterized by western pride and opposition to the eastern “monopoly.”5

44

Telecom Tensions

In Ottawa, municipal dissatisfaction with Bell and calls for government intervention culminated in a 1905 Select Committee to investigate telephones (the Mulock Committee), which heard from various parties complaining about Bell’s practices, some of whom argued for the nationalization of long-distance telephone service and municipal control over local service.6 In its defence, Bell could argue that it was working in the interests of the “general public” (Armstrong and Nelles 1986, 171), as its charter had been extended and revised in 1882 to define the company as working “for the general advantage of Canada.”7 The company’s private interests had thereby been explicitly defined as benefiting the Canadian public, enabling the company’s spokespersons to wrap Bell in nationalistic rhetoric, and helping provide some legal protection against municipalities seeking control over Bell’s poles and wires. However, Bell’s charter could not guarantee that the company would be free from federal government regulation, or the threat of nationalization. Critics could even turn the language of its charter against Bell when pointing out its neglect of rural areas (Dagger 1915).8 The Mulock Committee of 1905 did not immediately result in significant changes to the political economy of telephony, securing Bell’s position as an incumbent in much of Canada, and rejecting the possibility of nationalization. However, by 1906 Parliament decided that the company would be regulated by a permanent and neutral commission.9 Since Bell had an effective monopoly on long-distance service in much of Canada, its new regulator deprived Bell of the ability to unilaterally refuse interconnection with other companies (interconnection being crucial for local telephone systems to make calls beyond their local areas). The regulatory commission could also examine Bell’s costs in order to set “just and reasonable” interconnection rates.10 The installation of neutral regulatory commission to oversee telephony established a framework that would spread across the United States at roughly the same time, where Bell companies also became stewards of the public interest, were regulated by independent commissions, and in exchange received certain state guarantees or protections (such as monopoly rights). In both countries, this “public service liberalism” (Stone 1991) was ended in the 1980s and 1990s, but many of its fundamental dynamics have continued today under the crtc. For instance, the wholesale internet regime described in chapter 4 mandates interconnection between incumbents and smaller interme-

History of Canadian Intermediaries

45

diaries, and leaves the two perpetually battling before the crtc over tariffed rates and what incumbents deserve to be paid for access to their facilities. But while the wholesale internet regime has been justified primarily as a policy to enable competition, mandated interconnection in early telephony was justified by the “public interest” of connecting the subscribers of different companies (Canadian Independent Telephone Association 1911, 483). Competition in telephony was actually seen as undesirable for much of the twentieth century in Canada (see chapter 4), but regulators did work to restrict the anti-competitive practices of telecom companies under another doctrine that would come to be relevant to internet intermediaries, known as “common carriage.” As is discussed in chapter 5, common carriers are transportation and communications institutions that are seen as serving the public, and are thereby prohibited from refusing service or discriminating between customers. Railway companies were required to carry freight from different clients on equal terms, and telecom companies (who shared the same regulators as the railway companies for much of the twentieth century in Canada) were likewise eventually required to transmit communications without interference or discrimination. In effect, common carriage meant that any institution that acted as a “carrier” of information (what would come to be called a “mere conduit”) was barred from doing anything other than ensuring that this information reached its destination, and was obliged to treat all communications on an even-handed basis. Previously, telegraph companies had been intimately involved in the news-gathering process, and gave preferential treatment to their press partners. Common carriage meant they could no longer act as gatekeepers for journalism or play favourites among their clients. The legacy of this regulatory principle continues to have significant implications for internet intermediaries, affecting copyright enforcement, net neutrality, interconnections between isps, and the extent to which incumbent isps can take advantage of vertical integration. MAINTAINING NATURAL MONOPOLY AND END - TO - END CONTROL

Power relations in Canadian telecom largely stabilized after the onset of the 1920s, with municipal tension easing and the majority of independent telcos gradually going out of business. Bell and the provincial

46

Telecom Tensions

telcos consolidated their territories, and found ample reasons to cooperate as they became interconnected parts of a national long-distance network.11 The principle of common carriage became well-established and adhered to. Finally, as telephone connectivity became indispensable to modern life, the principle of “universal service” gradually developed, with the public policy goal of providing service to all Canadians. This became a key justification for why the telephone system was best treated as a “natural monopoly,” since it was argued that universal service was only possible through a large monopoly that could use profits from urban areas to subsidize more expensive rural lines. Another key principle justifying the natural monopoly of the telcos was the idea that a company’s control over its network should be complete and centralized for technical reasons. Telcos argued for the importance of maintaining the “integrity” of the system when opposing attempts to connect new devices to the network. By the end of the 1960s, Bell was increasingly concerned that new technologies and a regulatory shift towards liberalization would threaten its dominant position, and endeavoured to convince policy-makers that “end-toend” control of its network was key to its past and continuing success (Mussio 2001, 107–8). While these arguments for total control were driven by fears of change to the political economy of telecom, the basic idea that telephones needed to be managed as a single system had been raised as far back as the Mulock Committee of 1905, and was likewise the basis of at&t’s argument in the United States for a single, centralized telephone network under its control (MacDougall 2013). A network cobbled together from multiple heterogeneous parts, or used to connect whatever assorted devices customers decided to plug into it, was portrayed as unwieldy and dangerous. Hence, arguments for “systemic integrity” and end-to-end networks opposed the attachment of “foreign” devices (not owned or approved by the telco) to the network, which could threaten or “pollute” the system (Babe 1990, 143–4). Today, we can recognize this argument as contradicting a key principle underpinning the architecture of the internet. Natural monopoly in telephony was justified by the need for centralized control over a singular system, including the telephones at the endpoints, but for the public internet to develop, users needed to be able to connect their own computers to telephone lines. The internet has its own “end-to-end” principle, which is understood to mean that most features should be implemented through the computers at either end of

History of Canadian Intermediaries

47

the network, with everything in between limited to faithfully carrying data packets to their destination (Lemley and Lessig 2001). In other words, the “intelligence” of the network should be at the endpoints, with “dumb pipes” in between, and this simplicity at the core of the network allows various kinds of systems to interconnect using internet protocols. Like the end-to-end argument for system integrity advanced by the telcos in earlier decades, the internet’s end-to-end principle is a normative and political vision for what telecom networks should be, dressed up in the language of technical efficiency (see Gillespie 2006). Both arguments are premised on their respective networks’ supposedly inherent features, but they are also defensive moves against specific threats: the telcos feared the loss of their monopolies, just as internet activists fear the loss of a decentralized internet. However, these opposing concepts of “end-to-end” belong to separate discourses and historical periods. In order for the idea of the internet to emerge as a decentralized end-to-end network, the idea of the telephone system as a centralized end-to-end network first needed to be pushed aside. THE GRADUAL BREAKDOWN OF SYSTEM INTEGRIT Y AND NATURAL MONOPOLY

The notions of system integrity and natural monopoly were rooted in technological determinism; proponents argued that these were the conditions that the technology of the telephone naturally required. There were other telecom networks in the twentieth century, but these were kept distinct from telephony conceptually and in regulation. This sense of what was natural concealed the underlying political economy of telecom, so that monopoly became a necessity rather than a choice, and telephone, radio, and cable television occupied separate worlds despite their overlapping capabilities (see Babe 1990; MacDougall 2013; Winseck 1998). In the 1970s, this orderly world began to break apart. For example, while cable was eventually deployed throughout urban Canada and capable of carrying far greater volumes of information than the telephone’s “twisted-pair” copper lines, cable systems were prohibited (through their agreements with telcos) from stepping outside of their narrowly defined role as carriers of television signals.12 Many of these restrictions were relaxed by regulators in the mid-1970s, first under the Canadian Transport Commission (ctc)

48

Telecom Tensions

and then under the crtc. While terms such as “information highway” and “internet” had yet to be brought into use, it became increasingly evident that telecom networks, whether operated by telcos or cablecos, acted fundamentally as carriers of information, and that attempts to limit them to particular uses or configurations were not dictated by technological necessity. This realization was slow to spread, in large part because incumbent telcos and cablecos were keen to preserve the status quo (Winseck 1998). Telco incumbents resisted interconnection between their networks and other devices or systems, and western provinces were opposed to anything that might threaten their telecom monopolies. However, by the mid-1970s the federal government, under pressure from it companies such as ibm and the Electronic Industries Association of Canada, came to see interconnection as a crucial way to promote innovation and to maximize the public benefits of data communications services (Mussio 2001). A major regulatory shift occurred in 1976, as the responsibility for telecom regulation at the federal level was transferred from the ctc to the crtc. The move stemmed in part from the perception that the ctc was undemocratic and that its process worked in the telcos’ favour. Whereas the ctc had refused public participation in its hearings, the crtc promised to take a broader view of the public interest, holding less-formal hearings that were open to the public (Babe 1990; Stevens 1976). While the crtc did prove to be more open to hearing from voices that had previously been excluded from regulatory hearings, questions remained about just who would represent the public before the crtc, and about the best way to encourage public participation in its decision-making. These issues continue to challenge the Commission to this day. Not long after it stepped into its new role, the crtc dismissed Bell’s arguments for system integrity,13 compelling common carriers to allow new kinds of devices to be connected to their networks. These intermediaries were no longer just obliged to faithfully carry customers’ communications from endpoint to endpoint, but were also required to give up control over what those endpoints were connected to. The issue of terminal attachments was just one in a series of moves towards greater liberalization of the telecom industry in Canada (Kaiser 1981), where technical arguments for system integrity gave way to other concerns, such as economic impacts on common carriers, the social impacts on unions and universal service objectives, and political struggles over jurisdiction between the federal govern-

History of Canadian Intermediaries

49

ment and the provinces (Wilson 2000). The eventual result would be the breakup of the established monopoly regime, occurring just prior to the arrival of the commercial internet and the first Canadian isps. This allowed various small intermediaries to take advantage of recently “opened” networks to provide budding internet services, but also marked the beginning of an era in which it made sense to refer to pre-existing telcos and cablecos as incumbents. In a newly competitive, less restrictive, and privatized14 marketplace, the field would be open to a host of newcomers. But in the end, the many decades through which incumbents had established themselves would accord them a host of advantages, and compensating for these advantages would require constant regulatory balancing on the part of the crtc. REGULATORY LIBERALIZATION

In brief, liberalization is a form of state intervention that opens the telecom industry to competition and aims to increase the number of competitors. While often described as “deregulation,” this terminology is misleading and self-serving, since it suggests that regulation gives way to the ascendancy of the market. Instead, liberalization involves choices between different forms of market regulation (Mosco 2009) in which the state often takes an active role to further different interests, rather than the removal of regulation. The state might continue to exercise power to constrain or direct the market towards values in the public interest, such as universal service and affordability. However, liberalization typically means less state involvement in rate-setting and new kinds of state intervention to encourage increased competition. The state retains the power to intervene where competition is seen to be inadequate, when competitors abuse their positions, or in instances of market failure. These forms of market regulation have kept the crtc quite busy in the era following liberalization (see chapter 4). While the goal may be to create a reliance on market forces that requires only minor regulatory intervention, the state has been unable to fade from the scene as much as policy-makers have wanted. Telecom liberalization is not simply a process that Canada underwent in the 1980s and 1990s; it is a continuous mode of governance that presents no end of problems to be solved by the state. The end goal may be deregulation, and a reliance on market forces to drive efficiency, competition, and choice for consumers. However, in the never-ending

50

Telecom Tensions

interim, the market must be regularly adjusted, recalibrated, and its players repositioned. Liberalization has been a gradual and uneven process. Federal concerns about the duplication of facilities still surfaced in the mid-1980s – a decade characterized by a lack of coherent federal policy on telecom liberalization. Serious disagreements occurred between the crtc, Department of Communications (now ised), and federal Cabinet over the shape of liberalization in the late 1980s and early 1990s. Since the crtc can effectively have its decisions overruled by Cabinet, the commission was ultimately brought into line with its political masters’“de facto” competitive telecom policy (Winseck 1998, 235–8). But there was no grand plan or explicit policy of liberalization in this period. Instead, the slow path to liberalization was built through the alignment and accumulation of individual regulatory decisions, such as opening up competition in terminal attachments and data services, as well as long-distance and local telephone service (Schultz 1999; Wilson 2000). Telecom liberalization through the 1980s and 1990s also coincided with and was complemented by the privatization of (wholly and partly) state-owned intermediaries, including the provincial telcos in Alberta and Manitoba, along with NorthwesTel, Terra Nova, Telesat Canada, and Teleglobe. With the exception of a small number of municipally owned intermediaries and specialized networks such as those used for research and education (see chapter 3), a policy consensus formed around the idea that the state had no business being in the telecom business. Future state involvement in telecom would be limited to setting and adjusting regulatory regimes, contracting out public policy to the private sector, and engaging in public-private partnerships. Finally, telecom liberalization was also a process in which regulatory power was centralized and consolidated as a federal regime with explicitly stated policy objectives. Principally, this was achieved by a Supreme Court decision that affirmed the crtc’s jurisdiction over provincial telcos in 1989,15 and by the Telecommunications Act of 1993. Previously, the crtc worked under a vague mandate to ensure that rates were “just and reasonable” and to prevent “unjust discrimination.” The Commission had taken an ad-hoc approach to regulating competition and interconnection. Now, the crtc was finally granted a legislative mandate to regulate telecom, along with an explicit list of objectives to work towards, and truly national jurisdiction over intermediaries. While the Telecommunications Act was an articulation and

History of Canadian Intermediaries

51

formalization of a national policy of telecom liberalization after this had become irreversible, it also centralized regulatory authority under the crtc, and therefore the federal government. Liberalization turned monopoly telcos and consolidated cablecos into incumbents, and opened the field to new competitors that might challenge them in their established markets, or in new services such as internet access. Ultimately, however, incumbency would prove decisive. While the crtc pried open access to incumbent networks one piece at a time, the intent was never to dethrone these champions. Instead, the regulator deprived incumbents of certain methods of control over their infrastructure, and saddled them with responsibilities to facilitate new participants in the market. New competitors that might otherwise be crushed by incumbent power would be accorded some regulatory protection, but it was also imagined that the incumbents would vigorously compete with one another. These twentiethcentury masters of their respective domains would now be unmoored from their mutually agreed-upon territories. Bell, for instance, was free to come west, and telus (formed through the merger of Alberta and bc’s telco monopolies) could build networks and fight for customers in Bell’s traditional Ontario and Quebec territories. The extent to which these incumbents actually do compete with one another is the source of much debate, and will be taken up in chapter 4. Liberalization and Neoliberalism Canada undertook telecom liberalization at roughly the same time as much of the rest of the world. This period also overlapped with the opening of public access to the internet (although at the time it was often referred to by other terms, such as the “information highway”), and the rise of the first commercial isps. Indeed, the development of the internet in each country was intimately related to, or developed on the basis of, telecom liberalization (Kushida 2015; Noam 2001). From an international perspective, Canada’s commitment to marketbased telecom liberalism has been exceptionally strong (Birdsall 2000). Academic critique has sometimes identified the dominant post-liberalization rationality of Canadian telecom policy as neoliberalism (see Rideout 2003), and neoliberalism has become an influential description and theoretical explanation of the international transformations in governance that have taken place since the late 1970s. While the term has developed increasingly broad and contra-

52

Telecom Tensions

dictory meanings (Venugopal 2015),16 a consistently emphasized characteristic of neoliberalism as an approach to public policy has been deregulation, with the neoliberal state retreating and governing less as market forces take over. However, this narrative has often been inconsistent with actual policy developments, including in telecom. As documented by Vogel’s (1996) Freer Markets, More Rules, privatization and liberalization have typically been accompanied by a proliferation of regulations and regulatory institutions. It is therefore important to distinguish between neoliberalism as a set of claims or recommended public policy positions, and the actual set of policy developments that occurred during what is supposed to be the “neoliberal era.” As Braithwaite (2008) argues, the “neoliberal policy package” was never an accurate way of describing what happened in countries such as the US, the UK, and Canada during this period. While smaller government and deregulation was certainly what some business leaders wanted, governments in these countries proved adept at deploying the rhetoric of neoliberalism and practising extensive domestic regulation. Contrary to the prescriptions favoured by neoliberal advocates, the Canadian state maintains an active role in regulating the marketplace for telecom services, and regulations accumulate wherever profit-seeking behaviour results in undesirable outcomes. Most recently, the crtc established an “Internet Code” to govern the relationships between large isps and their customers in order to prevent misleading sales practices (crtc 2019b).17 While Canadian federal policies have had a great deal in common with the liberalized regulatory approach of the United States, the Canadian state has been more willing to intervene through policies to promote competition and regulate relationships between market actors. With the resurgence of nationalism in many countries and the erection of new international trade barriers, any claim that we are living in a neoliberal era requires significant qualifications. But Canadian telecom policy was never neoliberal to begin with, and remains fundamentally unchanged in recent years. Certainly, there have been many neoliberal arguments made to promote telecom policy since the 1980s, and for a time around 2006 these arguments seemed to become guiding principles of telecom policy. But the neoliberal “policy direction” from this period,18 which the crtc is still obliged to take into consideration in its decisions, has little bearing on how Canadian telecom is actually regulated. As in many industries around

History of Canadian Intermediaries

53

the world, Canadian telecom governance has only partially reflected the neoliberal agenda. This has meant embracing liberalization, as a shift to commodification and market logic, while rejecting (or indefinitely delaying) deregulation, with new markets creating a need for new regulatory regimes. Regulatory Capitalism and the Economization of Connectivity In light of the previously discussed weaknesses of the “neoliberal revolution” as a description of contemporary governance, some alternate conceptualizations have been developed to explain the transformation that has taken place since the end of the 1970s. Among these, the theory of “regulatory capitalism,” as articulated by David Levi-Faur, combines aspects of neoliberalism (privatization and liberalization) and the regulatory state (regulating markets for the public good). Under regulatory capitalism, the shift is toward “more capitalism [and] more regulation” (Levi-Faur 2006, 521), wherein “controls are achieved by the supreme command of the two regulators: the market and the state” (497). Rather than the “retreat of the state” often identified in neoliberal analyses, authors mapping the contours of regulatory capitalism show how the role of the state has changed – becoming less involved in some areas (privatization of service provision) and expanding in other ways (“steering” the market through regulation; see Levi-Faur 2006, 505). The specific path taken by Canada in this regard is unique, with some countries (such as the US) opting for fewer wholesale obligations, and other countries (including some in Europe) pursuing a more radical approach known as “structural separation,” in which the operator/provider of a network is banned from competing with the service providers who use the network (oecd 2011). However, the fact that telecom liberalization in Canada has resulted in more extensive regulation by a larger number of regulatory agencies is consistent with developments elsewhere in the world. Rather than neoliberalism dismantling anything that opposes market logic, we have the “regulatory capitalist reality of hybridity between the privatization of the public and publicization of the private” (Braithwaite 2008, 8). In Canadian telecom, we see this in the expectation that the public good of connectivity should be provided as a commodity by private industry, but also in the various responsibilities imposed on private industry to serve the public good. Markets distribute connectivity as a good, but

54

Telecom Tensions

public policy steers markets toward public ends (such as a more equitable or competitive distribution). While regulatory capitalism provides a more accurate conceptualization of the relationship between state power and private industry than the theory of the neoliberal transformation, it remains a broad generalization that can only serve as a starting point for analysis. It is more helpful to specify particular governmental techniques in play, such as increases, decreases, or transformations in particular forms of regulation, rather than ascribing these to a governmental project as vast as neoliberalism or an order as diverse as regulatory capitalism. When we pay close attention to Canadian telecom regulation, it becomes clear that the logic of liberalization has itself shifted and been subject to dispute since the 1990s, with regular contests over the shape and extent of competition. I will examine these specific governing techniques and logics in the following two chapters, but first there is one more broad generalization to cover that is related both to neoliberal arguments and to the reality of regulatory capitalism – what I characterize as the “economization” of connectivity (see W. Brown 2015). In Canada, regulatory liberalization may fundamentally be about opening new domains to market competition, but this has been accompanied by a tendency to value the economy above all else, and to approach the problems of connectivity primarily in economic terms. In this context, intermediaries and individuals are conceptualized as market actors. This is demonstrated in the centrality that the crtc and the federal government accord to the “digital economy,” wherein the primary value of connectivity is not to enable citizenship, social relationships, or human flourishing, but for Canadians to “fully participate in the digital economy” (Government of Canada 2016).19 This economization of connectivity is important because it means that certain justifications (the “business case” for connectivity) and forms of knowledge (economic analysis) are privileged in Canadian telecom policy discussions and through the regulatory decisions of the crtc.20 But while other forms of rationality are often marginalized, they are not wholly excluded, and, as seen in the following chapter, may even be significant drivers of particular connectivity projects.

History of Canadian Intermediaries

55

ISPS IN CANADA TODAY : NET WORK TOPOGRAPHY AND POLITICAL ECONOMY

Today’s internet still operates on the basis of some decades-old protocols and principles, but it bears little resemblance to the arpanet21 of the 1970s that provided its foundations. The early isps of the 1990s are best understood as the product of telecom liberalization, rather than as the technological culmination of packet-switching technologies that developed over the previous decades. While packet-switching was a transformative technological innovation, the internet that most people experience today is the result of specific public policies that shaped its commercialization. The first Canadian isps included publicly funded academic networks, non-profits such as FreeNets (see chapter 3), and commercial organizations. There was considerable overlap between these categories, with partnerships between government and industry, commercial uses of academic networks, and links between academia and FreeNets (CA*net Institute 2001; Hunt 2014). These organizations developed at the margins of public awareness in the early 1990s, but by the middle of the decade the World Wide Web had expanded the internet into a network with broader appeal and new commercial possibilities, forcing regulators to decide key questions of political economy. Even where these transformations happened in the absence of regulation, the choice not to exercise sovereign control was a public policy decision. For example, some of the first Canadian isps depended on the “Centrex” digital line service offered by telcos to provide connectivity. Start-up costs and fees for use of the service were low, and numerous small iisps (including FreeNets) competed with each other by utilizing telco infrastructure.22 In 1995, iisps appealed to the crtc and the Competition Bureau when incumbent telcos withdrew the Centrex service, offered a replacement for a higher price, and entered the isp market.23 When they rejected the iisps’ appeal, regulators effectively decided what kinds of organizations were legitimate competitors in the new market, and what power incumbent intermediaries would have over their dependents. With incumbent telcos and cablecos now operating as isps, many smaller competitors were absorbed or went out of business. At the end of the 1990s and throughout the 2000s, a shrinking number of large isps came to control more of the market (cmcrp 2019). Even as

56

Telecom Tensions

the crtc mandated new forms of wholesale access to incumbent infrastructure in order to promote competition from new entrants (see chapter 4), regulators showed a permissive attitude towards this growing concentration. Consolidation throughout the telecom industry was not just a matter of larger companies swallowing smaller ones, but also included sizable mergers, such as the one that created telus out of bc and Alberta’s telco incumbents in 1999. By 2004 (when mts moved to purchase Allstream), one telecom consultant noted that there wasn’t much left to consolidate, with the exception of municipally owned telcos such as those scattered across Ontario, or the power companies that possessed telecom infrastructure (Wire Report 2004). But the trend towards the consolidation of giant firms has continued, with recurring efforts towards merging Bell and telus (Marowits 2007; Nowak 2009)24 and Bell’s purchase of mts (Jackson 2017a) testing regulators’ limits of acceptable competition. Internet connectivity in Canada today is therefore dominated by telco and cableco incumbents, which have been transitioning to fibreoptic access networks where there is a business case. Incumbents do compete with one another, but not all incumbents compete all of the time, since many are satisfied serving their established territories and see no value in duplicating facilities where another incumbent is present. Hundreds of iisps of various sizes are still scattered across the country, competing with incumbents in urban areas (where multiple isps often provide access) or focusing on less-profitable rural areas where residents lack choices. iisps have maintained or expanded their market share against the incumbents over the past decade (cmcrp 2019; crtc 2020), but their continued viability is enabled by the crtc’s mandated access regime. While this regime affords iisps a measure of protection against larger competitors, iisps typically remain dependent on the incumbents for wholesale connectivity. This dependence is tied to control over the internet’s topography. The internet’s critical bottlenecks are the fibre-optic routes to and from major internet exchanges in Vancouver, Toronto, and US cities. If an isp can find a path for its traffic to one of these ixps, it can negotiate directly with major global players: international long-distance networks and services such as Google, Facebook, and Netflix. While incumbents control paths to multiple internet exchanges, iisps typically lack such direct access. This means that iisps rely on incumbents to connect to the broader internet, giving incumbents tremendous leverage, and it is the crtc’s mission to keep this “market power” in check.

History of Canadian Intermediaries

57

While the crtc adjudicates disputes over rates and access, an important stake in these debates is the very kind of governing rationality at play in telecom policy. The crtc’s broad reviews of wholesale and basic services (crtc 2015c) are not just contests over what the networks of the future should be like, but also over the underlying objectives, justifications, and methods of telecom policy. Many have pointed to the Telecommunications Act or subsequent “policy directions” (Government of Canada 2006; 2020) as specifying these objectives and the means to achieve them, but portions of these texts have been cited in different ways to justify competing visions of telecom policy and contradictory governmental techniques. Whereas previous analyses have focused on the transformations of the 1990s, characterized as deregulation (Wilson 2000), neo-liberalism (Rideout 2003), or convergence (Winseck 1998), this work is situated in a political economy that has largely stabilized under the regulatory capitalism of the internet era. What exists today was not the end-state imagined when major policy shifts were initiated some thirty years ago, and rather than settling into market equilibrium, its stability is maintained through regulation. Liberalization is a process that is always incomplete, and in Canadian telecom it has become a mode of governance that is continually being recalibrated by a number of state agencies. The specific policies being pursued through intermediaries remain open to revision and expansion, but what has steadied is a political economy dominated by incumbents and a regulatory state that both legitimates and limits the power of these corporate giants. The crtc is the most visible state actor that cultivates and elevates the economic forces that telecom incumbents represent, while restraining incumbents through its claim to sovereign authority. Under regulatory capitalism, society is governed by an alignment of private market actors and the public interest, but this does not specify what the public interest is, or the regulatory techniques to achieve it. The following chapter addresses what happens when connectivity itself is conceived of as a “public good,” and the variety of techniques that have been deployed for improving connectivity where competition and market forces are deemed inadequate.

58

Telecom Tensions

3

Connectivity for the Public Good

How should the internet be governed? The question is fundamental to a wide array of other concerns, branching into more specific policy dilemmas and intense public controversies. For instance, should isps do more to assist law enforcement or copyright owners? Should isps practise “net neutrality,” by not discriminating amongst traffic? Should incumbent isps be required to treat smaller competitors as wholesale customers? Underlying these debates is the shared assumption that connectivity is, in and of itself, a good thing. Or, as crtc Chairman Blais articulated in 2016, that broadband is “vital” for Canadian society (crtc 2016a). The least controversial objective of telecom governance is to increase the level of connectivity. The policy consensus is that the greater the number of people who are connected, and the faster their connections, the better. The public policy goal of improving connectivity is rarely questioned; it is the question of how to achieve better connectivity that has generated intense debate. This chapter will consider what happens when connectivity is treated as a public policy objective, with isps as the instruments to achieve it. Although connectivity in Canada tends to be provided by private actors as a commodity, it also bears the characteristics of a public good, in that we recognize a collective benefit to our society being more connected, and that some people being excluded across a “digital divide” is an issue of public concern. In this context, intermediation means that as connectivity increases in importance, so does the responsibility to provide it – a responsibility felt by both isps and state agencies. Where private actors do not meet public needs, regulators can direct or shape their conduct towards these ends. Alternatively,

Connectivity for the Public Good

59

public bodies can play a more active role in building infrastructure and providing services for the public good. Efforts to provide connectivity as a public good create persistent tensions and role conflicts around isps’ expectations following telecom liberalization. Telecom networks share many of the characteristics of other kinds of traditionally public infrastructure, such as transportation networks and utilities. But because of the expectations applied to public organizations and private intermediaries, role conflicts result when commercial intermediaries provide connectivity as a public good, and when public organizations operate in a role similar to that played by commercial intermediaries. Ever since the 1990s, the governing rationality has been for intermediaries to compete with one another, and for connectivity to be treated as a private good allocated through market forces. This means that governments are reluctant to pursue connectivity as a public policy objective through dedicated public institutions, and those public organizations that play the role of intermediaries must carefully delimit themselves so as not to compete with private industry. Private industry, on the other hand, has repeatedly been enlisted toward public policy ends, but this has been accomplished by trying to make public policy profitable, either by designing incentives for improving connectivity or by simply paying private industry to build and operate infrastructure. The track record for such projects has been mixed in Canada, and the contradictions of profit-seeking actors serving the public good are manifest in these outcomes. However, role conflicts also often grant considerable discretion to the intermediaries that must resolve them, as seen in the range of approaches that have been taken to resolving the problem of connectivity. PURSUING CONNECTIVIT Y

Public policy interventions have typically tried to address gaps and inequalities in connectivity, rather than improving connectivity for all Canadians. While the internet may stretch around the globe, connectivity remains geographically uneven, often following the lines of legacy infrastructure. In some ways, these inequalities deepened in the 1990s, since the internet originally operated through a universally available telephone infrastructure, but was then extended through the cable video network and more exclusive fibre-optic deployments. In

60

Telecom Tensions

short, access to the internet has been and continues to be limited by access to its infrastructure, and this means that for many residents of a country as vast as Canada, connectivity remains a problem. Since the 1990s, successive federal governments have committed to extending and improving connectivity in various ways. Provinces have pursued their own broadband programs, and some municipalities have also taken steps to manage internet infrastructure for public benefit. Numerous private isps have applied for government support or competed for contracts to build networks where market forces have failed to deliver. Some not-for-profit and charity organizations have also worked to address gaps in access. Finally, as internet connectivity has become recognized as a prerequisite for social participation and access to essential services, there have been debates over whether Canadians have a right to expect a certain level of connectivity, and whether isps have an obligation to provide it. One view, more fully explored in the following chapter, is that connectivity is best pursued through private actors operating in a competitive market. But this rationality is frequently challenged by circumstances that can be characterized as “market failure.” The most evident example of market failure has been in the provision of rural broadband, an issue that closely echoes the debates over rural telephone service in the twentieth century. Connectivity has been lacking in many rural areas, since companies have been reluctant to invest in rural broadband where the “business case” is weak. However, this is not the only instance in which market-based solutions to society’s connectivity problems have fallen short. While access to high-speed and affordable internet is a problem for much of rural Canada, even urban areas face connectivity challenges. Some populations, such as wealthy and highly educated Canadians, are better able to use technologies to benefit their lives (Crang, Crosbie, and Graham 2006). As scholarship in the field of community informatics has repeatedly emphasized, merely providing internet access is a partial solution to the digital divide.1 Individuals need the skills to use tools and services effectively, and a level of understanding to see the possibilities that internet technologies enable. Improving connectivity is a straightforward objective when this is defined as connecting more devices to the network at higher speeds. The goal of connecting people is a more complicated public policy objective that requires more than a technical solution. People can face a host of disadvantages in accessing online services irrespective of the speed of

Connectivity for the Public Good

61

their internet connection. It makes little sense to say that someone is connected to the internet when they cannot exercise agency, because they are unable to effectively use their device as an interface, or have difficulty navigating a web browser. The telecom industry is not in the business of evenly distributing connectivity throughout society. Commercial isps are interested in gaining and retaining subscribers where it is profitable to do so, and see less value in helping these subscribers use the internet more effectively to meet their goals.2 Some companies have championed affordability for low-income families since 2013, and this form of businessled social assistance is now being extended across the industry by the federal government.3 A more common and longstanding approach has been for governments to contract commercial isps to serve the public policy objective of improved connectivity, or to have this role played by public institutions. In both relationships between the public good and private markets, some degree of role conflict almost invariably results. This chapter will review several types of public connectivity initiatives, namely publicly funded networks and isps that are actively engaged in using connectivity as a means to meet larger social objectives. A wide range of approaches and organizational forms have been used to address the problem of connectivity in Canada, and this chapter will include an explanation of some of the different configurations of public and private institutions that have come together towards this end. The following chapter will analyze the pursuit of competition as the dominant expectation for intermediaries under regulatory capitalism, including how this expectation applies to publicly funded networks, but first I must explain what the idea of connectivity as a public good entails, and the role of public or not-for-profit institutions in enabling connectivity. CONNECTIVIT Y AS A PUBLIC GOOD

The public good and related ideas (like the public sector and its public policies) only make sense as part of some kind of public-private distinction. This is an easy distinction to make in a general sense, but a difficult one to delineate specifically. Distinctions between public and private, or state and civil society, are not universally understood, but are tied to different historical and cultural circumstances (Rosenblum and Post 2002). The concept of “public policy” is rooted in an

62

Telecom Tensions

Anglo-American governmentality that does not always translate across borders and languages (Regonini 2017). It is also difficult to determine who comprises “the public” in a specific situation, given the many lines of difference and division in a society (Calhoun 1998). And yet, these terms can have powerful meanings for those who use them. Individuals working in different government agencies to improve connectivity have repeatedly told me that their efforts are in the public interest, providing both a motivation and a justification for why these municipal administrators, chief information officers, and community champions are engaged in a pursuit that has traditionally been left to the private sector. Public servants in different jurisdictions also do not generally see each other as market competitors, making it easier for them to collaborate and exchange information to serve their respective publics. It is possible to take a cynical view of the above and to see the public interest as a self-serving and fictitious conceit. We are, after all, deeply suspicious of politicians who claim to act on behalf of the public. Even private corporations can wrap themselves in the public good in order to better align with or influence public policy, such as when Bell’s charter was revised in 1882 to serve Canada’s “general advantage” (see chapter 2). In the US, Bell’s embrace of the language of civic ideals and the public interest during the period around 1913 was partly a response to the threat of nationalization (John 2010). Today, incumbent telecom companies emphasize that their investments in infrastructure serve the public interest, and that these investments would be threatened by regulations to make these companies less profitable (Rajabiun and Middleton 2015). These actors invoke the public good to point to something nobler than their own interests, or to argue that in pursuing their own goals, they are actually producing collective benefits. But just as the public and private realms are not mutually exclusive, we need not expect the public good to be a selfless pursuit, or to always be positioned as an alternative to market competition or the pursuit of profit. So what then is the public good? Most fundamentally, this notion points to something larger than particular or private interests. The public good is what concerns all of us, or at least those of us who are included in a given conception of “the public.” To identify with the public good is to identify with the universal and collective, even if in actuality there are no truly universal concerns, even if there is no single unified public, and even if people disagree on what is good. In

Connectivity for the Public Good

63

John Dewey’s famous formulation, distinguishing the private and public rests on the difference between two kinds of actions and their consequences, specifically those consequences “which affect the [private] persons directly engaged in a transaction, and those which affect others [the public] beyond those immediately concerned” (1927, 12). This leads to a conceptualization of the public as “all those who are affected by the indirect consequences of transactions to such an extent that it is deemed necessary to have those consequences systematically cared for” (16). By extension, public goods would be those “which are so important as to need control, whether by inhibition or by promotion” (15). This is as good a starting place as any, precisely because Dewey does not rigidly demarcate the public and private, highlighting that this distinction must be “deemed” into existence on the basis of consequences and perceived importance. Also, these introductory statements say nothing about how public goods are to be governed or how the public interest is to be represented, which remain deeply controversial questions with many possible answers. The public good is a contested concept: first, in the different ways its development can be traced historically (often beginning with Aristotle and continuing through Dewey); second, as a contemporary category that economists, policy scholars, and political theorists variously define and operationalize; and third, as an object of political struggle, where members of a society argue over what is public and what is good (Geuss 2009; Kaul and Mendoza 2003; Mansbridge 1998). In this third respect, what counts as a public good is politically contingent – defined and redefined through a political process (see Stone 1991), the outcomes of which come to represent the interests of the public. Underlying this dynamic construction of the public good are significant differences and disagreements, rather than the formation of a general consensus or unity (Calhoun 1998). In a liberal political order, to distinguish something as private is to argue that the state should not interfere with it (Geuss 2009). Public goods are linked with collective interests and shared consequences, but in a political economy that privileges market logic, public goods are often defined synonymously with whatever cannot be left to private industry. A standard definition of a public good in economic theory is something that is “non-excludable” and “non-rivalrous” in consumption; people cannot be excluded from having it, and one individual’s use does not make it more difficult for others to access (see Milleron 1972). In practice, whether something is a public good

64

Telecom Tensions

is a normative question that depends on political processes rather than strictly meeting these criteria. However, political judgments about exclusion are crucial to determine if it is acceptable for people to go without a particular good, be that health care, park access, or internet connectivity (Kaul and Mendoza 2003). In regard to connectivity, the question is whether we would accept a situation in which some people want internet access, but cannot have it. If the answer is no, then public policies to promote connectivity are required. It is important to recognize that in Canada and elsewhere, public goods may be provided by private or non-state actors: not-for-profits, various kinds of collectives, or for-profit businesses. Public goods can also be distributed in ways that involve harnessing markets or “marketlike mechanisms” (Ostrom and Ostrom 1999). Governments around the world have championed a variety of pro-internet public policies, many of which complement a rationality in which private industry takes the lead in building and operating infrastructure, with public funds used to make investing in certain geographic areas more profitable (Gómez-Barroso and Feijóo 2010). To say that internet connectivity is a public good is to say that it should be publicly accessible. While this might require a significant departure from market logic, such as a publicly owned utility, it has been more common to see modest regulatory adjustments that try to steer markets towards the public interest. Proposals to treat the internet as a public good have surfaced in Canada at all levels of government, including crtc discussions about whether all Canadians should have access to broadband, provincial and regional efforts to extend connectivity by funding new networks, and debates over municipalities’ roles in managing broadband as public infrastructure. Proponents of public policies to expand connectivity may not explicitly invoke the public good as a rationale, but they associate internet access with broad benefits and see a role for public agencies or public funding. Typically, this means aligning private isps with the goal of improved or more equitable connectivity, whether through government-imposed requirements or through government contracts. More rarely, public bodies (namely municipal governments) have undertaken the task of building and operating entire networks. However, this approach contradicts the prevailing liberal ethos that networks should be owned and operated by private companies, and is typically criticized as government interference in the market.

Connectivity for the Public Good

65

In instances where isps provide connectivity as a public good, connectivity is rarely promoted as an end in itself. Instead, broadband is understood to be an enabler towards other public policy objectives. Particularly in rural regions suffering various forms of decline or inequality as compared with urban Canada, broadband is promoted as a way to attract desirable elements to communities (including new residents and businesses) and to overcome various challenges posed by distance and isolation. During the covid-19 pandemic, broadband allowed many Canadians to work, study, and maintain social relationships from home. Information flows are crucial for markets, decisionmaking, and contemporary society in general (Castells 2000; Melody 1987). As articulated in a court decision that granted charitable status to the Vancouver FreeNet (see below), “information is the currency of modern life,” the free exchange of which “has long been recognized as a public good” and “essential to the maintenance of democracy.”4 Ideally, internet connectivity enables people to participate in the public sphere, to access and share valuable information, and to benefit from the public good of journalism (McChesney 2013). There are public benefits, or what economists call positive externalities, to more people being connected as part of a communications network (Katz and Shapiro 1994). The telephone network grew more valuable as it connected more people and services, until universal access could largely be presumed. Similarly, agencies and organizations have increasingly been presuming that citizens, users, and customers can be reached through the internet. The more closely the realities of internet access match this assumption of universality, the fewer people are unjustly excluded from various opportunities and forms of civic participation. While benefits such as greater community solidarity and democratic participation have been presented to justify some efforts to expand public connectivity, in the current Canadian milieu, the value of broadband tends to be rationalized in economic terms. New networks are “sold” to potential backers as a way of improving competitiveness, enabling local entrepreneurship, and attracting new businesses. A particularly effective argument in favour of municipal networks is the promise to reduce municipal spending on contracts with private isps, possibly even generating municipal revenues through publicly owned infrastructure. Economic arguments tend to carry the day, and are far less controversial in Canada than notions of a “universal right” to internet access or the idea that broadband should be regulated as a utility.

66

Telecom Tensions

Publicly funded networks are typically justified as exceptions to the rule of private capital. The most uncontroversial of these justifications are that public networks serve the needs of public institutions, and that public investment is warranted where it can ultimately help private industry to thrive. Both of these rationales supported the internet’s early development through publicly funded research and education networks (rens), which connected Canada’s public universities with a mandate to promote private-sector innovation (CA*net Institute 2001; Shade 1994). Today, these rens survive by serving research and educational institutions, and to a lesser extent by facilitating business innovation and entrepreneurship. By limiting their competition with commercial isps, public-sector networks can avoid the criticism that they are unfairly interfering in the market for connectivity. Likewise, when intermediaries work towards the public policy objective of reducing inequalities in connectivity, this can be justified by the argument that when the market fails to meet everybody’s needs, alternate approaches are required. However, tensions and contradictory pressures remain where the public good meets private interests. Public-Private Tensions and Role Conflicts Our ideas about connectivity are often in tension with two divergent public policy assumptions. The first assumption is that connectivity has some of the characteristics of a public good, in that it has broad and indispensable public benefits. The second assumption is that connectivity is best provided through private networks as a commodity.5 When combined into a single rationality, this is expressed as the ideal that everyone should have access to broadband, and that this universal access should be the product of profit-seeking choices by market actors. The components of this rationality diverge because markets do not provide a uniform distribution of goods. As a result, markets for connectivity are governed in the public interest, and public bodies fill gaps that markets leave behind. A role conflict results when a publicly owned isp serves customers who might otherwise turn to a privately owned isp, given the fact that state actors are expected not to compete against private actors. The idea of governments providing public connectivity, while relatively common in the early twentieth century, carries a “whiff of quasisocialism” following liberalization (Lombardi 2006). Critics, particularly those associated with private industry, say that publicly owned

Connectivity for the Public Good

67

intermediaries interfere in the market or make unfair use of public money for competitive advantage (Jackson 2017b; Lie 2003, 35). US telecom companies have repeatedly waged legal and political battles against public networks (Guttentag 2018). In Canada, conflicts have been more muted, but the prevalent idea that government should stay out of the telecom business can prevent public connectivity projects from even being considered. Most public networks that have gone ahead since liberalization are justified as non-competing, or as somehow promoting competition through open access (see chapter 4). A different role conflict can occur when a privately owned isp is paid to serve the public good, which can also violate the expectation of fair market competition, since the state partner receives public funding to build infrastructure that can entrench their position (R. Shaw 2012). Intermediaries and state agencies can choose to shrug off such criticism, but there is a more intractable role conflict in such public-private partnerships, given the contradictions between market logic and the public good. Public-private partnerships in telecom often impose obligations on the private partner to limit profit-taking, such as mandating that any infrastructure benefiting from public funding be open to use by competing intermediaries, or prohibiting the company that built the infrastructure from using it to provide services. Private companies seek profits, and so the terms of a public-private partnership must be carefully written to foreclose on the possibility of for-profit exploitation. Companies that are expected to both maximize profits and serve the public may prioritize the former, as long as they can argue they are meeting their legal obligations. Contracts become a means of disciplining private actors to serve the public good, but contracts do not enforce themselves and must be carefully written to be effectively used as legal instruments. For example, New York City has been engaged in a long legal fight with Verizon over whether or not the company has lived up to its end of a deal to build a fibre network. There, the dispute hinges on the definition of what it means for a home to be “passed” by fibre – a term not actually defined in the contract (Brodkin 2017). In Alberta, the terms of Axia’s contract to operate the SuperNet (Government of Alberta, Bell Canada, and Axia SuperNet Ltd. 2005) restricted the ability of the company to use government-funded facilities for their competitive advantage, by stating that Axia could not offer services as a retail isp over the network. Axia did eventually enter the retail market but claimed this was consistent with its legal obligations. Despite concerns, government found

68

Telecom Tensions

its ability to audit the contract or to determine non-compliance was too limited to take action, and this troubled partnership continued on for years amid “differing interpretations of contract terms” (Auditor General of Alberta 2018, 9). When public connectivity is treated as a contract between parties rather than as public policy (Kozak 2010), the public interest can only be extended as far as the limits of contract law. In short, saddling private, for-profit companies with public obligations that limit their ability to profit introduces a role conflict that must be carefully managed in the public interest. The alternative of public intermediaries competing with the private sector contradicts the rationality of liberalization. Given these tensions, there are various possibilities for governing connectivity through both public and private actors, but before discussing such projects further I will consider an alternative approach that was quite different from the top-down implementation of public policy. The FreeNet movement of the mid1990s provides some of the best examples of isps providing connectivity towards the public good, under an equitable and redistributive governmental rationale that stands in stark contrast with today’s dominant (commercial and public) arrangements. FreeNets: Distributing Connectivity as a Public Good The FreeNet6 movement was an extraordinary phenomenon that (alongside rens) helped define the early years of the internet in Canada. Eventually, ideas of free access and community networking were overtaken by the business models of the commercial isps that remain dominant today. However, in the heady days of the mid-1990s the future of the public internet was still wide open, with vast but largely untapped possibilities, and so it seemed quite reasonable that connectivity should be conceived and provided as a public good. Groups of individuals scattered across the country, learning from one another, pursued an egalitarian and community-minded vision of digital networking. The details of these visions varied from case to case, and access to the broader internet was typically only one of the services provided by these organizations. But in general, the idea was that connectivity should be available free of charge, or that cost should not be an obstacle for anyone wishing to be connected. FreeNets (sometimes identified under the broader category of “community networks”) depended on volunteer labour, corporate donations, and government assistance. Instead of subscribers, FreeNets had members, some of

Connectivity for the Public Good

69

whom donated a regular amount and thereby subsidized those with free access. In addition to making basic connectivity affordable to all, these not-for-profits developed online information that was locally specific, coordinated volunteer assistance, and promoted the use of connectivity in the service of lofty social values, such as community development, democratic participation, and “digital inclusion” (Graham 2011). The earliest FreeNets preceded the commercial internet; the first was established in 1986 in Cleveland, and 1992 saw the birth of the Victoria Free-Net Association as well as Ottawa’s National Capital FreeNet (ncf; see Shade 1994). In August 1993 the International Freenet Conference was held in Ottawa, and the following year Telecommunities Canada was founded as an umbrella group and champion of various kinds of community networks (Clement et al. 2012). It is difficult to say how many FreeNets and community networks existed across Canada at the movement’s peak, but around forty are listed on a webpage archived in 2001 (Scott 2001).7 Today, we might only be able to count three or four operational descendants of those early organizations (most notably in Vancouver and Ottawa). These survivors are a far cry from the initial optimism and revolutionary potential of the FreeNet movement, which was never able to emerge from the dial-up era once incumbent isps made the shift to broadband. For Bodnar, the inability of so many of these networks to remain viable into the 2000s indicated that “the once-common objective of providing free or low-cost Internet access and communitybased resources in the service of public access ... [was] no longer as straight forward as it once seemed” (Bodnar 2007). And yet, some of the core problems that FreeNets tried to solve – inequalities in connectivity, effective use, and the need for community development – are no less relevant today. However, our capacity to “think otherwise” about the future of the internet has narrowed as its political economy stabilized. The mid-1990s were a crossroads, and the path then taken onto the privately managed information highway has precluded further consideration of alternatives. In retrospect, the birth and apex of the FreeNet movement occurred during a brief window in the internet’s development, specifically between 1993 and 1994. This window saw the internet outgrow its publicly funded incubator and extend into public access, but only lasted until the telecom industry stepped in with its own business plan. One of the founders of the ncf, Jay Weston, remarked that if the

70

Telecom Tensions

organization had been established in 1995, “it is unlikely that the telecommunication industry and various governments would have been so unconcerned, or benignly neglectful, of what the freenet implied” (Avis 1995, 86). Also in 1995, the internet’s US backbone switched from a publicly funded network (nsfnet) to several interconnected commercial backbones, effectively privatizing the internet. The World Wide Web emerged, and FreeNets such as ncf had to decide whether (and how) to provide access to it. Private isps were founded across the country to serve the growing demand, and by 1996 Canadian telecom incumbents had entered the market and become dominant isps. In 1997, early Canadian notions of the internet as a public good received both a last gasp and fleeting government recognition. The federal government’s Community Access Program (cap) had been rolled out to tackle the digital divide, by providing internet access from public locations such as libraries, community centres, and schools. While administered by Industry Canada, cap worked in partnership with different levels of government and non-profit groups who operated these sites (Clement et al. 2012), resulting in differences of opinion about the internet’s potential. In the context of these political controversies, Industry Canada partnered with Telecommunities Canada to release a Framework for Co-Operation ... to enhance the ability of Canadian communities to utilize electronic public space. This remarkable document shows that even at the cusp of the internet’s commercialization by incumbents, Industry Canada was willing to acknowledge a vision of the future in which connectivity would promote local autonomy, and one in which the “community network extends the idea of community into a shared electronic public space, a new not-for-profit transaction space” or “public commons” (Telecommunities Canada and Industry Canada 1997). According to a longtime champion of community networking involved in the process, this official recognition was actually an attempt to neutralize a perceived political threat from Telecommunities Canada. However, the community organization lacked the energy and resources to implement its part of the ambitious agreement,8 and afterwards, it was clear that the federal government’s public policy approach to connectivity favoured the commercial rationality of incumbents. At that point, FreeNets were no longer charting the future of Canada’s internet, but simply trying to adapt to the changing environment.

Connectivity for the Public Good

71

FreeNets offered a competing vision for intermediaries’ governing rationality; opening internet access to all, developing information and skills that were tailored to local needs, and directing group energies toward social objectives. They were able to ride the first wave of public interest in the internet, and while they seemed to be an open-ended instrument for the public good in a networked era, a lack of stable funding and the rapidly changing environment left FreeNets and other kinds of community networks struggling to stay afloat. A dependence on incumbent telco infrastructure left them vulnerable as the material underpinnings of the internet were upgraded and user expectations changed. In general, those that did survive transitioned to functioning as low-cost isps, rather than the community-development organizations that were originally imagined. With high-speed internet proliferating, and as government funding proved unreliable or dried up entirely, these organizations had to become sustainable on the basis of regular member contributions.9 Still, FreeNet descendants such as ncf continue to play a unique role in society unmatched by commercial isps. These survivors have found a niche by serving certain marginal populations as an “isp-oflast-resort,”10 supported by those who prefer a community-based nonprofit to an incumbent isp. ncf depends on volunteers for many basic tasks, and would cease to exist without them. Members who have problems can come into the organization’s storefront, and receive more personal attention than the customer service provided by commercial isps.11 Those members who pay for connectivity or donate to the organization, support services for those members who cannot. And while the few FreeNets that have weathered the transition to being basic low-cost isps have limited their efforts in public outreach and digital literacy training, their broad mandate and nonprofit status leaves the possibility open to future efforts. Even the original vision of FreeNets as information distribution platforms is far from obsolete. True, there are now a wealth of online sources of information as compared to 1994, but these often lack local context, and there remains ample opportunity for intermediaries to better address the information needs of local populations. As one example, in September 2014 Vancouver Community Network (vcn, formerly the Vancouver FreeNet) launched the Street Messaging Service to send locally relevant updates to street-involved people (Vancouver Community Network 2015). Social media platforms can perform a

72

Telecom Tensions

similar function, but these services are designed and governed according to their own priorities rather than those of local communities. Local values are sometimes used to justify new public connectivity projects, but a more prominent rationale has been addressing inequalities, or bringing residents of an area up to a level of connectivity available elsewhere. BRIDGING THE DIGITAL DIVIDE : ISPS AS INSTRUMENTS OF EQUITABLE CONNECTIVIT Y

The early FreeNets were based on dial-up access. They took advantage of the existing telephone infrastructure and could be established using modest hardware investments. As the incumbent telecom infrastructure transitioned to high-speed broadband, much of the original FreeNet vision was left in the dust. And yet, this is not because connectivity became more equal across Canadian society. Instead, a new sort of inequality was created based on the distribution of material infrastructure. Certain forms of connectivity were either present or absent in an area, and those who remained on dial-up often had no other choice. Mobile connectivity required towers and spectrum, and digital flows were underpinned by fibre-optic backbones. In some areas, fibre-optics were extended all the way to a subscriber’s premises, but only where isps decided this investment in infrastructure would be worthwhile. With every passing year, connectivity is required for a growing number of social interactions and relationships. It becomes harder to opt out of these digitally mediated interactions without losing access to important services and opportunities, and this also brings inequalities in connectivity into starker relief. The internet is now society’s connective tissue, and poorly connected segments will atrophy. The more essential connectivity becomes, the less need there is to justify its benefits for everyday life, since the consequences of lacking access to broadband become readily evident. Rather, the key points of debate have come to hinge on the question of how to best provide connectivity to everyone, with all levels of government retaining considerable discretion in this regard. The governing rationality of liberalization assumes that internet provision should result from a healthy, competitive telecom industry. But many Canadians have been poorly served by this market-based approach, leading to alternate models of connectivity such as those discussed in this chapter.

Connectivity for the Public Good

73

Inequalities in connectivity have often been characterized as a digital divide between privileged and disadvantaged groups. In academic elaborations of this concept, authors have identified disparities in connectivity between rich and poor, old and young, urban and rural, men and women, as well as different racial and ethnic groups (Howard, Busch, and Sheets 2010; van Dijk 2006). In short, all of the traditionally recognized social inequalities are reflected in unequal access to the internet, or in unequal ability to make “effective use” (Gurstein 2003) of new technologies. Many of the networks and organizations discussed in this chapter operate to bridge these digital divides. These isps are either the outcomes or the instruments of policies to distribute connectivity in a more equitable manner. In many cases, public sources of funds (tax revenues, infrastructure loans) are directed to building out networks in areas where internet penetration has been deficient. Discussions of the digital divide would not vanish as the majority of Canadians gained internet access, but the conversation would remain focused on the material preconditions of connectivity. Even as academic discourse came to conceptualize the problem as being deeper than a lack of physical access, public policies that had funded various community networking initiatives and social supports for technology use were abandoned (Clement et al. 2012). The public policy approach that survived was more narrowly directed toward extending private infrastructure. Access was identified as the ability to choose and consume a commercial internet service, and the role of public policy would be to create an environment favourable to private infrastructure development. The types of intermediaries considered in the rest of this chapter have a narrower mandate than the early FreeNets, and the public policies that have supported them focus largely on physical connectivity. And yet, a broader argument of this book is that the roles of internet intermediaries are not inherently circumscribed, but are open to expansion. The fact that the present belongs to the incumbent isps and their visions of what an intermediary should be, is the result of the political economy of the internet described in the previous chapter. The dream of the FreeNet remains associated with the era of landline telephones and dial-up modems, but there is no technological reason why the sorts of intermediaries that were imagined in the mid1990s could not operate over leased access to fibre-optic cable or wireless infrastructure (see Antoniadis 2018). Finally, even as incumbent

74

Telecom Tensions

commercial networks became accepted as the natural order of things, and public policy was limited to fixing the gaps these networks left behind, progressive ideas about the social roles that intermediaries could play never vanished. Arguments that local intermediaries can be instruments of equality and development have been a basis for the establishment of numerous municipal, regional, and provincial networks in Canada. Publicly Funded Networks to Meet Municipal, Regional, and Provincial Needs Some Canadians live with inferior levels of connectivity because geography affects the business decisions of isps to invest in networks. Upgrading the network in a region might win a company more customers, or provide a competitive advantage against inferior networks, but digging trenches and installing equipment quickly turns into a multi-million-dollar expense. As with the issue of universal telephone service in the twentieth century, there are limited reasons for private isps to serve remote rural regions, where either the costs of building infrastructure cannot be recovered, or returns are smaller compared to urban areas. This has left many rural communities wondering how to improve their situation without waiting for the arrival of market forces. Where markets fail to produce the desired outcome, governments are often called upon to step in, and improving connectivity is a public policy problem that any level of government can take on. But the federal government, which has ultimate authority over telecom in Canada, has historically been unwilling to provide the billions of dollars that would be needed to fund nationwide broadband infrastructure.12 Federal governments have recurrently provided one-time funding for specific projects, such as seed money for a local isp in an underserved town. These grants have worked to plug individual holes in broadband connectivity (see Rajabiun and Middleton 2013; Van Gorp and Middleton 2010), but they do not provide lasting support. If the private sector does not invest of its own volition, and when higher levels of government cannot be counted on, some municipalities,13 provinces, and regions14 have pursued their own broadband strategies. A common approach has been for governments to split the cost of new infrastructure with local isps. The result can be a grant of public money, a public-private partnership, a contract impos-

Connectivity for the Public Good

75

ing public interest obligations on the isp, or the creation of a cooperative.15 In these situations, government does not end up owning a network, but pays to align a privately owned network with public policy objectives. More rarely, a new intermediary is created from within the public sector. While public ownership is a reasonable approach for public goods, it is also contrary to the rationality of telecom liberalization. As previously mentioned, one way to avoid this conflict is for publicly owned intermediaries to limit their competition with the private sector. They do this by serving an exclusive class of customers, such as public-sector clients, or by operating infrastructure for use by the private sector. Some public intermediaries have developed from a local government’s initial desire to connect its facilities, followed by the realization that the same infrastructure used to link public buildings could be used to improve connectivity within the municipality. Local incumbents can usually meet a city’s telecom and networking needs, but administrators in various Canadian municipalities have decided they could reduce costs by building and operating such a network themselves, or entering into an agreement with a private partner to build and operate a network. Because the cost of installing a hundred strands of fibre is trivial in comparison to the costs of digging up the streets, a network built for a narrow municipal purpose can easily be installed with greater capacity than necessary and then utilized in other ways, such as connecting other sorts of buildings that happen to lie near its path. This approach does not result in a last-mile isp that connects to all possible kinds of endpoints (homes and businesses) in a city, since achieving complete coverage radically increases capital costs. Instead, a municipality begins with the limited ambition of connecting particular sites (city hall, fire station, recreation centre), and builds on that fibre footprint. One influential model of this kind of municipal network exists in the Vancouver suburb of Coquitlam, with its QNet network. The project began as an effort to connect Coquitlam’s traffic lights together, with the network extended to municipal buildings in order to reduce the cost of relying on incumbent telecom services in 2004.16 The city installed more fibre strands than it required, which it turned into a fibre-optic backbone that private isps can pay to use in order to serve their customers. The QNet model is that of an “open-access” network, meaning that the infrastructure is open for other intermediaries to

76

Telecom Tensions Municipal M unicipal F Facilities acilities

ISP IS P

IInterconnection nterc onnection Facility F acilit y / IXP IXP

Last L ast mil mile ile

Middle-mile M iddle - mile ISP IS P

PoP Po P

Last-mile L ast- mile ISP IS P

Last L ast mil mile e

Figure 3.1 Open-Access Municipal and Middle-Mile Networks In an open-access model, internet infrastructure can be used by multiple competing isps. In a municipal open-access or “dark fibre” approach (top), a municipality builds a network to connect its own facilities, and extends this network to homes or businesses which can be served by an isp making use of the municipality’s infrastructure. In a middle-mile approach such as the SuperNet (bottom), a middle-mile network is built to transport packets over long distances and to connect to the rest of the internet, and last-mile isps can build their own networks to connect to it at designated meet-me-points or points-of-presence (PoPs).

provide services over. In industry terms, the municipality installed “dark fibre” under its roads, and various parties can pay to “light” this fibre, or obtain leased access to the infrastructure. Some version of this model has been adopted by a number of municipalities in Western Canada.17 A similar logic has informed the design of regional networks including Alberta’s SuperNet, the Columbia Basin Broadband Corporation in bc, and Ontario’s eorn and swift. Rather than connecting directly to homes and businesses, these broadband projects were developed to support connectivity for government facilities and private-sector isps. The public sector’s involvement in telecom can thereby be justified as a way of enabling private-sector competition. The SuperNet, for example, was built at a time of provincial bud-

Connectivity for the Public Good

77

getary plenty to connect public facilities across Alberta, to eliminate the price difference between urban and rural connectivity, and to cultivate local competition. It was imagined, at the turn of the century, as an open-access network that would be used by public institutions, incumbents, and iisps to move packets with ease between Alberta’s cities and the hundreds of small communities spread across the province. A government-owned-and-operated network was out of the question, given the policy orientation through the 1990s of privatizing provincial infrastructure wherever possible and promoting competitive markets (Mitchell 2007). Instead, a locally based “middlemile” isp (Axia) would manage and operate the network, selling wholesale bandwidth to all other isps, and effectively becoming a private-sector steward of public connectivity. The company would be contractually aligned with public policy objectives, such as equivalent pricing for rural and urban areas, and last-mile isps would bridge the gap between SuperNet and individual homes and businesses. In this case, market failure was not used to justify public ownership and management, but this was instead portrayed as an attempt to reconstitute the market. As Axia’s ceo said at the time, “We make money when the government gets its policy outcomes. We make money by creating competition in rural Alberta as opposed to making money by having limited competition” (Collison 2003, 3). While the results did not exactly measure up to the vision and the project was marred by numerous contractual disagreements,18 the SuperNet shows how public connectivity projects can proceed even within an exceptionally pro-business regime of regulatory capitalism. A more-involved public-sector model for providing connectivity has been implemented in Olds, Alberta, as mentioned in chapter 1. In this approach, the government-backed intermediary competes directly with the private sector across the last mile, defying the logic of liberalization. However, establishing a full-service isp was not what the project’s champions originally planned when they began working through the non-profit Olds Institute for Community and Regional Development (oicrd).19 In the early years of the project, the plan was for an open-access fibre network for incumbents to utilize. In the end, no incumbent partner was willing to provide services over a network built by the town, and Olds was left largely on its own (Settles 2012). Olds is hardly unique in Canada as a municipality that has pursued improved connectivity as a public policy objective. It is unique in how it has pursued this objective – by establishing an institute for eco-

78

Telecom Tensions

nomic development, which conceived of and founded an isp that extended a fibre-optic network throughout the town, deployed public Wi-Fi, and offered its expertise and services to other Alberta municipalities.20 The creation of O-Net in Olds shows that the rationality of liberalization can be challenged over the last mile, and related lessons can be drawn from the small number of other municipal isps scattered across Canada that also serve their citizens. However, these either operate in communities that are smaller (and less profitable) than Olds, or serve only particular kinds of subscribers, such as organizations and businesses. With the exceptions of Prince Rupert and Thunder Bay, which never privatized their hundred-year-old telephone companies (now CityWest and Tbaytel), urban municipalities in Canada do not extend connectivity to residential subscribers, who remain the responsibility of the private sector. THE CHALLENGE OF PUBLIC CONNECTIVIT Y

So what kind of governing nodes are public isps? What sets them apart from private-sector intermediaries? What is the rationality of these actors, or toward what ends do they govern or expand their networks? In some cases, the differences can be minor. The SuperNet was entrusted to a private company, which has since been acquired by Bell. Municipal isps can be organized along corporate lines, and run to maximize revenues.21 But they are often much more than profitgenerators, and their champions may be government bureaucrats, social visionaries, or dissatisfied citizens convinced that they can do better than what is offered by private industry. Local connectivity projects can arise from local needs, much as the independent telephone companies did more than a hundred years ago; instead of a shopkeeper installing lines to speak with clients, today’s local entrepreneur might be a rancher who invests in a wireless tower for their business, and sells excess bandwidth to neighbours.22 Municipal isps and “community networks” (including co-operatives) are often guided by notions of collective interests or the public good. While their sovereignty over local affairs is circumscribed by higher levels of government (including the crtc), municipal governments are explicitly mandated to operate in the public interest, and to serve all of their citizens.23 Although few municipalities see their role as ensuring that every citizen has access to broadband, many recognize that telecom is interrelated with a host of collective or public concerns, such as social

Connectivity for the Public Good

79

and economic development,24 and management of physical infrastructure such as roads, power, and water (see Federation of Canadian Municipalities 2009). Unlike private firms, public intermediaries such as those discussed throughout this chapter can be established with the explicit purpose of serving public policy objectives, but these goals are as open-ended as the public interest itself, and place few limits on the possibilities for these institutions. A municipal intermediary or regional network can be founded to bring broadband into the lives of all citizens, to support innovative social programs and technologies, or to invigorate rural communities. Politics can be a forum for a diverse range of interests, and the values and rationality that guide the public pursuit of connectivity cannot be foreclosed on. Private industry selectively deploys networks where profitable, but public agencies work to fill the gaps, guided by the view that connectivity is too important for the disadvantaged to rely on market logic. Where private industry maximizes profits, public connectivity projects might increase public revenues, but are just as likely to encompass notions of collective development, local autonomy, equitable and accessible connectivity, or attracting and retaining residents. When resources and support are provided from public sources, intermediaries are accountable to the sorts of outcomes and terms specified by these arrangements, such as connecting remote or disadvantaged populations, or serving public-sector clients. Public intermediaries’ dependent links to state agencies can be as recipients of funding, or as beneficiaries of special regulatory treatment, or intermediaries may operate under the auspices of a public authority (as QNet operates under the City of Coquitlam). Even when the intermediaries are non-state or civil-society organizations seeking to promote local autonomy, this is typically only possible within a framework defined by state institutions and by mobilizing other governing nodes, such as municipal, regional, and provincial government agencies. Finally, because they are often in a marginal position in telecom’s political economy and are less likely to see each other as market competitors, public institutions can cultivate cooperative relationships with one another, where they confront common challenges, benefiting from combining efforts or interconnecting networks. This chapter has demonstrated how adaptable intermediaries can be to various public policy objectives and how the dream of connectivity can promote locally distinct values. I will return to these ideas,

80

Telecom Tensions

and the possibilities engendered by alternative models of connectivity, in the concluding chapter. However, it should be noted that while there are numerous examples across Canada of internet access being treated as a public good, the various dreams and desires attached to connectivity can exceed what is achievable, and project outcomes often fail to meet expectations. Many planned public networks were never realized, some were sold to private industry after completion, and some (such as SuperNet) fell well short of the transformative rationale used to promote them. Relative successes, such as QNet and O-Net, are a testament to challenges overcome through long-term dedication as well as the presence of fortunate circumstances.25 However, the material and operational challenges of building and operating networks are not the main obstacles to the spread of connectivity as a public good. Rather, it is the very notion of connectivity as a public good that contradicts the governmental rationale underpinning telecom liberalization: that connectivity is best provided through private networks as a commodity. It is this contradiction that has become the main obstacle to public connectivity projects, or an enduring source of tension within their constitution. The tension between public connectivity and private infrastructure is evident in recurring debates over how to make a “basic” level of broadband available to all Canadians, or the careful line that rens and open-access networks have to walk to avoid competing with private isps. Publicly funded networks and government-mandated connectivity run contrary to the rationality that the public good is served by government sticking to what it “does best,” and staying out of realms delegated to the private sphere. Governments have generally taken this to mean a reliance on market forces where possible, with public-sector intervention only appropriate where the market has failed, and a preference for aligning private intermediaries with public goals. While there continue to be calls for greater government involvement in providing telecom services or building and operating networks, these arguments take place at the margins of public discourse. The assumption is that even if broadband is now some sort of essential utility, it is best provided by the private sector. Exactly what competition means in this context, and how a particular understanding of competition has been regulated into being and imposed on Canadian intermediaries, will be the topic of the next chapter.

Chasing Competition

81

4

Chasing Competition We all thought that when we got there, when competition happened, there’d be no more regulation. Well, here we are. Ken Engelhart, 5 June 2012

Among isps’ roles and responsibilities as instruments of public policy, none have sustained a greater volume of heated debate within Canada’s telecom industry than those intended to promote competition. Since the 1990s the crtc has set the rules for how private companies must relate to each other and has granted non-incumbents access to key infrastructure, in the hopes of fostering a competitive telecom industry. These efforts were meant to create the conditions for a new structural arrangement, which would reorder the incumbent-dominated telecom landscape inherited from the monopoly era. Only once such an outcome was achieved would regulators be able to step back and allow market forces to govern. However, rather than retreating from the market, the crtc has become ever more entangled in the relationships between isps, regulating the role conflict imposed on companies that are required to serve their competitors, in what I characterize as a state of “competitive dependence.” In short, incumbents are expected to provide the means for smaller companies to survive, while simultaneously competing against them. What was meant to be a temporary crutch to help the industry move into a deregulated world has become an enduring constraint on the forces unleashed through liberalization. Early hopes that liberalization would result in deregulation and a free, competitive market have been replaced by regulations that keep the power of incumbents in check and preserve something resembling the status quo.

82

Telecom Tensions

This chapter explains the Canadian telecom regime of “regulationfor-competition” (Levi-Faur 1999), its goal of creating competition where it is lacking, and the resulting state of competitive dependence. These developments can be understood in the context of regulatory capitalism, combining aspects of neoliberalism with new forms of state regulation. According to David Levi-Faur, “[w]hile neoliberal systems of control are manifested in the marketization of social and economic life, regulation supplies the means to promote, sustain, and control marketization” (2006, 497). Competition is therefore treated as a condition that can be regulated into existence through the obligations that participants in the market have toward one another, rather than as an alternative to regulation. In this context, “intermediation” refers to the process by which powerful intermediaries are saddled with responsibilities towards their competitors. However, a sufficiently competitive market can stubbornly refuse to emerge despite a regulator’s efforts. As a result, while regulation-forcompetition is still justified as a way to create a sufficiently competitive industry, the more immediate results are limits on the exercise of power by certain actors. Without such limits, further concentrations of power and industry consolidation would predictably follow. In other words, rather than forging a bold new liberalized world, regulation-forcompetition has kept the market relatively steady and prevented the ultimate triumph of the giant firm. In the face of two unacceptable policy alternatives – further concentration of powers or a forced structural separation of powers – the mandated access regime and its state of competitive dependency becomes a middle road and a default choice. I argue that the enduring contradictions in regulation-for-competition are the unplanned consequences of failed public policy, but failure to effect change and the preservation of the existing order can also be redefined as success. First, I will explain how current expectations of isps have been shaped by liberalization, through definitions (and redefinitions) of competition, and delineations between those intermediaries that are expected to compete and those that are prohibited from doing so. THE IDEA OF COMPETITION IN CANADIAN TELECOM

Competition is a problematic idea, which is to say that the history of the concept is tied up with ways of expressing different problems,

Chasing Competition

83

including socialist critiques of early capitalism and neoliberal critiques of market regulation. Socialists decried the harmful social consequences of competition, and neoliberal economists sought to either protect market competition from state interference, or promote competition through public policy (Gane 2020). Today, competition is something that many public policies are designed to cultivate, and Canadian telecom regulators look to it as a fundamental value (Government of Canada 2020). But Canadian telecom policy continues to be shaped by decades-long struggles over the question of where competition should be allowed, and what forms it should take. For much of the twentieth century, competition in Canadian telecom was seen as inefficient, wasteful, and undesirable. Rather than seeking to have multiple companies operating in a given market, the brc (Canada’s first telecom regulator) actively took steps to suppress competition by discriminating against companies seen to be competing with Bell. Early debates over whether to allow competing telephone companies or whether to treat the telephone as a “natural monopoly” (Dagger 1910) were gradually settled in favour of the latter view, which held that competition in telephones was undesirable and unsustainable. This principle was not only espoused by Bell and its regulators, but was also echoed by municipal governments, which actively opposed competing telephone companies operating over city streets (MacDougall 2013). By 1925 all direct competition with Bell had been eliminated and the number of independent companies serving rural areas gradually dwindled through the twentieth century (Babe 1990). Provinces beyond Bell’s territories in Ontario and Quebec had monopolies of their own, which formed dependent and cooperative links with Bell as the century progressed (Rens 2001). Regulators set rates for telephone service directly, and the problem of telecom companies competing for space on utility poles was no longer the concern it had been in the late nineteenth and early twentieth centuries. Overlapping wires and fragmented networks were replaced by an integrated system of regional monopolies, which enjoyed generous government support,1 including guaranteed access to rights-of-way. This history is relevant because the state of play in Canada’s telecom industry today is directly descended from the relative position of the players as they were in the early 1990s, following liberalization and at the beginning of the current era of competition. When liberalization was pursued, telco incumbents enjoyed the advantage of own-

84

Telecom Tensions

ing vast communications networks in an industry that was now open to new entrants. The first cablecos had developed in the 1950s through an uneasy dependence on the telco monopolies and leased access to telco infrastructure (Babe 1975; 1990). By the 1990s the cablecos owned infrastructure that had become largely separate from that of the telcos, with an effective monopoly on the last-mile of coaxial cable. Much of this infrastructure is still in place today, or has since been upgraded one piece at a time. The legacy of the monopoly era is inscribed onto the present through the placement of telecom conduits, utility poles, and the copper wires in our walls. This presents a material obstacle to the liberalization-era idea that separate networks should be built to compete with one another, since inheriting an infrastructure monopoly provides an enormous competitive advantage against anyone who might dare to compete. New companies wishing to establish themselves as isps in the 1990s therefore faced monumental challenges. The country was already wired by the telcos for universal phone service (which could be used to provide internet as dsl), and cablecos offered a superior alternative in urban areas. Without the ability to piggyback on this existing infrastructure, new entrants would have to find a way to build expensive new networks from scratch before they could hope to attract any customers of their own. This was a “barrier to entry” that new isp upstarts could not surpass, thereby limiting competition to incumbents (generally, one telco and one cableco in any urban area), or to incumbents and new entrants that depended on access to incumbent infrastructure. In other words, incumbents possessed “market power” in their control over the “bottlenecks” through which connectivity had to flow, giving them the ability to raise prices and limit competitors from entering the market (crtc 1994). If state agencies allowed this imbalance to go unchecked, the result would be “market failure.”2 All of this relates to a tension at the heart of any competitive activity. Participants must be formally equal at the outset, but unequal at the conclusion. Competitors are constrained by norms of equality, playing by the same rules, but in pursuit of inequality (Davies 2014). If the starting conditions are unequal, a referee can try to compensate by imposing advantages and disadvantages, but what happens when a competitor wins and is the last one left standing? Rather than beginning another race, in telecom this amounts to a return to monopoly. Hence, in the mid-1990s, public policy “remained focused on creating and maintaining corporations that would be able to survive not only each

Chasing Competition

85

other’s competition but also the rigour of the international market” (McDowell and Buchwald 1997, 713). Competition policy in Canadian telecom has maintained these protective dimensions ever since. In Canada’s telecom industry, intermediaries regularly call for a “level playing field” when seeking or opposing various forms of government intervention. Often, these are calls for different classes of players (such as incumbents and their competitors) to be subject to the same regulatory treatment, but “levelling the playing field” can also justify regulatory discrimination to compensate for existing or potential imbalances. If a player is perceived to have an advantage, then a level playing field can be achieved by imposing restrictions or requirements on that player, or compensating other players with special treatment.3 This regulatory logic can be expressed through some unlikely pairings of metaphors, such as when the crtc applied a “head start rule ... to ensure a level playing field” (crtc 2006).4 Regulatory interventions are intended to better balance the relative positions of parties at the outset of competition. Ideally, once this is achieved, any subsequent inequality or advantage can be attributed to a player’s superior competitiveness. But there has been no end to claims that the playing field of Canadian telecom is uneven, or that the game favours certain competitors, forcing regulators to constantly recalibrate the rules.5 Of particular interest to this chapter is the notion that a level playing field can be created by mandating access to incumbent infrastructure. During the mid-1980s, law professor Hudson Janisch and Bell economist Bohdan Romaniuk feared that, in the transition from monopolyera regulation towards competition, Canadian public policy would be trapped in an unending “transitional period” of “regulated competition.” They warned of three dangers that accompanied such a halfway form of deregulation: “a sense of procreative responsibility [would] incline regulators to keep new entrants in business ... an irresistible temptation [for regulators] to play the role of handicapper [would exist]; and, third ... opportunities [would be] created for strategic use of the regulatory process” (Romaniuk and Janisch 1986, 612). Among isps, the government’s “procreative responsibility” has not been sufficient to prevent numerous smaller companies from being swallowed by incumbents. However, a mandated wholesale access regime has been developed as a form of “handicapping” to compensate for incumbents’ historical advantages. This regime sustains the surviving new entrants and imposes obligations on the incumbents that deal

86

Telecom Tensions

with them. Finally, regulating relationships between incumbents and iisps has certainly led to “strategic use” of the regulatory process, wherein parties attempt to secure crtc decisions that lower their costs and raise those of their competitors. While the crtc is not the only federal agency that governs competition in Canadian telecom,6 the regulatory regime it has created for this purpose is the most extensive and elaborate. Relationships between intermediaries are subject to different categories of crtcimposed obligations, advantages, and disadvantages, depending on the parties involved. The public policy justifications for this regime have shifted over time, through different formulations of “facilitiesbased competition” and “telecom populism,” but the pursuit of a competitive market has remained central to its underlying logic. Competition between networks is prized for its potential to spur the building of more and better infrastructure, delivering lower prices and better choices for consumers. In other words, the public policy objectives are improving connectivity and serving the needs of consumers, with competition between intermediaries being held up as the means to achieve these goals. Unfortunately, competition in the telecom industry does not emerge out of nowhere; concentrations of power inherited from the monopoly era can easily come to dominate the field. As a consequence, regulators craft and revise rules about when intermediaries can and cannot compete, and under what terms. State agencies use public policies to calibrate and configure isps as instruments in the production of a competitive market, with the understanding that a competitive market will produce material benefits that are in the public interest. WHO COMPETES ?

Before turning to the question of how competition is structured, or the obligations imposed through regulation-for-competition, it is important to recognize that Canada’s liberalized telecom regime is not one in which everyone competes with everyone else. As discussed in the previous chapter, since the governing rationality is premised on the assumption that connectivity is best provided by competing private networks, public intermediaries take steps to insulate themselves from the argument that they are somehow competing with private industry. They do so by serving the needs of a delimited public domain such as public institutions, by filling gaps neglected by pri-

Chasing Competition

87

vate industry, or by leasing capacity to private industry and claiming this supports a more competitive market. Originally, internet access was limited to publicly funded research institutions, and to non-commercial uses as specified in acceptable use policies (aups). During the commercialization of the internet in the mid-1990s, growing demands for connectivity came from various other organizations, including private business (CA*net Institute 2001). As private provision of internet access crystallized and incumbents began offering access, the internet gradually moved further from its public-sector roots, and the rens of academia were left playing a limited role adjacent to the private market. For example, bcnet (a provincial ren) originally served a variety of public- and privatesector clients, before agreeing to restrict its scope and not compete with private isps. After dabbling in the marketplace of early internet access, by the 2000s rens had been limited to serving their traditional research and educational institutions, leaving the market for residential and business clients to private industry.7 While rens limited their competition with private isps by focusing on a distinct set of users, some governments employed isps as instruments of connectivity for a larger public. As discussed in the previous chapter, to avoid competing with private industry for the same clients, or to counter claims that they were unjustly favouring certain private partners, governments often favour some version of an openaccess approach that mandates access to infrastructure for competing intermediaries. Where the network remains in the hands of a private company under a public mandate, its operation is often subject to rules regarding competition. In the case of the bc government’s agreement with telus (telus and Government of bc 2011), the Alberta SuperNet contract (Government of Alberta, Bell Canada, and Axia SuperNet Ltd. 2005), and others (eorn n.d.; swift 2016), funding to extend rural connectivity is accompanied by the goal of promoting rural competition. As the price of benefiting from a public mandate, isps can be compelled to serve their competitors, or to avoid behaving as a competitor in a particular market. A role conflict develops when a company is expected to maximize the value of a network’s points of control against its competitors, while simultaneously being obliged not to compete. Private companies seek to increase profits and their competitive advantage but may be obliged to give other companies access to the most profitable customers. This role conflict is expressed

88

Telecom Tensions

through two contradictory views of how competition in telecom should operate: whether control over telecom “facilities” should be a source of competitive advantage, or whether there should be competition in “services” over a shared infrastructure. FORMS OF COMPETITION

While the policies discussed in this chapter are intended to promote competition or “competitiveness,” these terms have several distinct meanings that lead to very different governing rationales. For instance, competitiveness might refer to the number of competitors in a market, or the extent to which companies offering similar services compete. Competitiveness can also be interpreted as some form of efficiency, for which a monopoly or duopoly are arguably well-suited. And finally, there is “national competitiveness,” in which nations are seen as competing with one another for economic prosperity (Davies 2014). When it comes to governing the various components that constitute the internet, there is also the question of where competition is desirable. Is it in the public interest to have competing networks, with multiple sets of wires running above and under public roads, or would it be better to have multiple internet services available over a common physical infrastructure? The last of the above-mentioned distinctions is typically known as the debate between facilities-based or service(s)-based competition. Public policy can try to promote competition in either facilities (the material infrastructure of networks) or services (offered over a shared material infrastructure). Access to material infrastructure is key in both forms of competition, but for facilitiesbased competition this means being able to access where wires can be hung or buried, or towers built and spectrum used, while for servicebased competition access to interconnection facilities (such as the carrier hotel visited in chapter 1) is vital so that one network can make use of the reach offered by another. From the mid-1990s to 2019, the federal government officially promoted facilities-based competition, but also recognized the advantages of sharing infrastructure to promote a number of public policy objectives, including efficiency and national competitiveness. In specifying federal telecom policy, the crtc alternated between insisting on the long-term guiding principle of facilities-based competition, and recognizing that not all facilities can or should exist as multiple, overlapping sets. Rather than specifying what an ideal world of

Chasing Competition

89

facilities-based competition would look like, the crtc tended to look at the industry for indications of competition and market power, to find not enough of the former and too much of the latter, and prescribe mandated access to infrastructure as a solution. As described in chapter 2, telecom liberalization took place gradually in Canada, picking up steam in the 1980s, but with a great deal of uncertainty over the parameters of a newly competitive market. The Telecommunications Act of 1993 set out several broad objectives for telecom policy in Canada, as entrusted primarily to the crtc. These policy objectives included enhancing “the efficiency and competitiveness, at the national and international levels, of Canadian telecommunications” and “increased reliance on market forces for the provision of telecommunications services” (Government of Canada 1993, sec. 7). However, this said little about exactly what competitiveness entailed, how efficiency and competitiveness were to be determined, or where regulation was to give way to market forces. The importance of greater competition in telecom facilities, products, and services was affirmed through the mid-1990s in a number of government studies and policies, an “information highway” action plan, and crtc telecom decisions.8 In 1997, the crtc stated that “the full benefits of competition can only be realized with facilities-based competition.” In the Commission’s view, “efficient and effective competition” was best pursued through competing infrastructure – separate cables, wireless antennas, cabinets, and routers. Otherwise, competition might be cultivated at the retail level (through mandated wholesale access to incumbent networks), but incumbents would retain a wholesale monopoly. The supposed gains of competition – increased investment, innovation, and efficiency (with resulting price reductions) – needed to be realized at the level of facilities, where they would have the greatest impact (crtc 1997). As the rationality of facilities-based competition became regulatory common sense in Canada, other countries (particularly in Europe) were following a different course of liberalization that depended on mandated wholesale and leased access to telco infrastructure. This European approach (known as service-based competition) was dismissed by Canadian regulators, because unlike Europe, Canada had a well-developed cableco infrastructure to compete with telcos.9 Finally, facilities-based competition was preferable to mandated wholesale because mandated access requires more regulation. In a mandated

90

Telecom Tensions

wholesale regime, the prices charged for access to facilities need to be regulated, which places the regulator (the crtc) in the difficult role of determining incumbents’ costs of operation (Denton 2013). Mandated wholesale would keep the crtc engaged in setting rates, forestalling further deregulation and the transition to market forces. In Canada, the facilities-based orientation of competition was emphasized through crtc decisions into the 2000s and reached its high-water mark in the Minister of Industry’s 2006 policy direction (Government of Canada 2006). The policy direction was a remarkable and unprecedented statement intended “to remind [the crtc] that Canada is a capitalist country, a country of freedom, and that regulation must be as limited as possible, to allow market forces to play out” (Standing Committee on Industry, Science and Technology 2006). But even as the Conservative government committed the crtc to reduce regulation and govern through market forces, regulatory decisions were extending and solidifying a regime of mandated wholesale access through incumbent facilities. In other words, despite a commitment to foster the “construction of competing telecommunications network facilities” (Government of Canada 2006), regulators were extending obligations to share facilities. Much of the rest of this chapter is devoted to explaining this apparent contradiction. MANDATING WHOLESALE

Wholesale access to some telco facilities was mandated in the very same crtc decision that affirmed the importance of facilities-based competition (crtc 1997), and this obligation was subsequently extended “until such time as the market for such facilities is sufficiently competitive” (crtc 2001). While presented as a “transitional” approach (crtc 2002, sec. 26) that would allow new market entrants to climb the “ladder of investment” (or cross the “stepping stones”) to owning their own facilities (Cave 2006; Van Gorp and Middleton 2010), mandated wholesale access to incumbent networks became an enduring feature of Canadian telecom regulation. A series of “microregimes” (Levi-Faur 1999) were developed to mandate access and to set tariffed rates for specific networking infrastructures and technologies. Beginning with telco facilities in 1997, mandated access was extended to cablecos in 1999 (crtc 1999). Most recently, mandated access has been extended to last-mile fibre networks,10 is currently being debated for wires inside apartment and condominium build-

Chasing Competition

91

ings (crtc 2019c), and is likely to be introduced to the wireless market in the near future (crtc 2019a).11 Incumbents with massive investments in facilities frequently disagree with the mandated wholesale regime, arguing that the Canadian market is already highly competitive with some of the best networks in the world. According to incumbent voices, supporting smaller competitors thorough mandated wholesale is an “artificial” form of competition (Dobby 2014) that “distorts” the market (isp Summit 2013). A common criticism of the mandated wholesale access regime is that it gives new entrants little reason to build their own facilities, since they can depend on the use of incumbent networks (Beaudry 2010). After 2006, facilities-based competition was still cited as a rationale for crtc decisions, but it became a hypothetical long-term end state, to be achieved when the market for facilities was competitive enough. Deregulated facilities-based competition could only begin when the playing field was levelled and the advantages of incumbency neutralized. Incumbents’ responsibility to provide wholesale access was justified as an important levelling technique, despite the fact that the constant extension of this responsibility demonstrated just how ineffective it had been at erasing the advantages it was meant to address. While, in theory, Canada’s liberalized regulatory regime allows new players to enter the market and compete, in practice the “barriers to entry” have meant that any new entrant would have to invest billions of dollars to build facilities comparable to those of the incumbents. These high barriers to entry, resulting in inadequate levels of competition, are referred to by some economists as a form of market failure (Gómez-Barroso and Feijóo 2010), with incumbents maintaining considerable market power over new entrants (crtc 2015e). The mandated wholesale regime is meant to address abuses of market power, but it has created a fundamental role conflict for the incumbents it obliges to both compete and cooperate. iisps, meanwhile, are reliant on companies that often compete with them for the same customers, and because “reliance on a competitor is a source of conflict in the industry” (Competition Bureau 2019, 55), the crtc is left adjudicating the resulting disputes. COMPETITIVE DEPENDENCY

To summarize the developments detailed above: competition became a guiding principle of Canadian telecom regulation in the 1990s, sup-

92

Telecom Tensions

planting the previous arrangement of regionally based telco and cableco monopolies. Facilities-based competition was the governing rationality, and this meant encouraging the construction of competing telecom facilities. It was imagined that facilities-based competition could be achieved by compelling incumbents to make their infrastructure available for use by their competitors. This would allow new entrants to grow their businesses and invest in their own facilities, until they were no longer dependent on the incumbents. Incumbents would be forced to adopt a number of responsibilities in regard to their smaller competitors. They would have to coordinate customer ordering and installation, billing, troubleshooting, and the physical interconnection of equipment with these new entrants. The regulator (crtc) would set tariffed rates for access to incumbent networks, and review these relationships over time to see if they were still necessary. Once the consumer market was sufficiently competitive (a threshold that was never defined), regulated access to incumbent networks would be “forborne,” or lifted. The regime of wholesale access to incumbent infrastructure would be a transitory form of regulation until the imbalances inherited from the monopoly era had been corrected. But the legacies of the monopoly era would not be so easily amended. New entrants found the road to facilities-based competition either undesirable or impassable. Tariffed wholesale rates did not allow iisps much room to compete by offering lower prices than incumbents or to fund the infrastructure to successfully decouple from incumbent networks. iisps were more concerned with covering their expenses than raising billions of dollars to duplicate incumbent facilities. Many local and regional companies were satisfied with their limited footprint, and had no ambition to grow into carriers. As a result, the temporary wholesale regime that allowed competitors to climb onto incumbent networks became more durable, and regulatory conflicts gradually settled into contests over the details of these arrangements. These developments have resulted in a state I will describe as competitive dependency, in which dependencies exist between competitors, and wherein competitors are expected to cooperate. Competitive dependency exists where power relationships between competitors are highly unequal, and regulators extend obligations from the most powerful (the incumbents) to the least (the iisps). In Canadian telecom, this enduring arrangement came with a host of responsibilities for incumbents, who were compelled to provide support to their unwanted dependents, with the crtc acting as a referee.

Chasing Competition

93

Figure 4.1 Competitive Dependency Through mandated access, an iisp connects to an incumbent isp’s network as a wholesale customer. The iisp can use this connection to compete for the same retail subscribers as the incumbent, but the iisp depends on the incumbent’s last-mile infrastructure to do so.

iisps are sometimes characterized as being a burden for incumbents, as wholesale customers are less profitable than an incumbent’s own retail customers. However, the tension created through the obligation to serve a less-desired class of customers is not as fundamental as the role conflict underlying the mandated wholesale regime. This role conflict, which imposes contradictory expectations of behaviour on incumbents, results from the fact that iisps are both wholesale customers and retail competitors. As wholesale customers, iisps bring considerable revenue to incumbents and must be served according to crtc regulations. As competitors, iisps attract subscribers who might otherwise pay an incumbent even more as retail customers, and the regulatory regime presumes that incumbents and iisps will compete with one another for these customers. iisps are served by an incumbent’s wholesale division, which is dedicated to treating them as clients and addressing their needs. Wholesale account executives sell internet access to iisps and manage

94

Telecom Tensions

these relationships, sometimes competing with other wholesale access providers who serve the same area. Other employees maintain the interconnections between incumbent and iisp equipment, and troubleshoot technical problems experienced by the iisps. Some countries have pursued “functional separation” of incumbents’ wholesale and retail divisions to promote competition (oecd 2012), but in Canada no such requirement exists (Middleton 2011).12 Therefore, incumbent wholesale divisions vary in how independent, resourced, and driven they are to compete for new business. Conflict arises when the goals of wholesale and retail work against one another, or when wholesale becomes subordinate to retail, since a growing number of well-served wholesale customers is likely to result in fewer retail customers. In the end (at the literal end of the network, or the bottom of Figure 4.1), retail and wholesale serve the same customers. However, retail serves these end-consumers more directly and profitably, while wholesale does so through another intermediary that can be seen as a retail competitor. Because retail tends to be more profitable than wholesale, and iisps depend on incumbents for wholesale, there is always the danger that this wholesale dependency will be used against iisps by incumbents. First, incumbents can simply choose to neglect their wholesale business, providing service but only doing the bare minimum to meet their regulatory obligations. Secondly, incumbents can actively exercise market power against competitors, exploiting their control over “upstream” or “bottleneck” facilities to control competitors’ rates (crtc 2015d), service (crtc 2010a), or behaviour (crtc 2014a). Thirdly, incumbents provide the information used as a foundation for crtc decision-making, namely through cost studies. These document incumbent costs for providing services, are submitted in confidence, and are used to set “just and reasonable” wholesale rates. While the crtc can disagree with incumbents’ cost studies, it is up to iisps to challenge any rates approved by the Commission that may be unfair or result from “costing error” by incumbents.13 The crtc, as the delegated sovereign for Canada’s telecom networks, adjudicates a never-ending struggle over access to incumbent facilities. The Commission sets tariffed costs, monitors the “quality of service” that incumbents provide competitors (crtc 2018), determines when an incumbent’s behaviour counts as undue discrimination, and settles various disputes. Regulatory inequalities between incumbents and iisps are put on public display at crtc hearings,

Chasing Competition

95

where incumbents sometimes appear before the Commission with large panels of lawyers, executives, and experts, while iisps more typically pool their resources for regulatory affairs. In a regulatory process that is meant to protect them from the worst effects of dependency, iisps are at a disadvantage in terms of resources and participation (McNally et al. 2018). As one example of how competitive dependency operates in practice, iisps often rely on incumbent last-mile infrastructure, making the incumbent responsible for sending a technician to service an iisp subscriber’s home. In 2013, iisps (represented through cnoc) alleged that cablecos were unduly discriminating against them by making wholesale access difficult to obtain and providing inferior service to retail customers (cnoc 2013). TekSavvy’s subscribers experienced long disruptions of service that the iisp was unable to resolve on its own, and one group of customers reported that their problems were quickly corrected once they switched to become customers of the incumbent (CBC News 2013). Ultimately, the crtc ruled that iisps failed to demonstrate that they were being discriminated against, because they lacked a way to contrast their treatment with the treatment of the cablecos’ retail customers (crtc 2015b, para. 17). This highlights the informational asymmetry between iisps and incumbents, since only the cablecos are in a position to provide such a comparison to the crtc.14 The Commission ordered both groups to resolve their issues through a special working group, but this is just one example of the many disputes that have arisen between the two classes of intermediaries, due to the expectation for incumbents to treat iisps as both competitors and customers. While the wholesale access regime does impose some clearly defined obligations on incumbents, it does not constrain incumbentiisp relationships into a rigid mold. Incumbents maintain a great deal of discretion in how they approach their wholesale relationships and the service provider/competitor role conflict. At industry conferences that cater to iisps, some incumbents act as sponsors and try to attract wholesale customers, while others are remarkable in their absence. Relationships with incumbent wholesale providers can also change over time, particularly in the last decade as a number of incumbents have come to terms with the permanence of the wholesale regime and the sustainability of iisps. In some markets competition for wholesale business is so lucrative that these relationships cannot simply be ignored in favour of retail customers.15 But even when an incum-

96

Telecom Tensions

bent’s wholesale division offers an iisp a deal in order to win their business, the regulatory regime remains in place as a “backstop.” This is to say that without a mandated set of obligations, iisps’ ability to negotiate for favourable terms would be greatly reduced. In summary, competitive dependency in Canadian telecom arises from highly unequal power relationships, and the disciplining of incumbent behaviour by the mandated wholesale regime. The resulting condition is one in which iisps are protected from total dominance by incumbents, and which obliges incumbents to serve iisps. To promote competitiveness and to keep our connectivity from being disrupted by corporate squabbles, the regime mandates cooperation between rival isps. Role conflict regularly occurs as incumbents are expected to both compete and cooperate with iisps. The following section will examine why this regime has endured despite these tensions. ACCOUNTING FOR THE MANDATED ACCESS REGIME

Since the efforts of the crtc and successive federal governments to foster telecom competition have stretched on without end, it seems obvious that the telecom market has failed to develop in a way that meets public policy expectations. This raises the question of what has sustained a regulatory regime that has been so inadequate at achieving its stated objectives. As previously mentioned, incumbents have argued that vigorous competition can (and does) exist between a small number of giant firms, but this argument has not persuaded Canadian governments to abandon regulation-for-competition. The continuation of mandated wholesale has become a way of forestalling a broader form of market failure that would lead to the collapse of smaller competitors and further industry consolidation. Several factors account for the extension of mandated wholesale. First, mandated wholesale continues because incumbents still have market power, which they would presumably exploit if incumbent wholesale obligations were removed. The fact that the regime has been ineffective in eliminating this market power has not been taken as a reason to abandon the regime, nor as a reason to pursue more comprehensive forms of mandated wholesale, and so the existing approach carries on as a sort of middle road. Second, increasing consumer choice has been a way for the federal government to cater to populist sentiment, and since 2011 mandated wholesale has been seen

Chasing Competition

97

as an issue that can attract significant public interest. Finally and most recently, the objectives of mandated wholesale are no longer as stated in the 1990s. While facilities-based competition was officially sidelined in 2019, it had long since ceased to be the immediate goal. The Middle Course and the Triumph of the Giant Firm Incumbent dominance after liberalization can be partly explained by specific features of the telecom industry, such as high barriers to entry for facilities-based providers, but it is also consistent with outcomes in other industries where neoliberal rationality has supposedly prevailed. As Colin Crouch (2011) argues, neoliberal policies such as privatization and liberalization do not tend to create the competitive markets that justify them, but instead favour dominance by giant firms (often the same corporate actors supporting neoliberal policies). This forces us to consider whether “the virtues of the market [are] better expressed in the maintenance of competition, and therefore with the existence of large numbers of competing firms, as in pure economic theory, or in the outcome of competition, which may often mean the survival of a few giant corporations and diminished consumer choice” (Crouch 2011, 53). In other words, should a market-based approach support a large number of competitors, or should we accept the continued dominance of Canada’s telecom industry by incumbents as the predictable result of competition? Successive governments’ support for regulatory capitalism through mandated wholesale indicates a desire for more competitors and an unwillingness to cede the field to incumbent survivors. But this has also been the limit of regulatory intervention, since the idea of a transition to service-based competition contradicts the facilitiesbased rationality. Mandated wholesale is a path between a pure form of facilities-based competition and a full commitment to shared facilities, because either turn would have consequences that are deemed unacceptable. To turn away from a mandated wholesale regime and toward closed, competing facilities would mean openly accepting the incumbentdominated status quo that competition was meant to unsettle. Deregulation of wholesale could not be pursued if it would result in the downfall of iisps and the survival of the incumbents. While turning away from mandated wholesale would be disastrous for iisps, extending mandated wholesale to its logical limit would force a radical trans-

98

Telecom Tensions

formation on incumbents. This would mean abandoning facilitiesbased competition in favour of service-based competition over shared facilities (“structurally separated” wholesale networks), as is the case in a number of national regimes, primarily in Europe (see oecd 2012). While calls for structural separation in Canada have recurrently been made by some commentators and non-incumbent intermediaries, the dominant view has been that such a policy option is “off the table” (cpac 2010) or “not up for discussion” (Palmer 2015). Structural separation has typically been dismissed as unnecessary and dangerous (crtc 1994; Denton 2013), requiring a more radical disruption of the status quo than Canadian governments have been willing to consider.16 If regulatory capitalism designates a mode of governance in which the state focuses on “steering” (directing, guiding) while business concentrates on “rowing” (providing services and entrepreneurship; see Levi-Faur 2006, 505), then Canadian governments have kept the ship on a fixed course for fear of the alternatives. Avoiding the poles of facilities-based and service-based competition means the continuation of liberalization along existing lines, reinforcing the status quo. Mandated wholesale continues precisely because it has not been successful in reordering the playing field, and any different course would have consequences that regulators have been unwilling to accept. Telecom Populism Another factor leading to the federal government’s continued support for the mandated wholesale regime is a contemporary form of telecom populism, which has worked against the interests of incumbents. This has resulted in the coupling of government decisions on intermediary responsibility with the public interest in a way that has not been seen since the battles between Bell and the independents in the early 1900s. What distinguishes this from the previous era of telecom populism is the promotion of consumer choice as the objective of competition (cpac 2013a). While policies of marketization (the creation of markets) have a number of other justifications, such as efficiency and international competitiveness, a particularly important reason given in the context of telecom liberalization has been consumer choice. Choice for consumers has been presented as an objective of Canadian telecom regulation since the liberalization of long-distance service, but was explic-

Chasing Competition

99

itly championed under the Harper government’s brand of telecom populism between 2011 and 2015. Under this rationale, the greater the number of competitors, the more consumer choice, the better. This makes it hard to accept any policy that would lead to consolidation in the industry, even if such an outcome might reward the success of the most effective competitors. It was the public’s unanticipated interest in the relationship between iisps and incumbents in 2011 that convinced the federal government that there was public interest in the continuation of a mandated wholesale regime favourable to iisps. At the centre of the dispute that galvanized public opinion was a crtc decision about the costs charged by incumbents for wholesale access (crtc 2011a). However, this conflict (over usage-based billing, or ubb) quickly became a focal point for concerns over “metered” internet, incumbent dominance of the telecom industry, and the role of the crtc (Anderson 2011b).17 In an unprecedented turn of events, a billing dispute between Bell and iisps became a major political story, and a petition organized by OpenMedia to “demand access to an unmetered Internet” (OpenMedia 2011) was signed by half a million Canadians. In response, the Prime Minister and Industry Minister both expressed their displeasure with the crtc’s original ubb decision.18 This led the crtc to reconsider, thereby demonstrating the limits of the Commission’s delegated sovereign authority, which ultimately depends on the support of the federal Cabinet. For many members of the public who responded to OpenMedia’s (2011) Stop the Meter campaign, opposing the ubb decision seems to have meant opposing the commodification of bandwidth. For the Commission, the ubb saga demonstrated growing public interest in internet policy and the involvement of new forces (such as OpenMedia and iisp supporters) in hitherto obscure regulatory disputes. For the Conservative federal government, the issue would become a turning point, leading to more active government involvement in telecom matters and attempts to align policy with public sentiment.19 Subsequent to its reconsideration of ubb, the federal government would commit to putting “consumers first” (Government of Canada 2013) in the battle for greater telecom competition, with competition defined as greater choice for consumers. The current Liberal government has continued to emphasize the importance of consumer interests, most recently at the expense of facilities-based competition (Government of Canada 2020).

100

Telecom Tensions

Shifting the Goalposts of Competition Policy In February 2019, the crtc’s arm’s-length relationship with the federal Cabinet was once again feeling the strain of divergent ideas over competition policy. After sending two displeasing crtc decisions back for review,20 ised Minister Bains pulled the crtc’s arm and gave the delegated sovereign a course correction. He did this by issuing a policy direction – the first since the facilities-based guidance of 2006 – and in doing so opened up the possibility of redefining competition policy. The crtc had come to effectively ignore the 2006 policy direction, apart from the obligation to mention it in telecom decisions. It had applied further regulation to incumbents and accorded more legitimacy to leased access (Denton 2013). The Commission extended mandated access to last-mile fibre networks in 2015, and claimed that forcing incumbents to share their facilities would promote facilitiesbased competition. This competition would just have to happen elsewhere in the network, because it was neither “practical or feasible for competitors to duplicate” the incumbents’ wires over the last mile (crtc 2015e, secs. 134–6). The fact that the Commission was required to tie arguments into knots to explain itself, or could redefine the scope of facilities-based competition to justify shared facilities, was not particularly bothersome. The only body that could hold it accountable generally approved of these moves. Once the crtc (under the direction of its new Chairman, Ian Scott) began to behave in ways that made Cabinet uncomfortable, revisiting the policy direction became a way to instruct the Commission on how it was expected to behave. The 2019 policy direction was more open-ended than the directive of 2006, signalling that the crtc could intervene to promote “all forms of competition” (Government of Canada 2020). In effect, Cabinet was telling the regulator that it was not regulating enough, and that competition was still valuable even if it was not between competing facilities. Certainly, the crtc could try to flout the policy direction or interpret it narrowly as it had previously done, but this would escalate the conflict with Cabinet. When sovereignty is contested between the two, Cabinet will eventually have its way. In 2019, the rationality of facilities-based competition was officially demoted. The move had the effect of transforming an inconvenient policy failure into a horizon of possibilities. The vague fantasy that competition would lead to overlapping wires running into every home was

Chasing Competition

101

abandoned. In its place, consumer interests and competition were reaffirmed as core values, but competition could be defined differently than before. Facilities-based competition was still an option, and one evidently still favoured by the crtc even as it course-corrected after receiving its policy direction.21 But facilities-based competition was not the only officially recognized option for the state to promote. Competitive dependency would be part of the picture for the foreseeable future. Perhaps, success in competition policy could mean something closer to the status quo, and the middle road no longer needed to be justified as a temporary detour. Mandated access could now be the final destination. THE LIMITS OF COMPETITION

This chapter has traced the development of regulation-for-competition in Canadian telecom and the mandated access regime. Consistent with the larger phenomenon of regulatory capitalism, Canadian telecom is governed through norms and regulations that specify how intermediaries should compete with one another, and where they should avoid competing. Incumbent ownership of material points-ofcontrol (which could be fibre backbones, wireless towers, or last-mile wires) confers power to these intermediaries. As with intermediation more generally, this power is met by additional responsibilities, regulating access and industry cooperation. The mandated access regime has failed in its original objective to foster an arrangement of intermediaries that can be sustained by facilities-based competition, as evidenced by the continuation and expansion of mandated access (see crtc 2015e, secs. 121–4; 2019a, secs. 34–40). The reasons for the regime’s failure can be linked to flawed assumptions about how incumbents and new entrants would behave following liberalization, but I have been more interested in accounting for the regime’s continuation than its failure. The reasons why mandated access continues to endure as a central point of struggle over telecom policy includes the crtc’s and federal government’s unwillingness to accept either of the two main alternatives to the regime: allowing complete incumbent dominance (by eliminating mandated access) or removing incumbent control over both facilities and services (through structural separation). Dismantling the mandated access regime would contradict the government’s longstanding emphasis on maximizing consumer choice – a cause that was taken up

102

Telecom Tensions

fervently in the later years of the Harper government as a kind of telecom populism that opposed incumbent interests. Finally, it is worth remembering that while promoting competition remains the governing rationality for telecom regulation, what this means has changed over time. Rather than some imminent end-state where regulators can depart from the stage, the current direction of competition policy points to a world where some classes of facilities are shared while others compete, and regulators draw the boundaries of acceptable competition. Facilitiesbased competition is no longer an overriding policy orientation, but the general value of competition remains dominant. Network-building can be justified for reasons other than enabling private-sector competition, but the competitive ideal is never disputed. The problem is never too much competition, or too great a reliance on markets. If there is a problem, it is either that there is not enough competition or that it is the wrong kind (facilities- or service-based). Since intervention in the market is justified as regulation-for-competition, and because the ultimate vision of a competitive market desired by these public policies is never clearly laid out, the resulting arguments can be dizzying in their contradictions. Where potential competitors are forced to cooperate by sharing infrastructure, this is justified as a way to promote competition. Competition requires a level playing field, which is promoted through differential treatment – imposing responsibilities on the most powerful and sustaining intermediaries that would otherwise be unable to sustain themselves. Positive visions of the world that competition policy is meant to create have been vague, and by avoiding interpretive constraints (such as the ideal outcome of facilities-based competition), the crtc can maintain flexibility in how it rationalizes individual decisions. But even without an alternative vision of how to organize telecom infrastructure in Canada, regulators can make their decisions based on lines that should not be crossed, or conduct that should be prohibited. In this regard, preventing the giant firm’s ultimate triumph remains an important public policy objective. However, limiting corporate concentration is a shallow political objective compared to pre1990s notions of national digital infrastructure, or the principled pursuit of competition that was espoused at the outset of liberalization. And so, public policy that was once meant to shape a new arrangement of market actors is left maintaining the existing order – an order

Chasing Competition

103

that preserves incumbent dominance as well as the opposing tension brought by iisps. Ultimately, the debates in this chapter over competition policy are secondary to questions of power. This is often openly acknowledged, since notions of competition and market power are deeply interrelated in telecom policy. The crtc decisions underpinning the mandated wholesale regime have become a way of restraining incumbent power, rather than a means of promoting industry competition. Under the logic of liberalization, one is presumed to lead to the other, but this is not necessarily the case. Incumbent power is controlled through the mandated access regime, just as it is through spectrum auctions and merger approvals, but the result has not been a flowering of competition. Instead, liberalization has led to a concentration of power in the giant firm(s), and regulations have restricted this consolidation. While this chapter has focused on mandated access, the principles of common carriage and net neutrality have also provided important and complementary rationalities for governing the problem of market power. Where mandated wholesale addresses the problem of incumbent market power through control over bottleneck facilities, net neutrality is a response to isps’ ability to discriminate among the data flows passing through these bottlenecks, potentially privileging data flows that enhance their market power and discriminating against competing data flows. Mandated wholesale does impose requirements on how data flows between incumbents and their wholesale customers, but fundamentally what is at stake is physical access to networks. Net neutrality, on the other hand, focuses squarely on how data is governed by isps, and the consequences of discriminatory treatment for competition, information flows, and user autonomy. Net neutrality has implications for topics discussed in subsequent chapters, including copyright and cyber security, while also being an important dimension of regulation-forcompetition in Canada. The next chapter takes up this topic, which spans concerns over connectivity, competition policy, and the roles that isps play in governing traffic across their networks.

104

Telecom Tensions

5

Common Carriage, Net Neutrality, and Bad Traffic

INTRODUCTION

Many are now aware that the internet, in addition to being vital for our society, is also used to spread harmful content. The harms that have been top of mind in recent years include political interference and disinformation, radicalization, criminality, and abuse, all facilitated by the openness of internet infrastructure. Attempts to govern this conduct often focus on specific communications platforms, such as Facebook or Google, but isps provide the underlying infrastructure and can similarly be enlisted to identify and police misconduct. A key regulatory difference between the two is that unlike platforms, Canadian isps have inherited the responsibilities of “mere conduits” and “common carriage” from an earlier era, and are now subject to network neutrality requirements. The resulting role conflict is between the expectation that isps actively intervene in traffic and the expectation that they remain neutral carriers. More than just a battle over the acceptable scope of an intermediary’s conduct, or a disagreement over the internet’s foundational principles, the fight over net neutrality is a struggle over market power and its concentration in the telecom industry. Net neutrality is typically presented as the principle that all internet traffic should be treated the same. In practice, net neutrality’s many exceptions define the kinds of traffic that can be discriminated against, and the ways that isps can profitably exploit their positions as points of control. This may be a conflict about some of the biggest questions of internet policy, but Canadian net neutrality policies have been defined piece by piece, in relation to some very specific cases of bad behaviour.

Common Carriage, Net Neutrality, and Bad Traffic

105

Most recently, the struggle has been over piracy, when at the end of November 2019 some Canadian internet subscribers found their use of the GoldTV streaming service blocked, replaced by a notice from their isp informing them of a recent court decision against the website (Geist 2019). Unable to target the operators of GoldTV (the alleged pirates), who could only be named as “the John Doe defendants” due to their efforts to remain anonymous, copyright owners had instead targeted the large isps that most Canadians relied upon to connect to the site. The Federal Court, rejecting an argument based on net neutrality, ruled in the copyright owners’ favour, ordering isps to act as agents of copyright enforcement by establishing a websiteblocking process for the first time under Canadian law.1 Not long thereafter, subscribers attempting to connect to GoldTV.biz were redirected to a page notifying them that the streaming service had been rendered inaccessible. The intermediaries’ technical infrastructure had been repurposed for the fight against piracy; what users counted on to connect them to desired content had been transformed into a regime of containment for intellectual property rights. Because copyright owners knew that their targets were slippery, and could route internet traffic around a static court decision, they included a way to have isps to extend their blockade when GoldTV and its customers inevitably tried to circumnavigate it. Bell’s investigators covertly became GoldTV customers, tracing the service’s shifting domains and reporting these to other isps, whom the court had turned into responsive instruments of copyright enforcement (LeeMurphy 2020). Although this micro-regime was focused solely on GoldTV as a moving target, it represents a historic victory for copyright owners’ effort to steer intermediation in a direction that serves their interests, unobstructed by net neutrality and common carriage. As this chapter emphasizes, regulating net neutrality does not simply mean prohibiting discrimination – it is also a way of permitting discrimination, particularly against “bad” content and traffic. NET NEUTRALIT Y AND COPYRIGHT

Normally, isps are not in the business of policing copyright on their networks; internet subscribers expect to connect to whatever services and content they want, and while a subscriber’s contract states that they agree not to use their network connection to break the law, the isp is generally not responsible for assessing the intellectual property

106

Telecom Tensions

rights involved in a user’s data flows. Against this neutral stance, the argument that isps should take on a copyright enforcement role has driven one of the oldest and hardest-fought disagreements in internet policy. Copyright owners have tried to hold Canadian isps responsible for copyright enforcement since the mid-1990s (Copyright Board of Canada 1999). In 2012, internet and telecom companies were important political actors in successfully opposing website blocking under sopa in the US and an expanded monitoring role under acta in Europe (Horten 2013). So why did major Canadian isps either successfully champion website blocking against GoldTV in 2019, or allow it to proceed without opposition? Because isps play many roles, connectivity is often just one of the services or lines of business in which Canadian incumbents have an interest. “Vertically integrated” companies such as Bell and the major cablecos encompass broadcasting businesses as well as telecom services, producing and distributing profitable content through some subsidiaries, and operating general-purpose communications infrastructure through others. The specific role conflict this establishes is between the expectation that an isp provide the connectivity that users want, and the expectation that the isp serve the interests of its parent company, which might mean favouring some kinds of connectivity over others. More generally, the role conflict is between neutrality and discrimination, when discrimination is profitable for an isp that would prefer its subscribers access or avoid particular content. As seen in this and the subsequent chapter, isps can choose how to legally position themselves in copyright cases, and an important factor in this decision is an isp’s relation to media ownership. Companies such as Rogers Communications, bce Inc. (Bell), and Quebecor (Vidéotron) own television channels and other media subsidiaries, producing and licensing video for broadcast and streaming services. Internet subscribers can choose to watch these videos, or they can choose to watch videos from a competing service. In some cases, this competing service operates illegally, providing content that the isp’s parent company is invested in, but without permission or payment. For example, Bell profits when internet users pay to subscribe to the Bell-owned Crave streaming service, but internet users can watch the same content (including content produced by Bell Media) through “pirate” services like GoldTV. Hence, the fight over access to GoldTV is part of a larger policy struggle, in which vertically integrated media companies have

Common Carriage, Net Neutrality, and Bad Traffic

107

tried to create new regulatory regimes that would authorize and compel website blocking by isps (Fair Play Canada 2018). It is not surprising that companies with substantial media interests have been championing a more active role by regulators, courts, and isps in governing copyright online, while isps that are less affiliated with media have taken more neutral or opposing stances (House of Commons Standing Committee on Industry, Science and Technology 2018). Alignments of interests have important consequences for how copyright is enforced in Canada, including the liability and privacy of internet users. Those topics will be explored further in the following chapter, but the focus here is on the actions that intermediaries take to shape how users experience the internet. Specifically, this chapter addresses how and why isps discriminate between data flows, including blocking specific websites, “throttling” some kinds of traffic, or charging subscribers different rates for different uses of the internet. Today, these concerns are part of the policy debate over net neutrality, but in the telephone and telegraph eras, a similar role conflict led to the creation of common carriage responsibilities. Net neutrality is not just a rehashing of these earlier struggles, but has been built on the foundations of common carriage (Sandvig 2007) as well as the intermediate discourse of “open access” as a justification for mandated wholesale (Newman 2019). In Canada, as in the US, net neutrality became a way to challenge incumbent giants and their self-serving practices. The discourse of net neutrality allows smaller firms, public advocates, and activist groups to push back against undesirable forms of discrimination. In the GoldTV case, this pushback was initially unsuccessful, although net neutrality arguments are still being made in 2020 as part of TekSavvy’s ongoing appeal of the decision (Malik 2019). More broadly, net neutrality has become a prominent part of telecom regulation in Canada, thanks in large part to the work of nonincumbent actors including iisps, advocacy groups, and academics. However, what allowed these voices to succeed against Canada’s largest telecom companies was the fact that net neutrality was far from a fringe idea or radical set of responsibilities. Instead, as Newman (2019) argues, the discourse of net neutrality was aligned in fundamental ways with dominant assumptions about the commercialized internet, justifications for telecom liberalization, and even earlier conceptualizations of common carriers, conduits, and interconnection.

108

Telecom Tensions

COMMON CARRIAGE

In a widely interconnected world, where transportation and communication infrastructure is privately owned, control over such infrastructure can benefit owners in ways that go beyond the profits earned from subscribers. Governing communications can be especially lucrative when infrastructure owners also have other business interests, given the importance of connectivity for so many contemporary processes. The problem addressed by both common carriage and net neutrality regulations is the possibility that infrastructure owners will favour some uses and users of their network and discriminate against others because of these additional business interests, and that a “common” or “neutral” infrastructure must be maintained to serve all fairly. Precursors to common carriage regulations governed English transportation networks in the 1600s, before crossing into the North American context in the 1800s to deal with railroads and telegraphy (Noam 1994). Originally, in English common law, those who offered their services to the public were considered to be engaged in a “common calling” and were bound by a legal responsibility “to serve all who should apply” (Burdick 1911, 522). This expectation of nondiscrimination could be extended to the “common” water carrier, baker, farrier, tailor, surgeon, ferryman, or bargeman. By the 1800s, offering services to all members of the public was no longer an exceptional practice, and so this special obligation to serve was applied more narrowly to innkeepers and the “common carriers” who operated transportation and communication networks (Noam 1994). Liability for common carriage developed as a safeguard to maintain reliable flows of people, cargo, and communications. As dependence on these infrastructures grew, common carriage regulations were intended to keep their operations dependable. The key historical example in Canadian common carriage is the dependence of the news industry on telecommunications. If news media do not control their own distribution networks, they must rely on another party to ensure that the news reaches an audience. In the internet era, fierce battles have been fought between news providers and online gatekeepers, but a hundred years ago the “wire” news services relied on telegraph companies, and telegraph companies (which were often affiliated with railways) were also in the business of news-

Common Carriage, Net Neutrality, and Bad Traffic

109

gathering and distribution. Conflicts became inevitable: in 1907 Canadian Pacific raised the rates to carry its competitors in the news business, and cut off news service to a paper in Nelson, bc for publishing negative stories about Canadian Pacific. In response, the brc imposed rate regulation on telegraph companies in 1910, Canadian Pacific got out of the news business, and Canadian telecom companies became common carriers (Babe 1990; Rens 2001). In today’s world of vertically integrated media and telecom conglomerates, the content of the news may once again conflict with a parent company’s other business interests. According to internal correspondence published by Winseck (2015b), Bell executives repeatedly tried to shape favourable coverage of regulatory issues through Bellowned news media. This sort of behaviour ultimately cost the president of Bell Media his job in 2015, after he intervened in news coverage of a crtc dispute (Bradshaw and Dobby 2015). Newsroom manipulation to promote a corporation’s agenda may be unethical and the product of concentrated ownership, but common carriage principles have a more specific application, as enshrined in the Telecommunications Act’s sections 27(2), 28, and 36. These apply specifically to carriers, which “shall not control the content or influence the meaning or purpose of telecommunications,” nor discriminate unjustly in the transmission of programming, the pricing of telecom service, or its provision. While a media company can shape its content in selfserving ways, it would be illegal for an isp to block a news site that criticized the company, to modify the content of news carried over the network, or to demand additional payment for carriage, simply to serve a business interest. For a time during the mid-2000s, these common carrier provisions were deemed sufficient by the crtc to address net neutrality concerns (French 2007), but much depended on the kinds of discrimination considered “unjust,” or what the regulator understood to mean “control” and “influence” over communications. As the net neutrality debate developed and the crtc began deciding issues more specific to discriminating between internet traffic, a new regulatory microregime was required to elaborate common carrier principles for the internet era. Rather than limiting intermediaries to mere conduits, these initial net neutrality regulations had to distinguish between acceptable and unacceptable kinds of discrimination, and to make isps accountable for their interventions in traffic flows.

110

Telecom Tensions

SHIFTING INTO NEUTRALIT Y

In the mid-2000s, the net neutrality debate spread beyond US policy discussions and became crucial to internet governance questions in different jurisdictions, including Canada. The origins of the discourse championing net neutrality are complex, preceded by US debates about regulating open access in the late 1990s, which were themselves partially informed by the development of mandated wholesale in Canada (Newman 2019). Tim Wu, an American legal scholar who had spent his early years in Canada, coined the term “network neutrality” to provide a policy alternative to “structural solutions” such as open access, structural separation, and mandated wholesale (Wu 2003). Wu’s original formulation of net neutrality as a principle held that vertical integration could be acceptable (even valuable), as long as isps were prevented from practising undesirable forms of discrimination. As debates on the topic continued in the US, the crtc implemented some of the first net neutrality regulations in the world, but Canadian isps also provided some early examples of what a violation of net neutrality might look like. One story in particular has gained international attention as a cautionary tale of online discrimination. In 2005, members of the Telecommunications Workers Union were engaged in a heated labour dispute with telus, their employer. A website operated by a union member included forum posts that suggested jamming telus phone lines and displayed photos of employees crossing picket lines. telus blocked access to the website from its network by targeting its ip address, which also disrupted access to hundreds of other websites hosted on the same server. The company claimed that content on the targeted site was a threat to its employees and its operations (CBC News 2005). telus’s action may have done the company more harm than good because of the unfavourable attention it generated in various media, but on its face it appeared to show an intermediary using its control over digital flows as leverage in a power struggle. Ultimately, the blockage was short-lived; within days telus reached an agreement with the website operator and restored access. However, it became easy to imagine a similar situation where an isp decided to limit connectivity to its critics, and the case raised questions about the ability of intermediaries to make these choices unilaterally (Barrett 2005). Even where an isp removed access to some kind of illegal content, the case demonstrated how “an entity whose own interests are at

Common Carriage, Net Neutrality, and Bad Traffic

111

stake may have an incentive to interpret more liberally what should be removed” (van Schewick 2010, 268). The legality of the company’s actions was never tested against its common carrier responsibilities, and telus’s website blocking may well have been unlawful even in the absence of net neutrality regulations (French 2007). However, the incident has been cited ever since by net neutrality advocates and defenders of open media, in both the US (Klinenberg 2007, 291; Newman 2019, 334) and Canada (Anderson 2011a). That an isp could cut off connectivity the way that telus did in 2005 seemed like censorship – a violation of the ideals and expectations of an open internet (Austen 2005; Barrett 2005). However, the sorts of problems that would lead the crtc to establish net neutrality regulations involved more flexible kinds of traffic discrimination as well as the rates charged by incumbents. Net neutrality would factor into debates over mandated wholesale, copyright enforcement, vertical integration, cyber security, and the “internet of things.” Concerns about net neutrality proliferated as a way to critique the process of intermediation as it unfolded along its many dimensions, wherever intermediaries became newly active in governing traffic. Intermediation and Net Neutrality Net neutrality is now presented as a key principle in Canadian internet policy, although the crtc only fully embraced the term after a round of public hearings in 2017. Subsequently, the Commission released a new policy framework that defined net neutrality as the principle “that all traffic on the Internet should be given equal treatment by isps. In other words, there should be no manipulation, preference, or discrimination, either through technical or economic means” (crtc 2017b). The very same day, the crtc also published a webpage explaining its view of net neutrality to the Canadian public, where the definition was similar to the above, with one notable exception: the “equal treatment” afforded traffic was less absolute. Instead, net neutrality meant “little to no manipulation, interference, prioritization, discrimination or preference” (crtc 2017a). A “little” manipulation and discrimination was still permitted, and net neutrality policy is a way of determining what falls under this “little” exception. Regulatory decisions concerning net neutrality involve answering specific questions such as whether an intermediary can offer unlimited streaming to some platforms but not others, or whether it is acceptable

112

Telecom Tensions

to limit bandwidth when a network is experiencing high levels of use. For the crtc, the lofty stakes of these decisions are to ensure “that all Canadians have access to choice, innovation and free exchange of ideas” (crtc 2017a). For intermediaries, net neutrality policies translate to limits imposed on action – limitations on isps’ ability to profit from some traffic, to exclude or block content, or to manage data flows in ways that are optimal for their networks. Net neutrality is invoked across so many debates in internet policy because the discourse allows people to engage with fundamental questions about what an intermediary should be, and to envision an ideal internet. However, net neutrality conflicts have also followed as a consequence of intermediation: the further an isp moves from the ideal of a passive, neutral intermediary and towards exercising agency over data flows, the more an isp discriminates. Net neutrality becomes a way of interrogating whether this discrimination is legitimate. A lot of the positive potential of the internet can be attributed to its openness; the end-to-end principle, mere conduits, and decentralized governance allowed for the development of innovative uses (van Schewick 2010). Net neutrality discourse emerged from debates over open access (Newman 2019), and net neutrality and “open internet” are often synonymous (fcc 2015). The discourse of net neutrality is rooted in the idea that intermediation is a problem for openness, and that isps, though their expanding interests, are in a position to govern connectivity in a way that forecloses on possibilities. At its most expansive, net neutrality is an “empty signifier” (Newman 2019, 32) that can be used to oppose nearly any kind of intermediation, or to push back any attempts by isps to expand their agency over data flows. While net neutrality advocates have often grounded their arguments in the neutrality of the early internet’s design and its end-to-end principle (Lawford, Lo, and De Santis 2009; van Schewick 2010; Wu 2003), critics and some network operators have argued that the “net was never neutral” (Newman 2019, 62) and pointed to desirable and regularly practised forms of traffic discrimination (French 2007; McTaggart 2006). However, there has been recognition on both sides that this is fundamentally a normative debate about ideals (McTaggart 2006; Wu 2003), and how isps should conduct themselves. In simplified terms, net neutrality advocates are opposed to isps discriminating among internet traffic. Net neutrality advocates sometimes warn that discrimination by isps will produce “walled gardens”

Common Carriage, Net Neutrality, and Bad Traffic

113

(Crawford 2013), hearkening back to the days of CompuServe and aol, but the more immediate concern for regulators is that intermediaries may grant “undue preference” to profitable traffic (crtc 2015a; 2017b). Therefore, while practices that target internet traffic for discrimination are problematic in the discourse of net neutrality, some forms of discrimination are more problematic than others. There are net neutrality advocates who accept the need for network monitoring and discrimination to deal with security threats, or for other legitimate network management purposes (Hecht 2015; van Schewick 2010). A more nuanced discussion of net neutrality shifts the question away from whether discriminating among traffic is a legitimate practice by isps, to ask which discriminations are legitimate. The most accepted forms of differential treatment involve security (see chapter 7), and the most strongly opposed are ways of directly maximizing profit, such as vertically integrated incumbents privileging their own content or providing certain apps a “fast lane” in exchange for payment (fcc 2015). In other words, discriminating for reasons of pure self-interest is the greatest concern, while discriminating against threats to collective interests is either tolerated or welcomed. Congested Pipes and Bandwidth Hogs Typically, net neutrality controversies involve the interconnections between an isp’s network and other networks, services, or servers. Somewhere, a material connection is made when a cable is extended from one network to another, perhaps at a location such as 151 Front Street West (see chapter 1), allowing decisions to be made about the data flowing through this cable. These are border conflicts, sometimes involving the use of deep packet inspection (dpi) equipment to examine traffic moving in and out of the network in order to categorize its contents and act accordingly (Bendrath and Mueller 2011). As previously discussed, one border that has long been rife with conflict is the one separating incumbents from their wholesale customers, or iisp competitors. The mandated wholesale regime relies upon neutral incumbents providing connectivity to iisps. Since incumbents are the upstream party, they are in a position to govern the traffic flows of their downstream competitors. Furthermore, some of these traffic flows may be considered illegitimate, unlawful, or a form of “network abuse” (crtc 2008), further justifying discriminatory actions. These were the circumstances during the mid-2000s, when incumbents were

114

Telecom Tensions

complaining about “congestion” on their networks caused by a small number of “heavy users,” leading the crtc to establish the first piece of its net neutrality policy framework (crtc 2009). In 2008, video streaming services had yet to be widely adopted by the Canadian public and internet policy was still closely related to the “copyright wars” (Patry 2009) over file-sharing services and their vilified users. Other “heavy users” of the internet, such as people who were livestreaming, transferring large files, or playing online games (Lawford, Lo, and De Santis 2009), were relatively uncommon. When the demands of subscribers exceeded what the incumbent networks were designed to handle, the response of Bell and incumbent cablecos was to use their dpi surveillance equipment to identify these so-called “bandwidth hogs” and “throttle” them – selectively narrowing their pipes (Parsons 2013). This move would be contested by some iisps and users, who took issue with their traffic being “managed” in this way by the incumbents they depended on for connectivity. The regulatory battle began when Bell notified its iisp dependents that their traffic would be subjected to dpi and the throttling of peerto-peer file sharing in March 2008. Bell pointed out that this kind of traffic discrimination had already been applied to Bell’s own retail customers, so that everyone was being discriminated against equally; iisps were not being singled out for special treatment by their larger competitors (crtc 2008). But as wholesale dependents, iisps had no say in the matter – their users’ traffic would be governed according to the policies of the upstream provider, instead of the iisp with which they had a contractual relationship. This controversy helped elevate net neutrality activism in Canada, leading to the birth of OpenMedia and the SaveOurNet.ca coalition. The resulting actions included a May 2008 protest on Parliament Hill, public events in Ottawa, Toronto, and Vancouver (in which the iisp TekSavvy played a prominent role), and the first effort to enact net neutrality legislation (Nowak 2008; SaveOurNet.ca 2009). The regulatory consequences of the dispute, however, would be up to the crtc. While the specific case involving Bell’s throttling of iisp subscribers was decided by the crtc largely in Bell’s favour (crtc 2008), this led immediately to public hearings on the larger issue of what isps were allowed to do with internet traffic, under what circumstances. After hearing from isps about their respective dpi and throttling practices, the crtc acknowledged that these were relevant to the larger net neutrality debate, and released the framework for Internet

Common Carriage, Net Neutrality, and Bad Traffic

115

Traffic Management Practices, or itmps (crtc 2009). The Commission’s new policy required greater disclosure by isps about exactly how they were intervening in subscribers’ traffic, but largely approved the use of traffic discrimination along existing lines: itmps such as throttling could be used by incumbents handling iisp traffic as long as incumbents applied the same practices to their own retail customers. Congestion problems were a valid reason to use itmps, albeit with the understanding that the best solution to a clogged network is to invest in greater capacity, rather than relying on traffic management over the long term. Finally, issues related to network security got a free pass; itmps could be used to control “malicious software, spam, and distribution of illicit materials” under the rationale that “such activities are unlikely to trigger complaints or concerns … and are a necessary part of an isp’s network operations” (crtc 2009, para. 44). Of course, all of this would be enabled by dpi equipment surveilling traffic flows, but since isps assured the crtc that they were only interested in packets rather than the individuals responsible for sending them, they could carry on as long as they did not take advantage of the broad surveillance capabilities that dpi made possible (crtc 2009, paras. 100–5). The itmp framework, which would be touted as the world’s first net neutrality regulations, largely gave the crtc’s blessing to techniques that isps were already using to govern data flows. One could argue that this was because net neutrality was not really much of a problem – discrimination was taking place to manage the circulation of traffic, rather than to give undue advantages for competitive reasons. But the itmp regulations legitimized isps’ control over their users’ data flows. isps were conduits, but not merely so, and they were required to be neutral only in the sense that they were “competitively neutral” (crtc 2009). The principle of net neutrality had been translated into criteria for determining when discrimination was acceptable, and these criteria would be adjudicated by the Commission as complaints came forward. Most Canadian internet users were unaware that any of this had happened. Net neutrality had emerged as a public issue, but itmps remained telecom policy jargon, and what was being done to users’ traffic by dpi gear embedded somewhere in a carrier’s network was largely a mystery to anyone except the network operator. One could, if they tried, find the now-required disclosure statements on an isp’s website informing users that their BitTorrent streams would be

116

Telecom Tensions

throttled between such and such times, but accountability for these practices relied on complaints from users who suspected something improper was being done to their packets. The next two years would see more than thirty such complaints, often from people who were unable to substantiate their complaints to the crtc’s satisfaction (Geist 2011b). The most notable case involved a two-year struggle between Rogers and online gamers who came to believe that their traffic was being throttled. Teresa Murphy was one such gamer, who learned enough about crtc procedures to write a complaint that was taken seriously in 2011, marshalling the sorts of evidence she could gather from other gamers at the edges of the network. Murphy then co-founded the Canadian Gamers Organization with Jason Koblovsky, and their work helped to publicize and sustain the issue, eventually leading to the conclusion that Rogers had mistakenly classified gaming traffic as file-sharing, and compelling the company to change its practices in 2012 (McKelvey 2018, ch. 7). Today, isps continue governing traffic flows to target bad actors and undesirable packets, blocking security threats, child abuse imagery, and (as with GoldTV) limited forms of copyright infringement. However, since the 2010s, “bandwidth hogs” and network congestion are no longer presented as major public policy problems, with file-sharing traffic having declined. Meanwhile, overall internet use by Canadians has continued to climb steeply, thanks to the use of streaming video services such as Netflix (Sandvine 2014). Since this growing demand for network capacity has been caused by commonplace behaviour and typical internet users, it could no longer be managed by targeting a minority. With the exception of wireless and satellite services, isps generally moved away from “technical” itmps involving throttling (Geist 2012a) and focused on meeting their subscribers’ streaming expectations. While the first net neutrality battles concerned discrimination over wired broadband, the next regulatory phase would focus on video over wireless. The debate over technical itmps had been about the ability of isps to restrict or throttle digital flows; the following debate over “economic” itmps would be about whether some digital flows could be privileged, and given away freely. Net Neutrality and Price Discrimination Net neutrality can be enforced to minimize the extent to which isps actively intervene in traffic flows, but the crtc’s (2009) itmp frame-

Common Carriage, Net Neutrality, and Bad Traffic

117

work recognized that traffic management does not have to involve directly acting upon data packets. A city can manage traffic congestion by blocking certain types of vehicles from using a roadway, or it can make driving more expensive, using economic incentives and disincentives. Similarly, an isp can use dpi to detect and throttle certain kinds of packets, or it can impose different prices on a subscriber for different uses of the internet. Canada’s common carriage regulations developed around concerns over price discrimination for news services, and an isp might similarly play favourites through its billing practices. At the time of the initial itmp framework, economic itmps were seen as less problematic than technical ones. Particularly for managing congestion on wireless networks with scarce bandwidth, the crtc preferred that isps impose greater costs on subscribers who used more data, rather than using technical controls over their data flows. This gave subscribers greater choice and put “market forces to work” in governing behaviour (crtc 2009, para. 40). Until their suspension during the covid-19 pandemic, these “data caps” for phone subscribers continued to be common practice in Canada, charging subscribers more when they exceeded a certain amount of usage. The crtc’s acceptance of these practices was based on the fact that they treated all data the same. However, just as an isp can discriminate between packets, subjecting some kinds of traffic to throttling but not others, it is possible to make some packets count towards a subscriber’s data cap (potentially resulting in overage fees) while privileging a different kind of traffic by making it exempt from data caps. This practice, known as “zero-rating,” compelled a regulatory novice to take on some of Canada’s largest companies, and win. At the time that he submitted his formal complaint to the crtc, Ben Klass was a graduate student at the University of Manitoba, with a deep interest in telecom policy. In the summer of 2013 Klass had been in contact with Jean-François Mezei, who had recently blazed a trail as an independent expert on crtc disputes. Mezei had attended the 2008 net neutrality protest on Parliament Hill, and ever since had actively participated in crtc proceedings, making a significant contribution to the ubb issue discussed in chapter 4 (Chase and Marlow 2011), and helping Teresa Murphy with her gaming complaints (resa1983 2011). By 2013, Mezei had developed an understanding of crtc procedures at a level normally possessed by regulatory lawyers, and when Klass explained his concern over differential pricing being a violation of anti-discrimination rules and the itmp framework,

118

Telecom Tensions

Mezei encouraged him to formulate his complaint as a “Part 1 application” to the Commission. Klass, in contact with Mezei, developed his draft application over several months and submitted it in November 2013 (Klass 2015). The precipitating event had been Bell’s announcement earlier in the year that it would be offering a mobile TV service which would not count towards a subscriber’s data cap in the same way as watching Netflix or YouTube. This perk made it more attractive to be a Bell phone subscriber, as long as you were using your phone to watch Bell’s media offerings. The company was maintaining data caps to supposedly address congestion and limit subscribers’ network usage on the one hand, while also encouraging its subscribers to use more data to watch content supplied by Bell, as an exception to the data cap. This practice was not unique to Bell, and Klass’s complaint would soon be joined by a similar one against Rogers and Vidéotron launched by public advocacy groups. A key issue in the legal back-and-forth that followed was straightening out the roles that vertically integrated isps such as Bell, Rogers, and Vidéotron were playing. As both isps and television providers, these companies’ practices are governed by two different statutes: the Telecommunications Act and the Broadcasting Act. While the crtc is responsible for administering both, an actor’s responsibilities under these two laws differ, and may conflict. For instance, while Canadian telecom law prohibits discrimination, broadcasting law mandates discrimination by requiring Canadian programs to be carried (Denton 2015). Broadcasters are seen as originators of content, choosing what to offer, rather than as neutral intermediaries that are required to carry others’ communications. The question, therefore, was whether vertically integrated isps would have to comply with the non-discrimination requirements of telecom law, or whether they were subject to broadcasting law, which was much more permissive of discrimination. Bell based its arguments on the latter, claiming that as a broadcaster, common carriage and net neutrality concerns did not apply to it. Rogers voluntarily stopped its zero-rating practices partway through the proceeding. The crtc then rejected the role distinction articulated by Bell, concluding that it (and Vidéotron) played the role of common carrier when they provided data connectivity to mobile subscribers. If those subscribers chose to watch programs on their phones from Bell or Vidéotron, this did not mean that the isps were suddenly “transformed” into broadcasters; they remained common

Common Carriage, Net Neutrality, and Bad Traffic

119

carriers (crtc 2015a). Because the roles of intermediaries are defined by what they do, rather than what they are, these roles can overlap or conflict with one another. In the legal arena, an intermediary can argue that one of its roles should take precedence, to the exclusion of others. The crtc did not accept this argument, and held isps responsible as common carriers, whether or not they were also acting as broadcasters. While the crtc (2015a) rejected Bell and Vidéotron’s practices of giving preferential billing to customers using their mobile TV services, it did not accept that this kind of differential pricing was an economic itmp, but treated it as a violation of the common carriage requirements of the Telecommunications Act. Despite the decision, Vidéotron continued to offer a zero-rated streaming music service, resulting in further complaints and the crtc holding public hearings on the issue of differential pricing. The hearings drew a broad range of participants, some of whom even contributed through a Reddit forum that the crtc established as a more inclusive way to gather public comments. The result, as quoted earlier in the chapter, was the Commission embracing the language of net neutrality, and creating a framework that restricted differential pricing practices (crtc 2017b). Today, Canadian isps can remain components of giant, vertically integrated media conglomerates, but regulators have some new means of controlling the boundaries between content and carriage. The fear is that intermediaries with other business interests will continue to find ways to privilege certain uses of their networks, turning subscribers of services into consumers of content. Since the desire to create different kinds of walled gardens by linking carriage and profitable content is an enduring one in vertically integrated telecom (Denton 2017), regulators can counter by erecting barriers of their own. The net neutrality framework, in combination with updated broadcasting regulations (crtc 2015f), means that if an isp wants to offer its own video streaming service, it has to make it available to all across the “open internet” and cannot exempt its media offerings from data caps. Discrimination and preferential treatment continue in a variety of ways, but within these defined limits. THE F UTURE OF NET NEUTRALIT Y

In March 2020, Canadian isps suspended data caps to address the public need for greater connectivity during the covid-19 pandemic

120

Telecom Tensions

(Schwartz 2020). Millions practised physical distancing, isolating themselves from unnecessary bodily contact, and becoming bandwidth hogs out of necessity – whether by working from home or occupying themselves with streaming media. The response to a dangerous pathogen circulating in society was to unleash internet connectivity for the public good; but digital networks also circulate their own pathogens, in the form of “malicious” packets and other online threats. As will be discussed in chapter 7, practising internet hygiene requires discrimination by isps, and the techniques of cyber security are largely exempt from net neutrality regulations. But distinguishing good and bad traffic will continue to pose public policy challenges for regulators, while intermediaries struggle with related questions of internet governance in their own operations. Exercising certain controls over internet traffic makes good business sense, but regulators must also be aware that business decisions can sometimes be presented as necessary responses to some supposed threat. telus’s website-blocking in 2005, throttling in the years that followed, and GoldTV in 2019 were all net neutrality disputes that involved isps protecting their business interests. isps justified these interventions by pointing to bad traffic and bad actors: illegal content, bandwidth hogs, and pirates. In the differential pricing decisions that followed Klass’s complaint, when isps were privileging content rather than restricting it, this justification was simply not available. Vertically integrated isps argued that they were providing greater value for consumers, or experimenting with new business models, or that their actions were not governed by telecom law. While these arguments were unsuccessful and the federal government now professes a commitment to the principles of net neutrality for all “legal” content (Chhabra 2018a), the future will continue to see isps discriminating against undesirable traffic and favouring certain services or data flows. We have been repeatedly told that the future of the internet will be a world of connected “things” communicating wirelessly. Tyson Macaulay, formerly a Bell Security Liaison Officer, describes net neutrality as a “threat” to this internet of things (Macaulay 2015; 2017), where densely interconnected devices using precious bands of the electromagnetic spectrum require traffic prioritization. United States fcc Chairman Ajit Pai, speaking to the 2017 Canadian Telecom Summit, likened his country’s short-lived net neutrality regime (which was also based on common carrier principles) to a “black cloud” that had

Common Carriage, Net Neutrality, and Bad Traffic

121

hung over isps, preventing innovation, stifling investment, and blocking progress towards the “digital future [of] smart infrastructure” (Pai 2017). To get the full benefit of smart technologies, these arguments assert that traffic discrimination will be required, and that net neutrality regulations can be an obstacle to technological progress. But the tensions covered in this chapter are not about what a technology requires from public policy; instead they concern the business practices of intermediaries. Common carriage and net neutrality rules are a way to keep an intermediary’s roles separate in a political economy that tolerates giant firms. Where vertically integrated conglomerates can expand (“diagonally”) into markets adjacent to telecom, or make deals with corporations in other industries, their interests broaden beyond simply carrying the communications their users want. If a pure intermediary is a relay, passing along messages or content, additional roles introduce additional preferences for how to handle communications: affecting their priority, cost, or whether some are blocked while others are allowed to pass. A telecom company that also produces media content would rather that its telecom subscribers use their connections to enjoy the company’s media rather than that of its competitors. telus stands out against its vertically integrated competitors in that it is not a major content owner (cmcrp 2019), but like its competitors, telus has in recent years been offering an internet TV service to its subscribers on an exclusive basis. These exclusive arrangements are permitted under the regulations covered in this chapter because they are delivered over a “managed service,” distinct from the rest of the internet – a distinction that might one day be challenged before the crtc (Nowak 2017). It is through these exceptions to net neutrality that giant firms are able exploit new business opportunities, and why it is important to pay attention to all of the ways that net neutrality regulations do not apply. Whether or not isps are vertically integrated, the process of intermediation continues through the accumulation of new roles and lines of business. One of the most recent trends has been for intermediaries to become more like platforms (Newman 2019, ch. 8), expanding their offerings into home automation, medical services, security, and surveillance systems (O’Rourke 2017; Wire Report 2019b), thereby becoming points of control for our networked devices, our safety, and our health. These services can be bundled, tied to specific hardware, and sold exclusively to telecom subscribers, increasing customer dependency and making it more difficult to switch providers

122

Telecom Tensions

(van Schewick 2010, ch. 6). If net neutrality is meant to address the possibility of isps constructing exclusive walled gardens for content and services, intermediaries are becoming increasingly active within the walls of our homes, and hoping that this expanded role will keep us within their interconnected “ecosystem” (Karadeglija 2019). Net neutrality needs to be understood as a contemporary policy response to the kinds of power relations that have developed between intermediaries and their users under regulatory capitalism. Given this political economy of telecom, it can be entirely appropriate to limit exploitation of the internet’s points of control, but this is not an argument against intermediation in general. Whether or not intermediation or net neutrality constraints are desirable depends on the set of responsibilities we are considering and our tolerance for the sorts of corporate entities discussed in this chapter, and relates to the concerns over competition policy covered in chapter 4. A conglomerated service provider may be able to meet a host of our needs, all bundled together in a monthly subscription, and this may even produce “innovative” and cost-saving connections between these services. However, it also becomes more difficult for competing services to gain a foothold against such a platform and the overlapping ways that customers come to depend on it. The more interrelated these lines of business become, the more important regulators become for maintaining neutrality among intermediaries and addressing concentrations of power. The discourse of net neutrality provides one way to challenge power relations in Canadian telecom, but the discourse of privacy is another; both prohibit intermediaries from adopting certain roles and limit what they can do with traffic. Like net neutrality, Canadian privacy law has its own broad exemptions permitting data collection and disclosure, but privacy law has also been used as an effective obstacle to certain kinds of intermediation, such as increased surveillance and commercialization of user data by isps. Unlike net neutrality, however, privacy protections also govern the relationship between intermediaries and state agencies, limiting the roles that isps can play in policing society. The next two chapters will address intermediaries’ roles in relation to privacy and security, with isps discriminating and taking action against various threats, while also protecting the privacy interests of their users.

6

Internet Service Providers as Privacy Custodians

INTRODUCTION

On 13 June 2014, Matthew David Spencer’s criminal case concluded at the Supreme Court of Canada, where he had been appealing a conviction for child pornography possession nearly seven years after the police had searched his home. The court upheld his conviction, which meant that Spencer was still guilty of the crime, but the reason the Supreme Court had decided to hear the case was that it raised questions about basic internet privacy rights. In this regard, the country’s highest judges found that the police’s investigation of Spencer had been unconstitutional. Spencer was convicted on the basis of evidence obtained from his computer after a court-ordered search, but what led the Saskatoon Police to the address where he lived was a digital search of ip addresses, through the voluntary assistance of an isp. Police investigators had done the detective work, but Shaw Communications had voluntarily provided the crucial piece of information that led to Spencer’s door. It was this “pre-warrant” digital search that the Supreme Court unanimously found was unlawful, in that it should have been carried out using a court order. While various police agencies and isps across the country had previously interpreted the law to allow such warrantless investigations, they were now informed that their actions had actually violated Canadians’ constitutional rights hundreds of thousands of times a year (Office of the Privacy Commissioner 2015a). Simultaneously, isps now had a redefined duty as privacy custodians to protect subscribers’ anonymity, “guarding the link between the information and the identity of the person to whom

124

Telecom Tensions

it relates.”1 isps can be surveillance partners or the public’s guardians of privacy, and as the next two chapters show, events such as the Spencer decision help to define the contours of this often-contradictory dynamic. This chapter is about how intermediaries govern the privacy of their subscribers and users, and how this role has developed over time. Where isps are conduits of information, they act as privacy custodians by guarding the privacy of these communications. Where isps have access to vast amounts of sensitive personal information about their users, including their browsing histories, interests, habits, and secrets, they protect privacy by keeping this information anonymous and refusing to share it. isps also act as privacy custodians by limiting the amount of information they collect, even when collection might be easy and profitable. Finally, the role of a privacy custodian extends beyond a relationship with users and their personal information and is played out through a broader relationship with the public. This includes efforts by some companies to educate the public about privacy, as well as the accountability regimes through which organizations report how they gather, handle, and disclose personal information. As privacy custodians, intermediaries must navigate role conflicts when they assist government agencies and other organizations investigating their customers and users. isps’ roles as guardians and caretakers of personal information have competed or conflicted with other pressures and public policies, including copyright, policing, and cyber security. In order for isps to operate as agents of surveillance, identification, and security, they must reconcile these roles with their conduct as privacy custodians. Privacy custodians have both positive and negative legal obligations, which is to say that there are certain things privacy custodians must do, and other things they must not do. The key privacy legislation that regulates Canadian isps is pipeda (the Personal Information Protection and Electronic Documents Act), which imposes positive responsibilities to ensure isps identify the purposes of data collection, obtain consent, ensure the accuracy of information, and use appropriate safeguards. However, the privacy responsibilities that have been at the centre of internet policy debates are expressed primarily in negative terms. These are obligations not to collect, use, or disclose personal information, or to place specific limits on these activities. pipeda’s negative responsibilities may conflict with positive pressures that encourage or require intermediaries to do the opposite.

Internet Service Providers as Privacy Custodians

125

It is important to note that the above distinction between negative and positive duties or responsibilities is not always straightforward. Many negative responsibilities have a corresponding and equivalent positive responsibility, and vice versa (M.G. Singer 1965). If we say an isp is obliged to be open and accountable about what information it collects, this has a corresponding obligation for the isp not to be secretive and misleading. However, limitations on disclosure (“shall not...”) are essentially negative, and require some additional reasoning to translate to a positive duty. This is precisely what some intermediaries have done in a number of examples covered in this chapter, by interpreting negative limitations as obligations to take privacyprotecting actions. Effectively, this means that some intermediaries have been adopting a more positive orientation towards their privacy responsibilities and playing a more active role in determining what it means to be a privacy custodian. For intermediaries in these cases, privacy is not just a legal standard to meet, but a value to fight for, even if this means challenging state actors or opposing court orders. This trend reflects the growing recognition of the important role played by privacy custodians, and the efforts of some organizations to differentiate themselves by demonstrating (rather than merely stating) their commitments to the privacy of their users. However, this trend has been neither consistent nor strong, since intermediaries’ roles as privacy custodians are continually in conflict with contradictory pressures, and organizations are selective in deciding when to exceed their minimum privacy obligations. Intermediaries, whether these are massive corporate isps or not-forprofit community networks, exercise a degree of autonomy when making decisions such as whether or not to cooperate with law enforcement, or what it means to safeguard their users’ privacy. We can see this through the different approaches that isps acting as privacy custodians have adopted regarding issues such as privacy policies, the amount of time they choose to retain personal information, how they respond to copyright notices, and how they share information for reasons of security and law enforcement. For each of these questions, intermediaries must find ways of resolving the tensions around their responsibilities as privacy custodians when these conflict with other roles, and are subject to contradictory pressures. These pressures can be applied by copyright owners or government agencies, or result from an intermediary’s own desire to collect and use personal information in the pursuit of greater profits.

126

Telecom Tensions

As privacy custodians, isps must safeguard communications and personal information, and be accountable to their users regarding these practices. In their surveillance roles, isps collect and disclose information, and are often prohibited from notifying their users or the public of such activities. Alternately, isps might have other reasons to surveil their users (for profit, or network management), and find ways of reconciling these actions with privacy commitments. Although advocates for surveillance practices often claim that intermediaries can collect or disclose personal information while respecting privacy, the contradiction between the two has been made clear through a number of legal contests covered in this chapter. For isps acting as privacy custodians, accommodating some desires or expectations creates tension with others. Satisfying police demands for information leads to dissatisfaction and criticism from privacy advocates and some users. In response, privacy custodians have frequently justified their actions as being limited to, or constrained by, what is required of them by law. And yet, companies operating in the same legal jurisdiction have taken rather different approaches in interpreting just what their legal obligations consist of. Indeed, citing what is “required by law” can obscure the amount of latitude that intermediaries have in their actions, including the choice to test or challenge the law. isps thereby minimize their discretion in public statements, even as they exercise it. In one remarkable example of the tensions between the intermediary responsibilities described above, telus took on a positive stance in 2010 by pushing back against an over-extended court order to disclose text messages.2 This move to resist the Crown was touted in a blog post as part of the company’s “commitment to privacy” (Blackburn 2013). The same blog post describes how the company also assists police “in obtaining customer information they need for an investigation – but only when telus is ordered by a court to do so ... We never voluntarily turn over customer information.” The post highlights how telus chooses to “fight for [their] customer’s privacy all the way to the Supreme Court of Canada,” but explains that when the company supports the police in fighting crime, it does so out of legal duty. Missing from the text is any acknowledgment that the company voluntarily provided information to police tens of thousands of times in 2013 and previous years, in a way that the Spencer decision would later find to be unlawful (telus 2014). In short, telus presented its defence of subscriber privacy as a choice,

Internet Service Providers as Privacy Custodians

127

while voluntary cooperation with police was unacknowledged, or presented as an obligation. telus was certainly not unique in quietly volunteering information to police while publicly defending its practices using the language of legal obligation (Ling 2014; Mediacaster 2011), but the conduct of privacy custodians across the industry has also differed greatly in some respects. The variability in how these companies govern privacy is partly related to the diversity of actors that can be considered isps, including small businesses, publicly funded institutions, and not-for-profits. Indeed, this diversity has been a problem for police and security agencies, but the roles of intermediaries as privacy custodians developed before pipeda and the internet, and it is this history to which I now turn. PRE - INTERNET INTERMEDIARIES AS PRIVACY CUSTODIANS

While the expectations applied to intermediaries in Canada varied over the course of the nineteenth and twentieth centuries, the idea that these organizations had a duty to protect the privacy of communications is almost as old as the idea that they had a duty to help law enforcement. Initially, however, the greatest threats to the privacy of users came from telephone company employees who were required to listen to calls, as well as from other subscribers listening on a “party line.” By 1918, companies were subject to provincial legislation in Ontario and Quebec that regulated eavesdropping and made it an offence to divulge the content of a telephone conversation (Martin 1991, 141–5). More general privacy laws developed gradually in Canada over the latter half of the twentieth century and into the twenty-first (Power 2013). However, until the 1970s, eavesdropping and surveillance restrictions often did not apply to police investigations, so formal requirements for intermediaries to assist police in intercepting communications were rarely stated explicitly (Cornfield 1967; MacDonald 1987). Rules prohibiting intermediaries divulging private communications could be overcome through some form of “lawful authority.”3 Intermediaries (specifically, telephone and telegraph companies) did apparently assist police investigations and enable wiretaps, either voluntarily or based on the understanding that they were required to do so, but the extent and nature of this surveillance is unclear.4 Eventually, concerns about electronic surveillance and the lack of police accountability (Canadian Committee on Corrections 1969) led

128

Telecom Tensions

to police guidelines, and then to the Protection of Privacy Act in 1974 (Rabideau 1991, 172–3). However, the Privacy Act was widely known as the “wiretap bill” while it was being debated (Winsor 1973), since its most notable feature was that it created a formal legal route through which the police could obtain wiretaps. Privacy protection was thereby legislated hand-in-hand with surveillance powers; while “invading” the privacy of subscribers became a criminal offence, police could legally carry out such invasions to gather evidence. The Privacy Act established a hierarchical relationship between the roles and responsibilities of telephone companies. On the one hand, intermediaries were constrained as carriers of communications, and were obliged to respect the privacy of subscribers. However, the exemptions listed in what would become Part VI of the Criminal Code effectively trumped this responsibility, specifying instances in which privacy invasions could be justified or legally authorized. Intermediaries now clearly had a vital role to play in safeguarding privacy, but their privacy obligations were secondary to their judicially authorized surveillance responsibilities. Additionally, the prohibitions against privacy invasion could be voluntarily overcome by intermediaries for other reasons, such as when necessary to provide a service or in the interests of safeguarding subscribers. By the 1990s, the legal authorities under which police could demand or request access to subscriber communications or records held by telephone companies were fairly clear, but the attempt to extend such rules to cover isps would set off lengthy and contentious debates. INVISIBLE HANDSHAKES AND INTERNET SURVEILLANCE

During the early public internet of the 1990s, police forces in general devoted little attention to combating “cybercrime,” and saw little value in online investigative and surveillance techniques until the end of the decade. According to Birnhack and Elkin-Koren (2003), ordering the online field was initially left to the “invisible hand” of the market and the new private intermediaries. In the 2000s, state agencies sought to make an “invisible handshake” with these intermediaries, recruiting and co-opting them towards state goals. Data packets may frequently cross international borders without inspection, but they are carried over a physical transport infrastructure that is, by necessity, a local asset, and therefore subject to state control (Goldsmith and Wu 2006). Governments recognized that the cooperation of isps

Internet Service Providers as Privacy Custodians

129

would be key to policing and surveillance in the online era, and some of these isps (the former telco monopolies that emerged as isps in the mid-1990s) were already familiar partners for state agencies. While some authors writing in the 2000s warned that isps would become close surveillance partners of state agencies (Birnhack and Elkin-Koren 2003) and copyright owners (de Beer and Clemmer 2009), these authors did not fully appreciate the ways that the interests of these groups were fundamentally misaligned. Such conflicts came dramatically to the fore in battles over copyright reform, where the business models of intermediaries and copyright owners were at odds (see Edwards 2011; Haggart 2014). When it came to state surveillance efforts, isps were more willing to act as deputies or partners, but even this cooperation had its limits, and the tension became more pronounced following the Snowden disclosures of 2013. In both cases – copyright enforcement and state surveillance – isps’ roles as privacy custodians would come into conflict with expectations that these institutions would act as internet police or surveillance deputies. However, it was in a legal dispute concerning copyright that Canadian isps would first assert the privacy interests of their subscribers, interpreting a negative responsibility not to disclose information as a positive duty to oppose such disclosure. PRIVACY AND PIRACY

In 2004 a group of incumbent isps took an active stand to protect the privacy of their subscribers against a group of copyright owners. In what would become a landmark legal decision,5 the isps acted to safeguard the interests of their customers: the “John and Jane Does” that copyright owners sought to identify and sue. The copyright owners had hired a copyright surveillance company, MediaSentry, which had recorded Canadian ip addresses sharing copyrighted music. In order to link these ip addresses to internet subscribers who could be sued for copyright infringement, the copyright owners needed the help of Canadian isps. However, isps had a legal obligation to limit the disclosure of personal information as a consequence of pipeda, which had recently come into effect. The Federal Court was in a position to decide between these competing demands, and had the power to order the isps to cooperate. The question being decided in 2004 was: Who can have access to information about us, and in what circumstances? Here, the data that

130

Telecom Tensions

copyright owners requested was limited to people’s identities, but the intent was to link those identities to online behaviour that had already been recorded. This is the link that allows a person to be sued for copyright infringement; it provides the John and Jane Does with proper names. But how do we make the connections between an ip address recorded by MediaSentry and a person in their home on a computer – a person who can now be served with legal demands and the prospect of paying a punishing fine? Should courts demand some standard of evidence before ordering isps to identify subscribers? When can information about digital networks, harvested by humans and algorithms, become a record that is linked to a body? How are bodies held accountable for online data flows (Poster 2006; Zajko 2015b)? The 2004 court case of BMG Canada Inc. v Doe, and its appeal the following year,6 only provided some preliminary answers to these questions. Effectively, the result was that copyright owners were invited to try again with better evidence, but this court decision came at the tail end of the period in which major US–based record labels had been suing individual file-sharers (Poster 2006). While the major record labels never did return to this legal strategy in their battle against piracy,7 this early file-sharing case is instructive for my analysis by showing the different positions taken by the five isps involved.8 All were acting as privacy custodians for their respective “Does.” The companies were entrusted as keepers of these subscribers’ identities, but how they acted on this responsibility varied. Importantly, some isps rationalized their role as privacy custodians in a way that included a positive responsibility to assert the privacy rights of their subscribers. Others adopted a more negative orientation – treating privacy responsibilities as a set of conditional obligations not to disclose personal information. The case would be an important episode in a process that would see some intermediaries adopt a more positive conception of privacy stewardship. The lowest degree of involvement in BMG Canada Inc. v Doe came from Vidéotron, which chose not to participate in the court proceedings. The company’s guardianship of its subscribers’ identities amounted to waiting for the court order to be issued before these identities were disclosed – a position that could be attributed to the company’s various media interests and anti-piracy stance. This was a negative orientation to the company’s privacy responsibilities, which were interpreted as a prohibition against disclosing customer information to a private party in the absence of a court order.

Internet Service Providers as Privacy Custodians

131

On the other hand, Shaw provided the greatest resistance to the motion, pushing back to protect the privacy of the company’s customers. The positive duty of a privacy custodian expressed by Shaw amounted to asserting the privacy rights of the company’s anonymous subscribers, by making arguments on their behalf to oppose the court order sought by the copyright owner. Meanwhile, Bell and Rogers both submitted privacy-related arguments, and telus complained about being conscripted to carry out investigations for the copyright owners, which might result in the company being sued by its own customers. In the end, the trial judge9 refused to grant a court order to compel the isps to identify their customers on several grounds. Shaw and telus publicly claimed credit for protecting the privacy of their subscribers (G. Shaw 2004), and the decision solidified the duty of isps to protect the identities of their subscribers, as well as the conditions under which such protection could be overcome in a lawsuit (Geist 2005; Knopf 2013). While this particular episode is a good starting point for considering how isps developed into the trusted privacy custodians upon which we are all now so dependent, it is certainly not the beginning of a straight trajectory through which intermediaries became increasingly vigorous defenders of subscriber privacy. BMG Canada Inc. v Doe was a case of isps asserting their positive responsibilities as privacy custodians, but this was followed by inconsistency, ambivalence, and contradictory attitudes by intermediaries towards the privacy of their users. In subsequent years, isps involved in copyright cases interpreted their responsibilities as privacy custodians in a variety of ways, but they rarely sought to use these cases to advance the privacy interests of their users. When file-sharing lawsuits were reintroduced to Canada in 2011, the three isps involved (Bell, Cogeco, and Vidéotron) did not oppose the copyright owner’s attempt to obtain a court order to identify their subscribers (McKenna 2011; Mediacaster 2011). In a recent case before the Supreme Court of Canada, Rogers contested a motion to identify its customers, but it did so on the issue of costs rather than privacy grounds.10 Privacy arguments were advanced by TekSavvy in a 2012 case, although the isp did so without opposing the court order.11 Distributel made a privacy-related argument and successfully opposed a motion to identify subscribers in 2013.12 Finally, in 2019 TekSavvy’s privacy advocacy and opposition to a court order led to success in Federal Court. The copyright owners had demanded subscriber informa-

132

Telecom Tensions

tion from thirteen Eastern Canadian isps, but TekSavvy alone appealed the court order and identified problems with the copyright owners’ evidence.13 These preceding examples demonstrate how Canadian isps operate as privacy custodians in the context of file-sharing lawsuits by copyright owners. In these cases, the negative responsibilities of privacy custodians have been fairly clear and uncontroversial under pipeda, as requiring a court order prior to any disclosure of personal information. Positive responsibilities have depended more on the discretion of the intermediary, with some choosing to be more assertive than others in defence of users’ privacy interests. In contrast, when intermediaries act as privacy custodians responding to police agencies, pipeda’s negative responsibilities have been more easily neutralized by exceptions written into the statute. Until the Spencer decision, intermediaries had wide latitude to decide when to insist on a court order in law enforcement matters (L. Singer 2012). The “lawful authority” exception has been a key point of contention in the story of lawful access in Canada, and its effective elimination in Spencer continues to animate law enforcement advocates. LAWFUL AUTHORIT Y

What is known as “lawful access” includes a number of positive duties, or requirements for intermediaries to assist state actors in obtaining information or carrying out surveillance. These duties involve complying with court orders, such as those mandating wiretaps. A related positive responsibility is the mandatory disclosure of information relating to child pornography (Valiquet 2011). However, the debate over lawful access has frequently focused on the negative responsibilities of privacy custodians, and the scope of the exceptions to these responsibilities. In other words, an intermediary’s responsibilities not to disclose personal information can serve as a barrier to lawful access, but one that can often be overcome by state actors through exceptions to privacy law. Privacy laws such as pipeda share some foundational principles, which state that consent must be given for personal data to be collected or disclosed, and that collection and disclosure must be limited to a specified purpose. This means that isps wishing to collect information about us must have our agreement, and we grant this consent by clicking “I Agree” on a page of lawyerly text many of us

Internet Service Providers as Privacy Custodians

133

have never tried to read. Once an isp collects our information, whether our names, credit card information, or browsing history, it can only use this information for the purposes to which we (supposedly) consented. The exception to obtaining consent for any disclosure of personal data is where such disclosure takes place “by the authority of law” (oecd 1980). Therefore, while a prohibition against the disclosure of personal information without consent is a key part of privacy law in Canada, pipeda (and related privacy legislation) includes a substantial list of exceptional circumstances in which the non-consensual disclosure of personal information is legally permitted (see Power 2013, 79–81). isps are legally permitted to disclose personal information to other parties during various kinds of investigations, to comply with court orders, in matters of national security, in emergency situations, and for contacting next-of-kin in instances of injury or death. In these circumstances, information that has been collected by an intermediary for other reasons (such as managing a subscriber’s account) can be shared without the knowledge or consent of the subscriber. The most important of these subsections in debates over intermediaries’ lawful access responsibilities has been section 7(3)(c.1),14 which permits disclosure to a “government institution” that has “identified its lawful authority.” This section was reportedly introduced into pipeda “by Industry Canada as a result of representations made by law enforcement and national security agencies. The intent as explained to Parliament was to maintain the status quo for these agencies to allow them to engage in pre-warrant intelligence gathering” (Morin 2011, 4). As the following chapter demonstrates, such “status quo” arguments have commonly been deployed by lawful access advocates. Where lawful access critics see expanding state surveillance powers being pushed through new technologies, lawful access advocates see a way to preserve the reach of the state, which is being challenged by new technologies (Nevis Consulting Group 2003). pipeda left lingering ambiguities about the contours of the “status quo” that its “lawful authority” section preserved. Eventually, court decisions would provide greater guidance as to how the law should be interpreted, but police agencies and isps needed some shared understanding of how to govern personal information without waiting for these precedents to be established, particularly regarding the issue of online child pornography (known as internet child exploitation). To this end, the Canadian Coalition Against Internet Child Exploitation (ccaice) was

134

Telecom Tensions

formed in 2005 as a partnership between incumbent isps, industry associations, and government agencies (ccaice 2005). Business competitors joined in partnership with one another and police forces to tackle a pressing moral issue that, in much of the world, has been the “vanguard” of internet regulation (Mueller 2010, 190). Where other illegal activities utilizing the internet are largely ignored by isps and police, the abuse of children and the distribution of child pornography have since the 2000s been fought using specialized investigatory teams and international cooperation. isps have also taken exceptional steps, assisting police without compensation, and using governmental techniques (such as internet censorship) that might be seen as illegitimate if used against other kinds of illegal content. The protocol for information disclosure designed through the ccaice in 2005 was intended to be one of these exceptional measures, which would be legal (or “privacy compliant”) under pipeda, but which participants agreed would only be used in internet child exploitation investigations (Morin 2011). When police obtained an ip address that they suspected was linked to child pornography, they could use an agreed-upon template letter to submit this information to an isp and receive in response an internet subscriber’s name and address information. No warrant would be required, under the justification that the information requested was not particularly sensitive and the process was akin to police “walking down the street and asking questions” (Morin 2011, 9). These requests for disclosure were based on the police officer’s “lawful authority,” and the cooperation of isps was voluntary. While the ccaice protocol established a shared interpretation of what pipeda permitted and what intermediaries would disclose to police on request, it did not preclude other intermediaries from interpreting pipeda’s “lawful authority” section differently, or adopting a stricter stance as privacy custodians (Lawson 2011; L. Singer 2012). Disclosures were made in cases not involving child exploitation, and police sometimes requested more than just name and address information (Ling 2014; Office of the Privacy Commissioner 2015a, 7). While, in some cases, a simple phone call from a police officer requesting information was enough for an intermediary to share its records, other organizations set the bar higher, and a small number refused any compromise of user privacy unless mandated by law. An intermediary’s resistance could be grounded in a commitment to its users or concerns over corporate liability for privacy violations. The difficulty

Internet Service Providers as Privacy Custodians

135

of obtaining the cooperation of recalcitrant intermediaries created investigative obstacles and delays, and according to proponents of lawful access legislation, this created a need for revisions to pipeda that would mandate and standardize compliance (Public Safety Canada 2012). After successive governments failed to pass lawful access legislation, isps’ discretion was curtailed by the Supreme Court’s Spencer decision, which strengthened privacy custodians’ negative responsibilities and effectively barred most voluntary disclosures based on “lawful authority.” However, the role conflict faced by intermediaries serving the needs of government agencies and acting as privacy custodians for their users remains an underlying tension. LAWFUL ACCESS IN THE BALANCE

The role conflict at the heart of lawful access has been the topic of repeated consultations and public debates in Canada, where surveillance and privacy have often been presented as trade-offs, or elements that must be balanced with one another. Lawful access proponents have rarely argued that privacy should be sacrificed in the interests of security, tending to present lawful access as respecting, protecting, or safeguarding privacy rights. However, such claims are frequently accompanied by the metaphor of “balance” – between privacy on the one hand, and on the other what is variously described as security, the needs of law enforcement, public safety, or the public interest. Supporters of lawful access legislation claim that new police powers strike an appropriate balance (CBC News 2009; Toews 2012), while critics argue that they tip the scales against privacy (Stoddart et al. 2011), or point to the incompatibility of privacy rights and the proposed surveillance measures (Cavoukian 2011). The metaphor of balance is often used to frame debates about surveillance and security in a liberal political context, and there are ample reasons to doubt whether security and privacy are somehow exchangeable (Neocleous 2008, 12–13). However, in Canada’s lawful access debate, “balance” became a way of talking about intermediaries’ role conflict, wherein companies were expected to act as both privacy custodians and surveillance partners. The limits of what an intermediary could do in one of these roles would end up defining the other. Until 2014 (pre-Spencer), intermediaries were able to resolve their role conflict by claiming compliance with negative privacy responsi-

136

Telecom Tensions

bilities while voluntarily disclosing personal information, either on an “ad hoc basis” (Public Safety Canada 2013c, 57) or through protocols established between industry and government (Morin 2011). In other words, the obligations of an intermediary as a privacy custodian extended only as far as pipeda’s “lawful authority” exception, which granted a great deal of latitude to disclose personal information on request. This latitude included the possibility of denying requests from government, and maintaining the confidentiality of subscriber information. Intermediaries were able to find their own “balance” in this role conflict, and could choose to extend either privacy protection or government cooperation to the legal limit. While uncommon, refusals to assist government surveillance were seen as unacceptable by lawful access proponents, who sought to replace voluntary with mandatory cooperation. In general, intermediaries avoided taking a public position on either side of the lawful access debate. Industry representatives accepted government needs as legitimate, but made their support conditional on having specific concerns addressed. Telecom companies were primarily interested in clearly determining their obligations under privacy law and lawful access legislation, wanting to know how legal changes would impact their operations. In other words, the industry’s main concerns were over the costs and obligations of lawful access (CBC News 2009; Industry Canada 2014a, 10–30; Nevis Consulting Group 2003; Public Safety Canada 2011b) and not over whether lawful access would make them less effective as privacy custodians. It seems that the freedom of individual intermediaries to resolve this role conflict on their own terms would readily be exchanged for legal clarity and financial compensation, and it is only recently that some intermediaries have been publicly asserting themselves as privacy custodians in this debate (see below). In 2012, the most significant attempt to pass lawful access legislation (as Bill C-30) was widely opposed and proved politically untenable (Ibbitson 2012), thereby failing to limit intermediaries’ discretion. Limitations on intermediary conduct were imposed by the Spencer decision in 2014, which effectively declared most warrantless disclosures to be unconstitutional. This fixed the “balance” of competing roles further in favour of privacy responsibilities than lawful access proponents had hoped. However, even with the restrictive implications of this decision, intermediaries have found novel ways to define themselves as privacy custodians. Additionally, lawful access

Internet Service Providers as Privacy Custodians

137

proponents (unhappy with the implications of Spencer) have renewed their push for surveillance capabilities and police powers, giving intermediaries another opportunity to take a public stance on these issues. REDEF INING PRIVACY CUSTODIANS THROUGH POSITIVE RESPONSIBILITIES

I have previously described how intermediaries developed as privacy custodians, and some of the different ways that organizations have defined themselves in this role in recent years. The lawful access debate highlighted the inconsistencies among privacy custodians dealing with a longstanding role conflict, and the overwhelming public opposition to Bill C-30 likely convinced some intermediaries of the importance that users attached to privacy. However, the lawful access saga was generally not used by intermediaries to assert themselves as privacy custodians, and privacy interests were primarily represented by other actors in the debate. It is only since the latest lawful access consultation in 2016 that some intermediaries have publicly championed privacy interests (Braga 2016; Solomon 2016), but even before this, there are a number of ways in which some intermediaries have recently asserted the privacy of users and redefined themselves as privacy custodians by adopting certain positive responsibilities. I have previously discussed this trend in relation to copyright cases, but recent years provide a number of other noteworthy examples. As described above, Canadian intermediaries have generally avoided pushing for stronger privacy laws, or (until recently) against weaker privacy obligations. However, some have fought legal battles to assert the rights of their users under the laws that do exist. telus and Rogers have both gone to court to argue against police overreach in lawful access cases.15 As in BMG Canada Inc. v Doe, these companies have rationalized their negative responsibility not to disclose (in the absence of a court order) as a positive responsibility to oppose an improper court order. In both cases, a judge ruled the court order to be inappropriate or excessive, but it is important to keep in mind that these instances are not representative of the thousands of court orders that either company annually complies with. While rare, these examples of active opposition provide privacy custodians a means of publicly affirming their privacy commitments (Blackburn 2013; Dobby 2016), and differentiating themselves from other companies that have presumably complied with similar orders (see Fraser 2016).

138

Telecom Tensions

In recent years, some Canadian intermediaries have also found ways to redefine themselves as privacy custodians without positioning themselves in court. The cases described above have set the rules about personal information and its disclosure, configuring “mechanisms of security” (Foucault 2007) such as Canada’s lawful access regime, which both enables and prohibits surveillance, constituting the procedures that hold people accountable for data flows. But some intermediaries have also volunteered their governmental capacities as a productive force, working to shape privacy-conscious subjects or well-informed “digital citizens” (telus 2020). The most notable of these has been the telus Wise program, which provides free seminars, school presentations, and online resources. telus Wise operates in partnership with MediaSmarts (MediaSmarts n.d.), a longstanding digital literacy organization that also promotes privacy awareness, and that has been funded by many Canadian isps. Rather than privacy custodianship being limited to the organization’s conduct, here we see organizations working to cultivate practices that promote privacy in the population. The 2013 Snowden disclosures prompted a great deal of privacyrelated concern among internet users, giving Canadian intermediaries an opportunity to promote themselves on the basis of privacy and security (Miller 2014). At a Canadian industry conference I attended in 2014, an executive from an incumbent isp presented the argument that service providers had an opportunity to gain a competitive advantage by offering better security, showing a photo of Edward Snowden as an answer to the question of why we care about privacy and security. In the wake of high-profile data breaches and revelations of sophisticated surveillance programs, many customers have wondered about the extent to which intermediaries are protecting their information. As a result, some privacy custodians have taken proactive steps to improve such safeguards, and to engender trust with consumers by communicating corporate privacy practices. One significant development towards a more positive role for intermediaries as privacy custodians has come in the form of “transparency reports,” and greater disclosure of these companies’ role in facilitating state surveillance. Since 2010, numerous (primarily US–based) companies have been issuing such transparency reports, documenting how they handle requests from both governments and private actors to remove content or disclose information. Canadian intermediaries began doing likewise after the lawful access controversy raised con-

Internet Service Providers as Privacy Custodians

139

siderable questions about the telecom industry’s disclosures of personal information, after Edward Snowden heightened concerns over state surveillance, and after a researcher with the Citizen Lab (Christopher Parsons) made inquiries (Freeze, Dobby, and Wingrove 2014). Even so, the release of the first of such reports in 2014 came in the absence of an immediately precipitating crisis, scandal, or “sudden focusing event” in Canada’s telecom industry (Parsons 2017, 113). TekSavvy and Rogers were the first to publish reports, despite the role conflict created through contradictory lawful access responsibilities that appeared to restrict such disclosures, with Rogers effectively declaring a new approach to privacy and surveillance in which “the needs of [Rogers] customers came first” (Bronskill 2014).16 Industry Canada responded to this private-sector initiative by producing guidelines for transparency reports (Industry Canada 2015), thereby contributing to their legitimization, and in the years that followed the majority of Canada’s large telecom companies publicly accounted for disclosures of personal information in some form. However, Canada’s largest intermediary (Bell) and many of its smaller competitors never issued transparency reports. Others have not always kept this information up-to-date, nor has there been any standardization of reporting across the industry. Companies can be selective in how they report on privacy governance; rather than levelling the information asymmetry between users and isps and enabling comparisons across the industry, transparency reports can be used to enhance an organization’s reputation or public image (Parsons 2017). While transparency reporting has been an inconsistent trend, it demonstrates how privacy accountability is now reputation-enhancing enough for some intermediaries to pursue beyond their legal obligations. THE FUTURE OF PRIVACY CUSTODIANS

Since this chapter argues that Canadian intermediaries have played an increasingly positive role as privacy custodians, an important question is whether we can expect this to continue into the future. While the preceding sections show an increase in the number of different ways that intermediaries have affirmed themselves as privacy custodians, any positive trend in corporate decision-making has not been a strong one. There are no telecom industry partnerships to promote privacy other than MediaSmarts, which produces publicly available educational resources on the topic. Companies that have gone to court to

140

Telecom Tensions

defend subscriber privacy have been exceptional cases rather than the norm, with inconsistency in copyright cases, and a generally cooperative relationship being maintained between isps and law enforcement. A company’s orientation to privacy interests in relation to other business pressures and legal concerns is something that can change on a case-by-case basis. While only some intermediaries have chosen to adopt a more positive orientation as privacy custodians, all have had to adapt to recent court decisions that have mandated stronger privacy protection for Canadians. The Spencer case joined a developing body of jurisprudence which recognizes that many kinds of “metadata” are personal information and therefore require privacy protection (Office of the Privacy Commissioner 2014). In 2016, a justice of the Ontario Superior Court ruled that Rogers and telus had “standing” to assert the privacy rights of their subscribers in opposition to a court order for a “tower-dump,” which would have disclosed information about tens of thousands of subscribers who had happened to be near one of several crime scenes. This meant that the companies could make arguments based on their customers’ privacy interests when opposing police demands to disclose information. The judge went as far as saying that Rogers and telus were “contractually obligated” to act in this way, because subscribers expected their personal information to be protected.17 For Michael Geist, this represented a legal “sea change,” affirming that companies have “a positive obligation to defend the privacy interests of their subscribers” (Geist 2016b). While these legal pressures can pull an intermediary in opposite directions (for and against information disclosure), an additional tension for privacy custodians arises from the business opportunity to collect and exploit personal information, which is becoming increasingly lucrative. In 2013, Bell announced that it was expanding its collection of personal information to include its subscribers’ internet browsing habits, as part of an effort to target personalized ads. Effectively, millions of wireless subscribers were told that information about their online activities was being monetized, in an effort by their provider to enhance its advertising business. Subscribers would benefit by being targeted with “more relevant” ads, or they could opt out by following Bell’s instructions, but their browsing would continue to be surveiled whether they liked it or not. After receiving “an unprecedented number of public complaints,” the Office of the Privacy Commissioner (2015b) investigated and found that Bell had not obtained adequate

Internet Service Providers as Privacy Custodians

141

consent from its customers to include them in the program. The company cancelled data collection following the Privacy Commissioner’s report, but has faced a class-action lawsuit by its customers alleging that their privacy was violated (Boutilier 2019). In recent years Bell expanded a similar marketing program with an opt-in approach to consent, and has been exploring other ways to monetize data about its subscribers for targeted advertising (Wire Report 2019a). Bell is not the only Canadian telecom company interested in monetizing personal information. In 2018 there were some concerns about how EnStream, a joint venture between Canada’s largest wireless companies, has been selling access to subscriber location data (Behar 2018). EnStream’s business is based on obtaining expressed consent from phone subscribers for this service, and there are no allegations that the service has been provided illegally. However, the potential for abuse in the “location-as-a-service” industry is evident in the US, where the fcc recently took action against the country’s major wireless companies for unlawfully fuelling a market in location data (Valentino-DeVries 2020). As in Canada, US telecom companies are legally responsible for protecting the privacy of their subscribers’ information, and are also responsible for ensuring a subscriber’s consent is given before their location information is sold. Subscribers have little choice but to trust that their intermediary is acting responsibly as a privacy custodian, or to trust that a regulator is ensuring legal compliance. A role conflict can occur when intermediaries play a double role by providing subscribers with connectivity and privacy, while providing another set of customers a service that depends on user data. These roles can largely be reconciled with legal obligations by satisfying the requirement for expressed consent from subscribers, but an intermediary that profits from what it knows about individuals has a qualitatively different relationship with subscribers than an intermediary that limits any non-essential collection and disclosure. US telecom companies were found to have chosen to profit from personal information instead of safeguarding the privacy of their subscribers (fcc 2020), and Canadian regulators will need to guard against the same possibility. The future outlook for intermediaries as privacy custodians is therefore quite mixed. The Spencer decision has restricted voluntary disclosures of personal information to government agencies, and with the exception of obligations for companies experiencing a data breach (Lithwick 2014a), new positive responsibilities have not been im-

142

Telecom Tensions

posed through law. Nevertheless, a number of intermediaries have adopted a more positive orientation to privacy of their own accord, by asserting the privacy rights of customers in court cases, by advocating for privacy interests in a public consultation, by issuing transparency reports, and by educating the public about the importance of privacy protection. However, such efforts have been wildly inconsistent across the industry, with organizations choosing to exceed minimum privacy obligations only in particular ways. Therefore, we should expect to see privacy custodians continuing to differentiate themselves on the basis of their positive conduct, but doing so in a selective manner. Future privacy scandals may lead to more comprehensive approaches by privacy custodians to meet consumer expectations.18 In the immediate future, we should be alert to how intermediaries respond to continuing pressures to monetize private information. Finally, the old tensions around lawful access remain, and battle lines around the issue may be drawn again if advocates are able to successfully move it back onto the government’s policy agenda (Parsons 2015). CONCLUSION

Intermediaries now occupy a pivotal role as privacy custodians, but this is by no means a culmination of the hundred-year history outlined earlier in this book. Public policies to safeguard privacy took time to develop, and did so alongside contradictory needs to monitor conversations, policies to enable state surveillance, and later in relation to copyright surveillance. Relevant legal regimes have sometimes imposed clear obligations and prohibitions, but intermediaries typically exercised a great deal of discretion in determining the nature of their conduct as privacy custodians. This meant making choices about the extent to which information could be collected, used, and disclosed to other parties. Intermediaries are more than rule-bound organizations complying with (or violating) their obligations as privacy custodians. Rather, intermediaries are collective actors that develop their own rationales for what it means to operate as privacy custodians, defining and redefining this role, while resolving the role conflicts that arise when they are subject to contradictory sets of expectations. Through their interaction with government agencies – in government consultations, through collaborative alliances like the ccaice, and in court or regulatory battles – isps have played an important part in determining what privacy governance means for individual users.

Internet Service Providers as Privacy Custodians

143

This chapter has argued that the responsibilities of privacy custodians can be differentiated as positive and negative expectations. Negative responsibilities, or prohibitions against behaviours such as information-sharing, also typically include exceptions that have been at the centre of role conflicts, public debates, and court cases involving isps. These exceptions include pipeda’s “lawful authority” section, which was used for nearly a decade to allow information disclosures to police. Court orders in copyright cases are another kind of exceptional circumstance. Finally, individuals can consent to their information being disclosed, but Bell’s 2013 advertising program violated a privacy custodian’s negative responsibilities, since the form of consent that the company relied upon was inadequate. While privacy custodians have often taken advantage of exceptions to their negative responsibilities, some organizations have been developing new positive responsibilities related to privacy. These include asserting the privacy rights of users in public (in court proceedings and a government consultation), educating the public, and being more open about how they govern privacy, such as through transparency reports. Intermediaries raising the bar through positive responsibilities has been a welcome development, and an important counterbalance to forces pushing in the opposite direction. There is no neat distinction between privacy custodians that take a positive orientation and those that meet the bare minimum of their obligations, but a more complicated array of possibilities in which each isp decides how far to go in pursuing each set of privacy responsibilities. The same organizations that champion privacy through one set of practices might collect, use, and disclose personal information through others, selectively publicizing only those actions that enhance their reputations as privacy custodians. This is the practical rationality through which privacy custodians navigate role conflicts and present themselves as responsible and trustworthy. Consumers trying to choose between isps on the basis of their decisions as privacy custodians are limited by what these organizations decide to disclose about their practices, and by customers’ ability to choose in a concentrated market (Clement and Obar 2016). In an era of growing concern over personal privacy (Office of the Privacy Commissioner 2017), amidst stories of data breaches, surveillance, and behavioural manipulation, the conduct of privacy custodians is of significant public importance. Whether we approach these topics as consumers trying to make an informed choice between

144

Telecom Tensions

intermediaries, or if we are concerned by the larger societal implications, it is not enough to hope that privacy custodians will act as good stewards of their own accord. While there are now compelling reasons for intermediaries to present themselves as being responsible, trustworthy, and even trailblazing when it comes to governing privacy, we are still largely dependent on what these organization tell us about their practices. Intractable role conflicts, where the expectations of a privacy custodian and state surveillance partner must be “balanced,” are too important to be left to an intermediary’s discretion about openness or transparency. Other role conflicts, which intermediaries enter into voluntarily, as they do when monetizing personal information, can pose unnecessary risks to users. isps now have critical responsibilities for both privacy and public safety, but they need not further complicate the picture through other profit-driven role conflicts. Ultimately, these are matters of public policy that cannot be decided through private relationships between users and their intermediaries, or complaints-based regulators.19 Whatever shape the policy debates of the future take, it is clear that intermediaries will also have their say, but less clear whether they will attempt to speak on behalf of users’ privacy interests. As the lawful access battle over Bill C-30 demonstrated, it will be necessary for public voices to articulate their own demands.

7

Internet Service Providers as Security Partners

The more vital that internet connectivity becomes, the more it is used for both licit and illicit activity, and the greater the harms associated with online threats. This chapter takes up the problem that the new “environment for innovation and social interaction has also become a vehicle for those who would do us harm” (Public Safety Canada 2016, 93). As Canadian society moved online, state agencies became concerned over what they saw as a lack of law and order. The drive to secure Canadian society in this new milieu sometimes led state agencies to assert sovereignty and compel intermediaries towards surveillance and security responsibilities. However, state agencies more often attempt to work through voluntary cooperation with private industry, particularly when governing the emerging domains of cyber security. The previous chapter analyzed the role that intermediaries play as privacy custodians, which developed alongside government requirements for intermediaries to serve as surveillance partners. Privacy is also implicated in a developing set of expectations around cyber security. This chapter will address these security responsibilities in detail, which are becoming increasingly important, albeit in tension with intermediaries’ evolving privacy expectations. Both privacy and security are domains of governance where intermediaries exercise considerable discretion. However, unlike privacy, security has been the aim of considerable efforts by the federal government to standardize intermediaries’ conduct and expectations. There are two types of security programs that have had dramatic implications for Canadian intermediaries in recent years: lawful access and cyber security. As described in the previous chapter, lawful

146

Telecom Tensions

access refers to requirements for intermediaries to assist state actors in obtaining information or carrying out surveillance against persons. Cyber security refers to efforts to protect both networks and persons against various kinds of online threats, and to govern data flows so as to control “malicious” traffic. In both lawful access and cyber security, government efforts have sought to achieve uniform standards of conduct for the hundreds of diverse isps operating in Canada following liberalization. For lawful access, this has included longstanding voluntary cooperation between industry and state agencies to fight internet child exploitation, and efforts to mandate lawful access compliance through legislation and regulation. Cyber security has taken longer to develop as a government concern in Canadian telecom, and has so far been addressed mainly through collaborative fora and confidential relationships. These relationships include secret programs disclosed by Edward Snowden, while publicly acknowledged (albeit little-discussed) “superstructural nodes” (Burris, Drahos, and Shearing 2005) operate to connect government and telecom representatives, to coordinate responses to specific incidents, or to work towards longterm strategic objectives. No topic in this book better exemplifies the conceptual emphasis of “governance” described in the Introduction than cyber security, with its various networked associations of state and non-state actors working to shape conduct across various institutional sites (Public Safety Canada 2017). While the developments described in this chapter can be understood as an expansion of security and surveillance responsibilities, the more specific ambition is standardization – creating uniformity across time and space through agreed-upon rules, allowing diverse actors to “work together over distance” (Bowker and Star 1999, 14), and facilitating the coordination of heterogeneous nodes towards some public policy objective. Often, such standardization is less about imposing a new standard for all to meet, and more about bringing some actors into line with a standard already met by others. As discussed in the previous chapter, privacy and security are deeply interwoven objectives in Canadian telecom, making it impossible to consider one without the other. Contradictory responsibilities of isps towards these two ends create role conflicts that intermediaries have a great deal of discretion in resolving. Additionally, some cyber security responsibilities are in conflict with net neutrality expectations. Standardization can be pursued to resolve or stabilize these tensions, but any attempt to legislate compliance risks turning role conflicts

Internet Service Providers as Security Partners

147

into political debates. As a result, there has been a proliferation of voluntary government-industry partnerships and telecom security standards, but Canadian isps’ roles as security and surveillance partners are far from standardized. This chapter makes a chronological argument, documenting how security concerns have expanded from individual subscribers and users (lawful access) to governing internet traffic (cyber security). Initially, widespread internet adoption led state agencies to demand surveillance access to packet-switched networks through analogies to circuit-switched telephone networks. Lawful access standards also had to deal with the internet-era problem of identifying individuals through digital traces – particularly on the basis of ip addresses. As lawful access standards were being developed collaboratively with the telecom industry, and alongside several attempts to legislate and regulate lawful access, isps increasingly began governing internet traffic to meet their own security needs and the needs of some clients. Lawful access ultimately targets the body, attempting to uncover the person behind an ip address, or collecting information about individuals. In contrast, cyber security deals with traffic and its circulation. Theoretical discussions of surveillance often reference Bentham’s Panopticon and Foucault’s Discipline and Punish (1995), wherein the sense of being watched produces a disciplined individual. But as numerous scholars of contemporary surveillance have argued (Haggerty 2006), the idea that surveillance has a disciplinary power over individuals can only take us so far in a world where surveillance is often hidden, or may not target individuals at all. The more helpful theoretical insight comes from the shift from discipline to security in Foucault’s later lectures, with security being concerned with “organizing circulation, eliminating its dangerous elements, making a division between good and bad circulation, and maximizing the good circulation by diminishing the bad” (Foucault 2007, 18). As the diversity, scale, and severity of cyber security threats have increased, standardization has been pursued as one solution to the problem of “bad circulation” (typically referred to as “malicious traffic”). While lawful access standards meet state surveillance needs, directed at individual users or subscribers, cyber security standards address how isps are surveilling their networks and responding to threatening or unusual traffic. These projects are by no means independent, and lawful access is sometimes now listed under the broader category of cyber security. However, the important development has been isps’ involvement in

148

Telecom Tensions

monitoring and discriminating amongst internet traffic, and the conflict this produces with the expectation that isps remain neutral, act as “mere conduits,” or serve as “dumb pipes.” LAWFUL ACCESS : INTERCEPTION AND DISCLOSURE RESPONSIBILITIES

The standardization of lawful access was achieved relatively late in the history of telephone networks, but the emergence of isps brought a new set of challenges. As previously discussed, electronic surveillance in Canada had gone from being largely unregulated before 1974, to a formalized regime of legal authorization and criminal penalties. But according to police and government spokespersons in the early 2000s, internet technologies left this legal regime in the dust. The law would have to “catch up,” or police would “lose” key investigative tools. Lawful access was not promoted as a new investigative capability enabled by digital networks, but instead as a vital tool that police had come to depend on in the telephone era, and which was now under threat from new technologies and new kinds of intermediaries. Lawful access advocates claimed they were attempting to “maintain” their traditional powers during these times of change (Government of Canada 2002, 5). To this end, isps were asked to accept new requirements to assist law enforcement and national security agencies, and these requirements would be standardized across an industry that now included a very diverse cast of companies. Lawful access was meant to turn these intermediaries into a consistent and dependable set of instruments in the fight against crime and national security threats. Beginning in the early 2000s, efforts to develop or “update” lawful access in Canada passed through several stages and multiple different versions of legislation, informed by numerous stakeholder consultations, behind-the-scenes meetings, and public debates. For much of this period, lawful access requirements encompassed two kinds of standardization and their attendant responsibilities. The first was a “design standard” (Timmermans and Epstein 2010) specifying the surveillance capabilities of telecom networks. Sometimes called the “interception component” of lawful access, these standards would require intermediaries to have surveillance capabilities built into their networks. Without this power, police or csis might find that an intermediary did not have the capability to wiretap a subscriber, even under court order.1

Internet Service Providers as Security Partners

149

The second set of lawful access responsibilities was a “procedural standard” (Timmermans and Epstein 2010), which would impose uniformity in how intermediaries disclosed information about subscribers when requested by state agencies. Rather than a way of monitoring individuals’ communications, this requirement was useful primarily in identifying individuals through the use of an isp’s records, as discussed in the previous chapter. While some incumbent isps had routinized the procedure of identifying subscribers for law enforcement, police complained about the lack of consistency across the entire industry and the ability of some isps to refuse to cooperate (Public Safety Canada 2012). Lawful access legislation was promoted to transform isps’ voluntary cooperation into a mandatory, uniform procedure for information disclosure. Government-Industry Collaboration on Lawful Access Intermediaries have not been passive targets of proposed lawful access legislation, raising various concerns and playing different roles throughout the lawful access saga (Parsons 2015). Many companies worried about the consequences of an expanded set of responsibilities for isps, and some became actively involved in meetings to craft lawful access policies. Industry concerns over lawful access included safeguarding subscriber privacy and the fear of being seen as agents of the state, but companies have been primarily interested in determining the details of what would be required of them, how easily they could meet these new obligations, and the costs of complying.2 Government-industry collaboration gave some intermediaries privileged access to information about the government’s plans and a channel for communicating their views, creating what Geist (2012b, n.p.) called a “two-tier” policy approach, where “secrecy and backroom industry talks” shaped surveillance policies before these were announced to the broader public. But while the relationship between Canadian intermediaries and government never turned adversarial during the development of lawful access policies, the telecom industry was also never a full partner in the process, and its concerns were not necessarily addressed by government. Federal governments were inconsistent in seeking industry input on various proposals as lawful access legislation was being drafted, leaving some intermediaries to argue that the industry was not being adequately consulted (csis 2012, 1). While extensive discussions and meetings between industry

150

Telecom Tensions

and government took place behind closed doors (Geist 2012b; Paperny 2012), not all intermediaries were equally represented in government consultations, with incumbents having the greatest involvement (Industry Canada 2014a; Public Safety Canada 2012). As different versions of lawful access legislation were being drafted and discussed, voluntary cooperation and collaboration continued under the existing legal regime, such as ccaice’s bsi disclosure protocol (see chapter 6). By and large, incumbents cooperated with police and maintained the surveillance capabilities the government sought to mandate. However, consistency remained a problem and even a single uncooperative isp could limit the reach of state agencies. Forging Consistency among Privacy Custodians While an uncooperative intermediary could simply be forced to disclose information with a court order, lawful access proponents argued that obtaining court orders for the sorts of information routinely requested by police would be either too onerous or impossible.3 Furthermore, a voluntary system, without a procedural standard or a universal set of obligations, created “an inconsistent, ad-hoc approach” (Public Safety Canada 2012, 143). Therefore, the policy solution pursued through successive iterations of lawful access legislation was to make the disclosure of subscriber information mandatory. Instead of allowing intermediaries to define their roles as privacy custodians under the flexibility of pipeda and their company privacy policies, lawful access advocates sought to impose consistent obligations throughout the telecom industry. The problem of inconsistency among privacy custodians in the telecom industry was often presented as a consequence of liberalization, which had resulted in a diversity of intermediaries of various sizes and with differing approaches to dealing with law enforcement. Some iisps operated with limited resources that made compliance with lawful access obligations an onerous burden, while other small intermediaries were alleged to be actively facilitating illegal activity, existing “solely to trade in child sexual abuse” (Bailey 2007). The lack of cooperation by some small isps and the fear that that these intermediaries would become safe havens for criminal activity were recurrently raised by lawful access proponents, while incumbents were generally recognized to be cooperative with police investigations. Incumbent isps already participated in ccaice, had procedures in place for handling

Internet Service Providers as Security Partners

151

lawful access requests, and maintained dedicated staff4 for these purposes. iisps, on the other hand, often had little or no experience assisting police investigations, and faced the greatest financial impact in installing and maintaining interception equipment (Arellano 2011). Lawful access legislation was intended to standardize responsibilities that most of the isp industry had already accepted, since incumbents generally already maintained wiretap capabilities and disclosed subscriber information on request. However, it would not be fair to say that small isps were the true targets of lawful access proposals, since incumbents would also be forced to modify their disclosure practices, or face clear penalties for failing to meet certain standards. Before liberalization, police could depend on a relationship with a single telecom provider in a given area, but some police departments were now paying large monthly bills to multiple incumbents for surveillance services that often had no regulated rates.5 Lawful access proponents were therefore keen to subject all intermediaries to a common regime, and isps of all sizes were acutely interested how lawful access would affect the costs of doing business, and what changes to their operations would be required. Even where proposed obligations were already largely being met by intermediaries through voluntary cooperation, new legislation would have the effect of standardizing methods, procedures, and equipment, along with the monitoring and enforcement of industry’s compliance.6 Standardization through Jurisprudence Successive federal governments failed to enact dedicated lawful access legislation during the period between 2000 and 2015 due to a lack of commitment, the contentiousness of surveillance and privacy issues, and (particularly during the Bill C-30 controversy in 2012) the inability of the federal government to make a persuasive case for lawful access (Parsons 2015; Whitaker 2012). Ultimately, greater consistency in the handling of warrantless requests for information would come about not through legislation to mandate cooperation, but through a court decision to prohibit this sort of “invisible handshake” (Birnhack and Elkin-Koren 2003) between the police and telecom companies. The 2014 Supreme Court of Canada’s unanimous decision in R v Spencer (2014) ruled that the voluntary disclosure of subscriber information by an isp in a Saskatchewan child pornography case could not be justified by pipeda (Lithwick 2014b). In the absence of a new legal

152

Telecom Tensions

regime for warrantless disclosures, the Spencer decision still remained open to some competing interpretations. As with the tower-dump decision discussed in the previous chapter,7 intermediaries were left in the position of assessing the constitutionality of legal demands before agreeing to comply and determining what kinds of disclosures violated their users’“reasonable expectation of privacy” (Therrien 2015). However, a clear implication of the Spencer decision was that many of the millions of voluntary disclosures by intermediaries since the early 2000s (see Office of the Privacy Commissioner 2015a) had been unconstitutional. According to the Supreme Court, such disclosures were in fact prohibited by pipeda, rather than being enabled by the law’s exceptions in section 7(3)(c.1). As a result, the routine disclosure of personal information by intermediaries to police underwent a massive change during the summer of 2014, with companies largely ceasing to comply with non-emergency “pre-warrant” requests. The police response to this new standard of privacy protection was to argue that their jobs had become more difficult, and as a result society had become less safe.8 The post-Spencer requirement for police to obtain a court order for subscriber information certainly means that the process takes more time and effort, and is limited to situations that satisfy a judge. There have since been renewed calls by police advocates for warrantless access to subscriber information (Geist 2017), but as things stand, intermediaries’ responsibilities as privacy custodians trump the discretion they once exercised as surveillance collaborators. To summarize the above, lawful access legislation was an attempt to standardize isps’ conduct in a way that resolved the role conflict discussed in the previous chapter, but the most powerful means of shaping lawful access have been collaborative fora and court decisions. In contrast, the federal government has largely avoided the legislative route in its work to standardize cyber security,9 where court challenges have also been less likely to arise. This is because the primary security and surveillance actors are isps themselves, eliminating the possibility of conflict between industry privacy custodians and state agencies. Additionally, while lawful access targets individuals who may be prosecuted, cyber security generally targets internet traffic through technical means. If you are an individual facing the prospect of a prison sentence, you might have good reason to challenge the laws that justified a police investigation. But a data packet that has been labelled as malicious traffic and “blackholed” by an isp’s filtering equipment will never have its day in court.

Internet Service Providers as Security Partners

153

THE RISE OF CYBER SECURIT Y

On the one hand, developments in both lawful access and cyber security are part of a gradually unfolding “securitization” of the internet (Barnard-Wills and Ashenden 2012; Hansen and Nissenbaum 2009; Zajko 2015a), wherein state agencies are increasingly asserting themselves through internet policy and requiring that intermediaries deal with particular threats. But cyber security in particular also encompasses a broader set of concerns that echo an earlier conception of security, rooted in European urbanization and the problem of managing the circulation of a town’s “mobile elements” (Foucault 2007, 20). Whereas eighteenth-century planners proposed ways to arrange the movement of people, animals, and goods through the town, while maintaining surveillance and hygiene within the city walls (Foucault 2007, 20–1), today’s concerns are with positioning “cyber sensors” in telecom infrastructure and maintaining network hygiene (csec 2011). Security was a major preoccupation for eighteenth-century political theorists, as a set of expectations that served as a foundation for society. These expectations came to be associated with the protection of private property and the smooth operation of the market mechanism underpinning the liberal order (Neocleous 2008). Twentiethcentury notions of “national security,” developed during the Cold War, seemed to prioritize the protection of the state and its interests. However, in Canada as in the United States, the security of the economy and that of the state were inseparable, particularly given the threat posed to both by communism. While global capitalism no longer faces such an existential threat, national security remains closely tied to safeguarding economic assets. As described in chapter 2, Canada’s federal government and the crtc have often described the importance of internet connectivity by using the language of “the digital economy.” Canada’s assemblage of cyber security programs are rationalized as protecting the digital economy against its existential threats – attacks on “critical infrastructure” and strategic hacking by foreign states (see Zajko 2015a). Much of what currently falls under the heading of cyber security has long been addressed (particularly by the private sector) as information, computer, and network security. However, cyber security encompasses a broader set of concerns, and its discourse operates by tying these diverse elements together. The project of cyber security ties personal responsibility to national interests, private partners to

154

Telecom Tensions

state agencies, and addresses a growing array of threats and social problems. In the process, information security becomes linked with collective referent objects, such as “society,” “the nation,” and “the economy” (Hansen and Nissenbaum 2009). Rather than simply privileging the security of the nation-state, cyber security is a collective endeavour and a unifying project.10 While lawful access is largely concerned with the problem of identifying, locating, and monitoring individuals (in “meatspace”) on the basis of digital records, cyber security typically operates at the level of internet traffic – detecting, filtering, rerouting, and manipulating data packets. Lawful access serves state surveillance needs, particularly for crime control. Cyber security, on the other hand, is divided amongst referents: it encompasses both state and private-sector strategies to secure networks against network-scale threats, as well as programs to protect individual users of the network from individually targeted threats. In addition to technical interventions, government agencies and private industry have implemented educational programs to promote safer conduct among users. In other words, cyber security covers an expansive and growing set of concerns. While the pivotal internet surveillance debates in Canada have thus far concerned lawful access, the critical issues of the future pertain to the larger category of cyber security. Cyber security is the advancing frontier of intermediary governance over data flows, potentially transforming isps away from the ideal of a “dumb pipe” (that simply carries traffic) and towards intelligent networks that maintain network hygiene and protect against cyber threats. In addition to its traditionally defensive focus on protecting information and networks, the pursuit of cyber security has expanded to include moral and legal conduct by internet users, for instance in regard to cyberbullying (Government of Canada 2018), as well as an offensive orientation against adversaries and intelligence targets (Wark 2016). Finally, cyber security now also encompasses lawful access, and its attendant political debates about privacy and surveillance. isps, as privacy custodians and gatekeepers of connectivity, are actively involved in shaping these developments. Intermediaries as Gatekeepers: From Child Pornography and Spam to Cyber Security A significant reason the last attempt to pass comprehensive lawful access legislation failed in 2012 was the response Public Safety Minis-

Internet Service Providers as Security Partners

155

ter Vic Toews received when he attempted to use the stigma against child abuse to gain support for the government’s bill. Instead of mobilizing support, he faced a public backlash from Canadians who were angry at his suggestion that opposing the bill meant “standing with” child pornographers (Public Safety Canada 2013b). While the strong response elicited by the topic of child pornography was badly managed by the government’s introduction of the Protecting Children from Internet Predators Act (C-30), the seriousness of the issue had already spurred Canadian isps into action in 2006. In addition to the ccaice protocol for disclosure of subscriber information discussed in the previous chapter, incumbents also agreed to block access to websites that had been identified as hosting child abuse imagery under ccaice’s Cleanfeed Canada initiative. This was justified as an attempt to “make the Internet safer for Canadians and their families by reducing their chances of accidentally coming across images of child sexual exploitation on the Internet” (Cybertip.ca 2006). In a country with an otherwise “open” internet, blocking access to web content was only acceptable given the nature of child abuse imagery. Through this limited action against the most universally despised kind of internet traffic, Canadian isps became gatekeepers rather than mere conduits, and faced no opposition in doing so. Milton Mueller calls the “child safety movement” the “vanguard of Internet content regulation” (2010, 190), with its rationale for website blocking based on the universal criminalization and condemnation of child pornography. However, programs such as Cleanfeed joined even earlier efforts by isps to use blocklists for filtering flows of malicious traffic that directly harmed users or affected how these networks operated. isps have long worked independently and collaboratively to block spam and phishing emails, and to filter traffic associated with malware or denial-of-service attacks (McTaggart 2006). Today, these traffic management processes continue to develop and some isps have become highly capable cyber security actors. Despite recurrent calls for intermediaries to expand their involvement in cyber security, role conflicts have limited the extent to which this transformation has taken hold across the industry. The role of traffic manager is not just an evolutionary step above the communications carrier; the orientation of one role contradicts that of the other. In becoming agents of cyber security, isps have been transforming themselves away from the model of a “dumb pipe” or mere conduit that simply carries traffic (thereby avoiding any responsibility for the

156

Telecom Tensions

contents of the pipe), and towards that of a “smart pipe” that detects and “cleans” internet traffic of threats. Some Canadian isps have long acted as security providers by offering such services to government and business clients (see Scalet 2007). However, extending such measures throughout the network moves intermediaries away from the ideal of net neutrality and the end-to-end principle, where discriminating between data is prohibited, and towards the routinization of discrimination. The role conflict this creates is between the common carrier or net neutrality expectation for isps to treat all traffic equally, and a smart pipe rationality where isps manage traffic according to how it is classified. Governing Cyber Security in Canada Mapping all of the different agents involved in Canada’s cyber security is beyond the scope of this chapter, but as in much of the rest of the world, securing telecom infrastructure and internet traffic is primarily the responsibility of the owners and operators of telecom networks. Officially, state agencies’ roles are to secure government networks, to help coordinate private actors, and to facilitate information-sharing. Since Canada’s Communications Security Establishment (cse) is a member of the international security alliance known as the Five Eyes, some materials relating to the agency’s secret conduct have also been revealed through Edward Snowden’s documents. These show examples of how cse gains access to internet traffic through “special source operations” (Wark 2016), involving one or more Canadian telecom industry partners. However, unlike the “special source” relationships in the United States (nsa 2009), these documents give us little insight into the basis of Canadian partnerships, so my analysis is based on the security partnerships with private industry that are openly acknowledged by the federal government. Recent years have seen a proliferation of industry-government cyber security associations in Canada, emphasizing various forms of information-sharing and standardization.11 These are what Burris et al. (2005) conceptualize as “superstructural nodes,” connecting “representatives of different nodal organizations … to concentrate the members’ resources and technologies for a common purpose but without integrating the various networks” (38). Canadian isps participate in several of these nodal assemblages, which operate horizontally at different

Internet Service Providers as Security Partners

157

levels of corporate hierarchy. What these organizations have in common is that while they either include or maintain relationships with police and national security agencies, none have the authority to compel isps to behave in certain ways. Instead, they rely on the voluntary cooperation of their members to address threats to a common interest. At the operational level, associations like the Canadian Telecommunications Cyber Protection Working Group (ctcp) and the Canadian Centre for Cyber Security (cccs) have had the task of responding to specific incidents and cyber threats, such as when a major attack threatens multiple networks in Canada, or when a foreign organization notices an attack that appears to be originating from within Canada. In addition to responding on an emergency basis by connecting security managers from competing firms with government cyber security staff, these organizations also carry out more long-term work by sharing information, analyzing threats, and standardizing processes. Here, their work is more likely to intersect with associations involving corporate executives and senior bureaucrats, such as the Canadian Security Telecommunications Advisory Committee (cstac), where information-sharing happens on a higher level and the emphasis is on strategic collaboration. cstac was intended to be a way for “industry to contribute to government policy,” but it was also imagined that government would use it as an “opportunity to contribute to industry network security policies and procedures” (Industry Canada 2014b, 1), such as through the development of industry standards. Another executive-level association, the Chief Executive Officers’ Advisory Committee on Cyber Security, was also established to provide the federal government with strategic advice. This group of ceos was instrumental in establishing the Canadian Cyber Threat Exchange (cctx), an information-sharing organization in which a vp from Bell has played a prominent role since 2015 (Public Safety Canada 2018). Much of what happens through these associations is beyond public view, or with carefully managed visibility.12 Documentation of their activities can be obtained through access-to-information requests, and while these files are often heavily redacted, they do provide a sense of what the day-to-day work of cyber security in Canada entails. Where cyber security implicates private industry, the federal government exercises limited sovereignty, and has been reluctant to assert its authority over how private networks secure digital flows.

158

Telecom Tensions

Canada’s Reluctant Cyber Sovereignty One of the internet’s myths is that it emerged thanks to private-sector innovation, with government staying out of the way (see Streeter 2011). In recent years, we are more likely to see the internet as an arena for international conflict and competition, state-sponsored hacking, and surveillance, as governments have learned to use it for their own purposes (see Morozov 2011). But Canada maintains a decidedly liberal approach, and state agencies have been reluctant to assert sovereignty over digital flows, even when addressing national security threats. In Canada, sovereign power in cyber security is largely concentrated in cse, a position likely to be reinforced through the sorts of reforms that were recently made to the agency’s mandate.13 However, cse’s digital interventions are typically secret, and do not impose private-sector responsibilities.14 Efforts to regulate cyber security by imposing requirements on telecom equipment vendors have been a more visible form of sovereign power, but regulations that would impose significant costs on private industry have either been a recent development, or floundered when confronted with industry resistance.15 cse’s public face is expressed through the ccss, a collaborative organization which can be used to channel information to intermediaries and respond to threats significant enough to impact government. In this and other public cyber security venues, the relationship between industry and the state is wrapped in the language of partnership and cooperation. For example, cstac was established in November 2010 as a “public/private collaboration” (Industry Canada 2014b, 15) and a “senior point of contact” between industry and government (89). It is co-chaired by one government and one industry representative, with private-sector membership composed of senior executives from telecom companies, and government representation including ised, cse, csis, and the rcmp. Modelled on similar partnership “committees” in the United States and United Kingdom, cstac was justified within government as a way to better understand telecom network vulnerabilities and the telecom industry’s ability to address various threats (Industry Canada 2014b). The impetus for its creation came from government, which had to convince telecom companies to participate, but these intermediaries had good reasons to influence the direction of cyber security policy, as cstac provided an opportunity to shape the standards that the private sector might be expected to adhere to.

Internet Service Providers as Security Partners

159

The first such project gave cstac members a chance to have a say on whether isps should have a standardized approach to “cleaning” internet traffic. Standardizing Network Hygiene In North America, ‘cleanliness’ and ‘hygiene’ have become common metaphors to describe isps’ expanded role in cyber security. One federal government document described maintaining network hygiene as “regularly performing the ‘bread and butter’ activities of network and information technology (it) security” (Public Safety Canada 2013a, 22), such as updating access privileges, applying security-related patches, operating intrusion detection systems, and educating users. While maintaining hygiene can refer to routine it security work, the term also has a more specific usage in regard to cleaning internet traffic. Mary Douglas’s (1966) explanation of the relationship between hygiene and social order is well-known, but this particular pursuit of social order – the “clean and safe” internet (Scalet 2007) – can only be achieved through continuous surveillance. This is because the best way to achieve “clean” pipes is for an isp to build “intelligence” into the network (making the pipes “smart”). This is to say, an isp needs to monitor internet traffic, discriminate between acceptable and undesirable traffic, and have a way of dealing with the latter. Ideally, these interventions are largely automated, with equipment inspecting and categorizing data packets or flows of traffic, applying set criteria to detect the malicious or unusual. Traffic flagged in this fashion can then receive special treatment from routers and servers, such as blocking, throttling, or blackholing. How isps practise cyber security has been left largely to their discretion, but as with lawful access, efforts have been underway to standardize the expectations of this role. As one of its first tasks in 2011, cstac went about establishing a “Cyber Security Standard for Telecommunications Service Providers” (Industry Canada 2014b). An initial formulation (at a time when Bell was cstac’s industry cochair) was dubbed the National Clean Pipe Standard (D. McMahon 2011). This was attempt to formalize and legitimate the role of isps as providers of network hygiene, which was in line with the approach Bell was already taking in regards to its network.16 Among the incumbents, Bell had an especially close connection to state cyber security

160

Telecom Tensions

efforts in this period, writing conceptual and technical reports for government, and analyzing flows of malicious traffic on its vast network to understand cyber threats (D. Hales 2014). The Clean Pipe Standard was based on work carried out under the Bell-led “DarkSpace” study, working in collaboration with government departments such as cse and the rcmp, and non-government partners including The Citizen Lab and Palantir (D. McMahon, Rohozinski, and Bell Canada 2013). It was imagined that isps that met the standard would provide a layer of security upstream from their enterprise customers, surveilling and filtering traffic, and responding against widespread botnets and denial-of-service attacks (Bell Canada 2010). Ultimately, the standard that emerged from the process (cstac 2013) was more general, did not include any certification or compliance mechanisms, lacked the language of clean pipes, and also avoided the word “standard.” Instead, it was renamed Security Best Practices – a change in terminology meant to better reflect its “voluntary nature” (Industry Canada 2014b, 136). As with lawful access, some cyber security standards govern the equipment used by isps, and others set procedural standards for how these institutions operate. cstac’s Best Practices deal mostly with the latter, but isps must operate specialized hardware and software to meet some of these procedural standards. For instance, isps should be able to monitor network traffic for “anomalies,” and have the capability to “throttle, filter, and block” categories of “malicious or inappropriate traffic.” Other Best Practices are strictly procedural and deal with social responsibilities of isps. These include participating in industry and government information-sharing processes, and notifying a subscriber upon learning that their computer has been the victim or perpetrator of an attack (cstac 2013). Cyber security now has an abundance of standards in addition to the above,17 but these have not yet produced much uniformity in the conduct of Canadian isps. Most recently, a review of telecom regulations recommended that the crtc update cstac’s “network security hygiene guidelines” (the Best Practices) and make them mandatory (btlr 2020) – a recommendation that has yet to be acted on. The push for certain kinds of cyber standards can tell us about the roles that are promoted for isps as security providers, or what is imagined to be their future. However, the limited success of these approaches speaks to the state’s reluctance to regulate in this domain, and the persistence of role conflicts that impede further efforts to “clean” the internet.

Internet Service Providers as Security Partners

161

Smart Pipes, Dumb Pipes, and Net Neutrality How would you like your internet served? Smart or dumb? Clean or dirty? In other words, would you like your isp to inspect and filter traffic coming to and from your device? Should an isp be able to block an app on your phone from communicating with a distant server if the activity seems suspicious? Should an isp be able to prevent you from viewing a webpage that might be harmful? Should an isp block an email from arriving in your inbox if it might be spam? These are all ways that an isp’s equipment, sitting between your device and the rest of the internet, can collect information about your communications and make decisions about how to handle them. They are also all questions that can touch off intense disagreements. The debate about whether isps should engage in these kinds of governance is the debate over smart pipes versus dumb pipes – a debate in which smart pipes create role conflicts with net neutrality and privacy custodian responsibilities. To avoid these conflicts, the dumb pipe remains the safer orientation, but smart pipes have repeatedly been promoted as inevitable or the key to the future. Tyson Macaulay, former Security Liaison Officer at Bell, stated that “the nightmare scenario for a carrier is to be a dumb pipe, and just carry the internet traffic, and add no value to it” (Macaulay 2015). While the dumb pipe is just a conduit for packets, the smart pipe is an actor – its intelligence is rooted in its ability to inspect and discriminate between traffic, to decide what should be permitted, prioritized, or cleaned. Whereas the dumb pipe refers to the mythical past of the end-to-end internet, the smart pipe belongs to the “digital future” (Pai 2017). As discussed in chapter 5, the practice of intermediaries granting “undue preference” to certain traffic for commercial reasons has received critical attention and regulatory disapproval in Canada, but the implementation of smart pipes has proceeded quietly in the interests of security, which is also the most accepted regulatory justification for Canadian isps engaging in “traffic management.”18 isps offer these security services to their enterprise and institutional customers, such as businesses and public-sector organizations who can configure these systems to their preferences (Scalet 2007). Network operators have an advantage over other security providers because of the view they enjoy into their entire network, allowing them to identify and predict threats on the basis of broader patterns of activity (Dickson 2015). Many experts argue that isps will have to play an active role in

162

Telecom Tensions

managing internet traffic to succeed in a future of the “internet of things” and next-generation 5G wireless technologies (Hecht 2015; Macaulay 2017). And yet, the broader predictions of smart pipe advocates remain largely unfulfilled, and other commentators argue that the future for isps will be to function as utilities, with some going as far as to embrace the “dumb pipe” metaphor.19 Even in the absence of legal considerations, the debate over dumb pipes or smart pipes as a viable business model in telecom is far from settled. Once we include regulatory regimes in our consideration, we can see that efforts by network operators to surveil and govern traffic in circulation are limited because they have not been mandated by government, and are either in tension or in conflict with the roles of common carrier and privacy custodian. From a regulatory perspective, there is nothing wrong with isps carrying out surveillance or filtering to meet a subscriber’s expectations. If a business or a school board wants their isp to filter certain kinds of traffic, whether for reasons of safety, confidentiality, or productivity, then that is their choice. It is when such interventions are extended to “the general public” that they potentially conflict with isps’ responsibilities as privacy custodians and with public expectations of net neutrality (Standing Senate Committee on Legal and Constitutional Affairs 2009). As far as the general public is concerned, isps predominantly remain dumb pipes, and much of the tension between smart and dumb pipes will not dissipate through technological progress, because it is not possible to innovate a way out of a role conflict. In the long term, isps are becoming ever more active agents, governing connectivity toward an expanding list of ends, and developing more sophisticated security offerings for particular clients. It seems likely that stronger cyber security regulations will eventually be imposed, perhaps as a consequence of a major attack or an especially consequential data breach. In the short term, without regulations compelling intermediaries to adopt a proactive posture against threats, Canadian isps manage traffic in circumscribed ways. Acting as a mere conduit is often the easiest choice, with more proactive kinds of security offered as a “value-added” service for those willing to pay.

Internet Service Providers as Security Partners

163

STATE PARTNERS , SECURIT Y PROVIDERS , AND THE STRUGGLE FOR STANDARDS

While the previous chapter addressed the discretion isps enjoy as privacy custodians, and how different organizations choose to pursue these responsibilities, this chapter dealt with efforts to promote the standardization of isps’ security responsibilities as a way of limiting discretion. For lawful access, standardization has been pursued through legislation that would mandate certain practices and surveillance equipment, while cyber security has thus far been addressed through voluntary standards and government-industry partnerships or superstructural nodes. Lawful access and cyber security can be thought of as routine operational practices for isps, which have historically partnered and collaborated with state agencies, and cooperated within the industry in pursuit of a common approach. Canadian isps generally recognize security as part of their public obligations, whether this means assisting law enforcement or countering network attacks, and they perform a great deal of this work beyond their legal obligations. Unlike regulation-for-competition, in which a regulator structures market relationships that would not otherwise exist, intermediaries often govern security of their own accord in ways that complement the interests of state agencies. But this has not been a harmonious realization of regulatory capitalism, since organizations exercise significant discretion in meeting their security responsibilities and conduct across the industry is far from uniform. Additionally, isps are forced to mediate when state and customer interests contradict each other, such as when police request customer information. As a result, security regulation has been an ongoing governmental priority since the start of the 2000s, leading to recurring attempts to legislate lawful access standards. However, Canadian regulators have shown a reluctance to impose standards for the dynamic and expanding set of techniques that govern cyber security, in favour of a more collaborative approach. Part of the reason for this reluctance is that governing cyber security means governing some of the fundamental ways through which the internet operates. Efforts to standardize lawful access reflect the persistent need of state agencies to identify individuals through digital traces and to intercept communications, both of which depend on partnering with

164

Telecom Tensions

intermediaries whose infrastructure makes communications possible. While the basic surveillance responsibilities of lawful access are outgrowths of telephone-era requirements, cyber security represents a much broader, future-oriented kind of rationality, in which isps govern internet traffic in the interests of security. Underlying the shift to cyber security is a fundamental transformation in the role of the isp as a governing actor – a transformation that has been underway since the early days of the internet’s commercialization and which deviates from what is commonly given as the internet’s founding ethos. This is the transformation away from the ideal of the end-to-end network, where isps merely act as conduits or dumb pipes, and toward the isp as a smart pipe that surveils and discriminates between data packets. The adoption of this new role and its accompanying rationality has proceeded with little controversy by targeting universally despised forms of traffic such as child abuse images, spam, and denial-of-service attacks, thereby eliding the net neutrality concerns that arise when surveillance and discrimination are used for other purposes. This transformation is indicative of the active role intermediaries now play in governing digital flows, the growing value of connectivity, and the corresponding increase in the harms caused by threats circulating through our digital networks. The limits of this transformation draw our attention to some fundamental disagreements and conflicts over the internet’s future.

Conclusion

Around the world, isps are implicated in a growing range of social relationships. In Canada, this trend also reflects all of the ways that market forces alone cannot meet public policy objectives, such as equitably distributing connectivity, protecting privacy, or maintaining public safety. Furthermore, under regulatory capitalism, the rationality of regulation-for-competition is based on the idea that the market itself is a problem that needs to be governed, by imposing obligations on isps to relate to each other in certain ways. As pure intermediaries, isps sit passively between the “endpoints” of our online interactions. Their responsibility is to enable our relationships over their networks. Since isps can be more than “mere conduits” or “dumb pipes,” these institutions also become actively involved in the flows passing through them. They inspect, discriminate, and police the connective fabric of society, either to meet requirements imposed by other actors or for their own reasons. Many of the social roles played by these intermediaries involve enacting distinct forms of rationality, such as the regulatory distinctions and justifications underpinning the mandated access regime, or the process of surveillance and classification that enables technical interventions against malicious traffic. Regardless of the specific role, all of these are instances of intermediation – a process that installs new intermediaries in social roles, or expands the roles of existing intermediaries. This is important to consider, because the internet has often been described as a disintermediating force that cuts out intermediaries. Intermediation and disintermediation are actually both occurring simultaneously, but it is important to reflect on how these major social shifts relate to each other.

166

Telecom Tensions

Part of the internet’s foundational end-to-end principle was that users at the edges could connect by using a shared protocol, and intermediate networks would simply pass these communications to their intended recipients. In this idealized model of the internet, isps effectively function as mere conduits, active only in making routing decisions. Connectivity removes the need for many kinds of pre-internet gatekeepers and go-betweens, such as retailers, distributors, and publishers, and reduces the ability of traditionally dominant institutions to control information flows. Hierarchies are flattened or collapsed, and for practical purposes, the result is an unmediated connection between the internet’s endpoints – a connection managed by isps. If disintermediation is “the movement of power out of the middle of the net” (Moglen 2012), then intermediation is a word we can use to describe the growing power of the “middle.” If disintermediation means dismantling hierarchies and moving to peer relationships, intermediation is a process that installs new “feudal lords” (Schneier 2012) and governs through these intermediaries (DeNardis 2014). Where previously each endpoint on the internet was reachable from any other, communications must now contend with intermediaries that may prove impassable. Relationships that used to be between two parties must now gain the consent of a third, and this third party is increasingly seen as a point of control where additional responsibilities can be applied. Online intermediaries such as Facebook are critically important, in large part because they sit atop vast quantities of personal data. But ultimately these are just services, and a world without Facebook is not difficult to imagine. Many online intermediaries provide services that might conceivably be handled by endpoints rather than in the “middle of the net,” but the same cannot be said for the isps that act as the internet’s ultimate go-betweens. isps cannot be disintermediated, since they provide the internet’s connective infrastructure (Kerr and Gilbert 2004). The only meaningful alternatives to isps, such as decentralized mesh networks, remain limited to localized contexts. There is no question that, for the foreseeable future, isps will continue to occupy their central position. The important question is what kind of intermediaries these isps will be. I have discussed isps as governing nodes in relation to privacy, copyright, security, and net neutrality, but this list is by no means exhaustive and seems set to expand. The trend has been for isps to provide more services, to take on greater public policy responsibilities, and to engage in more active

Conclusion

167

interventions in traffic flows. These include inspecting and blocking malicious traffic, prioritizing some internet protocols over others (Bendrath and Mueller 2011), and being more involved in managing subscribers’ ip addresses (Zhang 2007). The role conflicts discussed in the previous two chapters have impeded some of these developments, but traffic management is just one aspect of intermediation. As points of control over telecom infrastructure, isps are instruments of public policies to promote competitiveness in the telecom industry and connectivity throughout society. Even dumb pipes are expected to do more for the public good. Another way to characterize intermediation is as an accumulation of social roles for intermediaries. Some of this is due to our growing expectations of these institutions, but many isps are also interested in expanding their services. As intermediaries take on a growing number of important roles in society, role conflicts accumulate and become more significant. Any attempt to provide connectivity as a public good runs against the expectations of the liberalized telecom industry. Incumbents must simultaneously compete and cooperate with iisps. Government agencies expect isps to act as partners in surveillance and security, but these expectations can conflict with the duties of privacy custodians. New kinds of traffic discrimination and manipulation challenge expectations of net neutrality. isps exercise considerable discretion in addressing many of these conflicts, favouring some roles and expectations over others. Organizations that see isps as instruments of public policy have their own ways of resolving contradictory expectations, such as by forming partnerships with industry or mandating standardization. Some of the role conflicts listed above and their potential solutions have been among the most contentious public policy debates in recent Canadian history. Despite dramatic reversals on ubb and lawful access policy following public outcry, the fundamental conflicts at the heart of these issues remain and will resurface in new contests over what we expect from our intermediaries. ADDRESSING INTERMEDIATION

In summary, the internet has covered society with a dense connective fabric. This has allowed new kinds of relationships to be formed, some of which cut out previously installed intermediaries. Disruption and “creative destruction” became mantras as we did away with old

168

Telecom Tensions

and inefficient models in favour of decentralization and direct connections. But the connective fabric that covers the planet allows for new classes of intermediaries to manage our relationships – exchanging services for access to our data flows, providing security and convenience, and enforcing appropriate behaviour. This trend extends to the isps that own and operate telecom infrastructure, and that now contend with a growing list of responsibilities. The growing concentration of power in the middle of our networks can lead isps to exploit their nodal position, and reduces autonomy at the edges. An obvious solution to this is to stick to the internet’s “first principle” (Carpenter 1996) – namely, end-to-end architecture. As net neutrality advocates argue, this would mean privileging endpoints, limiting intermediary functions, and enforcing constraints on intermediaries’ discretion, so that control over a nodal position only confers limited power over data flows. But isps have acted as more than conduits since the beginnings of the industry. Alongside the intermediation that led to new service offerings and middleman roles, isps became instruments in a range of public policies, promoting connectivity (chapter 3), competition (chapter 4), privacy and security (chapters 6 and 7). Even in a simpler context, when an isp was little more than an “on-ramp” to the “information highway,” roadmaps for the future envisioned connectivity as a means of achieving political objectives such as cultural sovereignty (Industry Canada 1994) and local autonomy (Telecommunities Canada and Industry Canada 1997), and these visions themselves became an object of political struggle. In short, there is no going back to a mythical time when isps were apolitical and confined to simply providing connectivity. This was a state of affairs that never truly existed, and many of the new roles that isps have adopted are a response to the dramatic changes the internet has undergone since commercialization and liberalization. However, limits on the extent to which the internet’s points of control can be exploited are entirely appropriate. Privacy and net neutrality regulations are contemporary policy responses to the kinds of power relationships that have developed between intermediaries and their users. While we should not romanticize a past of unfettered connectivity that never existed, we can remember what made the internet so revolutionary: interactive connections between the endpoints, allowing for the removal or replacement of traditional intermediaries, and new kinds of social relationships. Preserving this transformative potential is particularly important given how much we now rely on a small

Conclusion

169

number of large internet companies. To address these stakes, we must take up the question of just how large and powerful we want the actors governing connectivity to be. DIVERSIT Y , CONSOLIDATION , AND CONCENTRATION OF POWER

Following liberalization, the trend in Canadian telecom has been towards consolidation, meaning that the giant firms which mediate our social relationships have grown bigger. While there is now less room for concentration in the Canadian telecom industry, the number of incumbents could be further reduced, and the survivors further enlarged, if regulators allowed. In addition, there are hundreds of Canadian isps that are protected from unbridled market power by the crtc’s regulatory regime. While iisps are usually dependent on incumbent infrastructure and are frequently referred to as “resellers” (of an incumbent’s product), these smaller intermediaries take a variety of organizational forms and approaches to the problems of connectivity. Their existence is marginal in many ways, and the ability of new entrants to innovate in the market or grow into true competitors against the incumbents is severely limited. But the capacity to create and operate alternative kinds of intermediaries is an important aspect of the internet’s foundational architecture. While internet access is not a natural monopoly, this does not necessarily mean that connectivity is best provided through market competition. There is no technical reason why telecom networks of various kinds and sizes cannot coexist, overlap, and interconnect. This is, after all, what the internet was designed to accomplish – allowing Autonomous Systems to communicate.1 The internet’s openness to new and diverse entities is a fundamental characteristic, but this openness depends less on protocols and institutions devoted to internet governance (such as icann and the ietf), and more on the isps that have ended up dominating in a geographic area. The reasons why there is often limited diversity among isps in a given “territory” include the concentrations of power that were inherited from the monopoly era (incumbency), the competitive rationality that conceives of isps as market actors (expected to exploit their differentials in power), and the material challenges of establishing a new network, which include access to property, infrastructure, and the electromagnetic spectrum. Under regulatory capitalism, all of these

170

Telecom Tensions

elements are subject to government regulation with the intent of controlling market power and promoting competition, but these regulations have been much more effective at the former than the latter. Still, even if the assumptions of regulation-for-competition have been mistaken and we cannot expect new competitors to challenge the incumbents, at the very least the regulatory regime maintains the possibility that other kinds of intermediaries can coexist and benefit from protections such as mandated access. This allows us to imagine intermediaries exercising their discretion in novel ways – the question being, what kinds of intermediaries might these be, and how would they be different? ALTERNATIVE INTERMEDIARIES

Internet governance takes place across various sites, but isps are in a position to be the most important governing nodes, given their direct control over infrastructure and the circulation of traffic. Scholars of nodal governance often make a normative argument in favour of local, decentralized, and diverse forms of governance that enhance democracy and empower weak actors (Johnston and Shearing 2003; Wood and Shearing 2007). To translate these values to internet governance is not to say that all isps should be small and local, but at the very least it must be possible for such intermediaries to exist. This is a possibility that can be denied when connectivity falls under centralized control, whether by a giant firm or the state, and it is important to ensure that new intermediaries can be created to address local needs and diverse values. A persistent pattern in Canada’s internet policy orientation since the mid-1990s has been the preservation of the status quo. This means legitimating the importance of dominant organizations, including incumbents and regulators, as well as using regulation to prevent or limit known problems, such as abuses of market power, or a final wave of consolidation that might clear the “playing field” of many remaining isps. These tensions are described in chapter 4, and leave the future open to three basic policy orientations. The first option is to embrace the giant firm, and allow the incumbents to dominate as the inevitable consequence of the market. The second option is to preserve the status quo of regulatory capitalism, maintaining checks on incumbent dominance while protecting the existence of other intermediaries (iisps) as a point of tension. This is

Conclusion

171

the “middle road” discussed in chapter 4, and is the least controversial because it conserves existing power relations. The third option would involve promoting some alternative conception of intermediaries, or a new framework for how isps relate to one another. When considering alternate approaches, we would do well to look to the actors and rationalities covered in chapter 3 that stand in contrast to incumbent-dominated liberalization. These include various ways of treating connectivity as a public good, or choosing something other than a market-based approach to the distribution of these networks. While even FreeNets must exchange money for services, this does not mean that commercial imperatives dominate. Alternative intermediaries are able to prioritize other values, such as inclusion, effective use, and local control. Some intermediaries continue to operate on the basis of these alternative rationalities within regulatory capitalism, in tension with its rationality of liberalization. However, we should also remember a time that seemed less constrained, before the die was cast. The mid-1990s were a formative stage, and not a golden era of the internet. But this was also an imaginative period, in which various dreams of connectivity and cyberspace circulated and were sometimes implemented, as in the FreeNet movement. While many of these dreams are easily dismissed as “cyber-utopian,” it does not follow that the existing political economy of telecom is natural, inevitable, or realist. Instead, we need to understand existing arrangements as the triumph of one rationality over others, which, despite its dominance, has yet to foreclose on other possibilities. In this book, I have examined intermediaries as instruments of public policy towards a variety of goals, and many of these objectives need not be uniformly or universally pursued. In other words, there is no one-size-fits-all answer to the problems of connectivity. The first public policy goal, connectivity as an end in itself (chapter 3), is the most fundamental, and echoes the universal service mandate of the telephony era. But this does not translate to monopoly control nor to a fibre connection for every home in the country, and how connectivity is promoted in an urban area should be quite different from its promotion in a remote First Nation. Other goals covered in these pages, including competitive success, privacy, security, copyright, and net neutrality, are politically contingent, with active debates about what aspects should be left to the discretion of intermediaries. Debates about standards and regulations are most important when subscribers lack meaningful choice between isps. But as I argued in

172

Telecom Tensions

chapter 3, the current regulatory regime is not the greatest obstacle to alternative intermediaries such as FreeNets, community networks, and isps that treat connectivity as a public good. Alternative intermediaries require support, and champions to mobilize this support. Organizations require internal governance. Networks require infrastructure and capital, as well as accessible sources of upstream connectivity. Finally, the dominant rationality of liberalization, in which connectivity is treated as a commodity best provided by private industry, is the chief obstacle to many projects that conceive of connectivity as some form of public good. To summarize, the process of intermediation is too complex to be generalized as desirable or harmful. There is little agreement about what intermediaries should be, or what values should shape their decisions, and we need not strive for such a consensus. Even local autonomy, a normative principle for both nodal governance scholars and community network advocates, is not a universal value in internet governance. Local actors or community members may not want to exercise control over connectivity. And yet, the ability to do so should be preserved distinct from the end-to-end principle. If endpoints no longer have the sort of decision-making power they once did, if we have come to depend on intermediaries’ discretion in many matters of increasing importance, then it must be possible for alternative intermediaries to exist, including those that serve a limited set of needs and values. This has nothing to do with reducing barriers to entry in order to promote a more competitive market, but is simply about allowing diverse and locally specific forms of governance to address the problems of connectivity. LOOKING TO THE FUTURE

Canada has a rich history of governing communications for public ends, and intermediaries have been political instruments since some of the earliest days of telecommunications. I have documented these historical precedents for current debates, showing that in many cases, internet policy issues did not begin with the internet. What has changed is that connectivity is more vital for society than ever, and internet intermediaries now occupy an unprecedented number of social roles, engendering new possibilities for the future. The scope of this analysis means that certain topics have remained largely unaddressed. By looking across Canadian institutions, the pre-

Conclusion

173

ceding chapters offer limited insight into international comparisons. In addition, we need more detailed accounts of what happens within organizations or governing nodes. How do organizations implement routine forms of governance? How do they undertake major decisions? How do individual agency and organizational structure shape collective agency? While it is safe to say that the crtc will remain an important sovereign authority in Canada and incumbent isps will maintain their dominant economic position in the near future, studies of alternative intermediaries, such as community networks and local connectivity projects, can also provide useful ideas for how to organize and govern connectivity. We need to better understand how these intermediaries can be established and structured, and useful strategies for keeping them attuned to the publics they serve. Going forward, it is important to remember what makes the internet unique: not just its pluripotency or its ability to assume new forms, but its architecture and governance. It is amazing that this assemblage of highly diverse networks without central coordination works at all. This underlying fact may become less remarkable as it becomes less true, if the trend continues towards an internet topography involving fewer, larger intermediaries. But the lessons of the internet – the dual movement of disintermediation and intermediation, the use of robust protocols to interconnect diverse endpoints – can be learned again as required. CONNECTING POSSIBILITIES

In June 2019 three public servants and a telecom consultant travelled for a series of meetings with small communities in the north-central region of Vancouver Island. This was a historically unusual approach to the public pursuit of connectivity; in previous years a project of this scale would have meant partnering with the local incumbent telus to extend service, but this time around the incumbent was conspicuously absent. Instead, an alternative intermediary would be built through the cooperation of regional, provincial, and municipal governments, dependent on the support of local publics and iisps. To attend the first two public meetings in June, I travelled into the island’s mountainous interior – to Gold River and Thasis, where cell phone coverage had yet to reach. Gold River had rejected telus’s cell service in 2014 due to some residents’ concerns over putting a tower in the middle of their village (McRae 2019), and similar worries would be

174

Telecom Tensions

raised at the third public meeting I attended on Cortes Island (R.L. Hales 2019). However, the team putting on these meetings did not come to talk about cell towers, but about a fibre-optic network that would soon be “landing” on citizens’ shores, who were invited to imagine some possibilities for this future of plentiful connectivity. The plan for the Connected Coast project is to lay thousands of kilometers of undersea fibre from Prince Rupert to the southern tip of Vancouver Island, stopping at more than one hundred fifty landing sites along the way. To develop and implement this plan, Connected Coast’s champions have had to coordinate between a regional government (Strathcona Regional District) and a municipal isp (CityWest), and to mobilize support from provincial and federal agencies. This has been ambitious, difficult work, with the potential for a variety of setbacks and unexpected challenges. When you read this, it is possible the cables will have been submerged and the glass inside them “lit” with laser light, or perhaps the project’s many obstacles will have left it unrealized. At the meetings in 2019, the network seemed imminent but immaterial – consisting of lines on a map and imagined possibilities. As I have previously discussed, the last mile can decide the success and failure of broadband projects. Many people live near a fibre-optic backbone without being able to access it, and many networks have been built across large distances, but failed to cover the shortest (and most expensive) final stretch to users. The Connected Coast network will be built by a cable-laying ship travelling along British Columbia’s shoreline and gradually unspooling its cargo, as workers prepare sites where the cables come ashore and local connections can be made. But what happens after these “meet-me-points” are built on land will depend on local actors, including municipal governments, First Nations, isps, and residents. Hence, there is a need for the project’s champions to visit local communities and to discuss their desires for the future. On Cortes Island, I accompanied the members of the project’s team as they were driven around the shoreline to scout the local infrastructure, surveying what already existed, and where the new cables might connect. Maps that were drafted in a distant office, with landing sites approved by regulators, had to be reconciled with the realities that exist on the ground – realities that encompass local people and organizations, cultures and histories. The culture of Cortes Island is rich and complex. Before Europeans decided its current name, its inhabi-

Conclusion

175

tants were the Klahoose, who still reside on the eastern shore where one of the cable landing sites is planned. The island was only connected by ferry in 1969 and to the electric grid in 1970, but the hippies and other residents who arrived in those years still appreciated the distance that it provided from modernity. Among the thousand or so residents today, many value permaculture and self-reliance, including gardening, farming, and oyster fisheries. Some are young entrepreneurs, connected to distant markets, and there are families with children that depend on the transportation link provided by the ferry service to Campbell River. What has made the island so particularly attractive is the fact that it is remote, but not isolated. As a consequence, while the internet provides greater connectivity, not everyone on Cortes sees this as a desirable development. Connectivity was the topic as some thirty Cortes Island residents gathered at a community hall to discuss plans for Cortes Island’s future. Victoria Smith, Connected Coast’s lead champion from Strathcona Regional District, opened with a presentation about the broad outlines of the project, and telecom consultant Don Sinclair showed where the network would arrive on the island. Some attendees interjected to ask about the network’s missing piece – the last mile. The General Manager of the local iisp Twincomm stood up to announce his plan to apply for public funding to cover this part of the network, and asked for the community’s support. Residents voiced their grievances with existing internet service, but most seemed willing to back their local intermediary across the last mile, expressing no support for the incumbent telco. Several attendees worried that once the cable was built, incumbents would swoop in, drive the local iisp out of business, and build towers to emit radiation across the island. Some were concerned about health effects, while other pointed to the benefits of telehealth. People explained that they depended on internet access, and that Cortes Island had long benefited from a community website that served local information needs; however, the woman who ran the community site had to leave the island for better connectivity. Another resident spoke up to say that if the cable arrived, he might be the next one to leave the island. By the end of the meeting, the project had cultivated some new supporters and a degree of enthusiasm, while a small group of residents huddled at the back discussing a petition to oppose it.2 Improving connectivity is not generally this controversial a goal, but the challenge of bringing social actors together in the public

176

Telecom Tensions

interest is a commonplace one. In this book I have examined just what makes connectivity possible. The answer, of course, is people. Technical expertise, equipment, and materials are all important, but getting internet infrastructure built is primarily a social concern. Before any fibre goes in the ground or under the water, people need to be convinced of a plan or vision, governance structures and funding models need to be established, regulatory approval secured, and competitors taken into account. Technology underpins our lives, but the underpinnings of technology – its development, design, operation, and use – are social. This means that, like society, technology can be organized differently, or tied to different values and goals. Technologies become objects of governance, even as they are instrumental in governing our conduct. Intermediaries are the socio-technical amalgams that connect us, the conduits of our communications. Among these, isps maintain the internet’s indispensable foundations, which are often obscured by the layers they prop up. We expect these organizations to direct our digital flows and to sustain the services we rely on. In addition to the roles they play in our lives, intermediaries are subject to public policy expectations, as a means for solving the problems of government. As this book has shown, communications infrastructure can be engaged to serve any number of visions for how society should be ordered. Given the vital importance of connectivity today, it is no surprise that it becomes a focus of governmental action, although this is not always for the best. The challenge today and in the future is knowing when we are expecting too much from our intermediaries, and when we are not demanding enough.

Notes

INTROD UCTION

1 I use the term “political economy” here in the narrower sense identified by Mosco (2009, 2), referring to “the social relations, particularly the power relations, that mutually constitute the production, distribution, and consumption of [communication] resources.” 2 See Hyslop 1984; Kolb 2001. 3 Canadian organizations’ willingness to comply with the state’s legal demands can largely be assumed, but Bangladesh provides an example of sovereignty being newly asserted over existing isps. For example, see New Age 2015. 4 While role theory in sociology remained focused on human actors, it was taken up and applied to collective or corporate actors in the fields of organizational studies, foreign policy analysis, and international relations. See Geser 1992; Thies and Breuning 2012. 5 As the Supreme Court of Canada ruled in a landmark copyright case (Society of Composers, Authors and Music Publishers of Canada v Canadian Assn. of Internet Providers, [2004] 2 scr 427), legal responsibilities have been deemed to depend on the role (or “function”) that an intermediary is playing. The “intermediary function” can be equated to acting as a “conduit” for a communication, with a “pure intermediary” acting “merely” as a conduit, and not otherwise exercising control or participating in the communication. In this book, the term “intermediary” does not imply such a “pure” or “mere conduit” status, unless explicitly stated. 6 Foucault wrote about human subjects in different ways, as actors as well as products (although usually emphasizing the latter). While on the one hand, Foucault seems to attack the very notion of intentional agency in much of his work (treating agency as an effect of discourse or power), he

178

7

8

9

10

11

12 13

Notes to Pages 11–18

also makes space for resistance and “acting otherwise,” and may have been moving towards a re-evaluation of the “free subject” in his last writings, specifically Foucault 1982; see Archer 2000; Bevir 1999; Caldwell 2007. Exactly what counts as a node is left ill-defined in the nodal governance literature, where nodes “may comprise individuals, groups (and parts of groups), organizations (and parts of organizations) or states” (Shearing and Johnston 2010, 501). They are “sites of knowledge, capacity and resources that function as governance auspices or providers” (Wood and Shearing 2007, 27). Importantly, nodal governance was not developed as an approach for analyzing how digital networks are governed, and so the nodes and networks these scholars refer to are principally organizational actors and their relationships, rather than the material infrastructure of digital networks. In my analysis, all governing nodes are organizations of some form, including private firms and government agencies. See Abdelnour, Hasselbladh, and Kallinikos 2017; Barnett 1993; Geser 1992; Thies and Breuning 2012; Wendt 2004. In social theory, Giddens’s (1984) approach to the problem of structure and agency has been highly influential (and contested), but the problem of collective agency has been addressed more directly by Domingues 1995; Gilbert 2014; Vandenberghe 2007. As many have pointed out, this phrase is not strictly accurate even for the online services (Facebook; Google) to which it typically refers. The basic commodity sold to advertisers is access to users’ attention; see Oremus 2018. In Canada, the two main locations have been 151 Front Street W. in Toronto and Harbour Centre in Vancouver, but a number of smaller ixps have been established in other cities in recent years; see cira n.d. Networks on the internet can either “peer,” exchanging traffic as equals without exchanging payment, or one network can pay to connect to another (either as “transit” or “paid peering”; see Norton 2010). A power imbalance (or imbalance in the amount of traffic moving between networks) can lead to one network demanding payment from another, and a refusal to pay these fees can escalate to disconnection, or “de-peering.” When this happens (not often), one part of the internet becomes difficult to reach from another. De-peering can also result when there is undesirable activity coming from a network, such as spam or malicious traffic; see Meier-Hahn 2017b. Netflix eventually paid and connectivity was restored; see Higginbotham 2014a; 2014b. I attended two regional (the British Columbia Broadband Conference

Notes to Pages 18–31

179

in 2014 and 2019) and four national industry conferences (the isp Summit in 2012, 2013, and 2014, and the Telecom Summit in 2014), two major international industry and policy events (ietf 88 and nanog 59, both in 2013), and several more specialized meetings dealing with various aspects of internet policy (including the 2014 ixp Symposium, the 2014 Canadian Internet Forum, arin on the Road in 2013, the cpi Citizen Planning Circle on Effective Use of Rural Broadband in Olds in 2014, and the Digital Futures Symposium in 2013, 2015, and 2017). 14 In total I carried out expert interviews with thirty participants, transcribed discussions at a two-day symposium with dozens of attendees (including individuals from local governments and the telecom industry), and had numerous informal discussions with industry executives, network operators, lawyers, consultants, police officers, and government officials. 15 I appreciate Ben Klass’s help in working through this categorization. 16 The crtc is sometimes described as being “arm’s-length” from government, but the actual distance depends on the government of the day. There have been periods in the crtc’s history where it was more regularly overruled and directed by government (Winseck 1998, 236–8), and the crtc generally tries to align itself with the direction set by government. It should also be noted that decisions of the crtc can be appealed to Federal Court. CHAP TER ONE

1

2

3 4

See Pickren 2018. The internet currently extends to the inhabitants of the International Space Station, with new protocols being developed to extend its reach further through our solar system; see Smith 2017. Telecom Regulatory Policy crtc 2016-496 created a $750 million rural broadband fund, to support projects that would provide speeds of 50Mbps download and 10Mbps upload. In 2018 (following Blais’s replacement by Chairman Ian Scott) these speeds were halved to 25Mbps download and 5Mbps upload, providing an easier target for projects to meet, with the intention that these projects increase speeds to the 50/10Mbps target in the future. See Karadeglija 2018. Most importantly, the Broadcasting Act and Telcommunications Act, the futures of which are frequently up for debate, as in btlr 2020. The Canadian Telecom Summit (http://www.telecomsummit.com), isp Summit (http://www.ispsummit.ca), and iic Annual Conference (http://www.iic-canada.ca).

180

An autonomous system (as) is a network, designated with a unique number, in control of a specified block of ip addresses. Between these cities, it is named the Queen Elizabeth II Highway.

5 6

Notes to Pages 33–44

CHAP TER T WO

1 British Columbia was a notable exception to Bell’s early ambitions, and the Northwest Territories (which until 1905 encompassed the lands that would become Alberta and Saskatchewan) was a vast territory in which Bell’s charter was disputed. See MacDougall 2013, 186. 2 The Act (Government of Canada 1880) was an incredible accomplishment for Bell and its head, Charles Sise, who achieved it without any significant political opposition through “lightning” speed (Rens 2001, 67–8). For Babe (1990, 69–70), the unsolved “riddle” of this accomplishment is that the company’s American connections never became a point of political debate, given the Canadian government’s nationalistic orientation. 3 This was part of a larger conflict between utility companies (many of whom sought a charter similar to Bell’s) and municipalities over control of infrastructure in Canada’s rapidly developing cities. See MacDougall 2013; Armstrong and Nelles 1986. 4 In a few cases, the company also offered discounted or free telephone service for as long as required to eliminate a municipal rival. See MacDougall 2013; Rens 2001. 5 Armstrong and Nelles 1986; Cashman 1972; Rens 2001. The Maritimes also gained independence from the Bell system in the 1880s. 6 Most notably Francis Dagger, a central figure advocating for municipalities and against the Bell monopoly in this period; see Armstrong and Nelles 1986. 7 Government of Canada 1882, sec. 4; see Rens 2001, 75. 8 Ultimately, Bell relied less on legal precedent or its charter to make its case before the Mulock Committee, and instead emphasized technical arguments for maintaining both local and long-distance service under a single system, which would be best left under its expert control. Bell’s monopoly was justified on the basis of technical necessity as dictated by the inherent properties of telephony; see MacDougall 2013, 180–2. 9 Rens (2001, 99) characterizes the move as a “tacit exchange of monopoly for regulation.” The regulator would be the Board of Railway Commissioners (brc), and in 1908, an amendment to the Railway Act extended the brc’s powers over all telecom (telephone and telegraph) companies under federal jurisdiction. The consequences of this move included grant-

Notes to Pages 44–52

10

11

12

13 14 15

16

17

181

ing more power to municipalities, and regulating interconnection between Bell and the smaller (or independent) telephone companies. The brc was renamed the Canadian Transport Commission (ctc) in 1938. This did not prevent these costs or their disclosure from being a source of dispute, and interconnection rates became the key point of contention before the brc in subsequent years. The Board decided questions such as whether Bell should be compensated for business it lost to its competitors through interconnection, and had to distinguish between which companies were competitors to Bell and which companies were not. See Babe 1990; Winseck 1998. In the 1920s, seven regional telcos formed the Telephone Association of Canada, and subsequently went about building the Trans-Canada Telephone System (tcts) – a long-distance line that would help unify Canadian telephony. Each company took responsibility for the tcts in its area, but Bell played an essential leadership role; see Armstrong and Nelles 1986; Rens 2001; Rideout 2003. While cable networks (cablecos) eventually became the telcos’ main competitors in the internet era, when catv was first deployed in the 1950s it was heavily dependent on telco infrastructure (such as telephone company poles, or entire systems built by telcos and then leased to the cablecos). Therefore, while the development of cable networks in the mid-twentieth century had the potential to radically transform the nature of communications networks in Canada, this potential was kept in check by the incumbent telcos, who had no desire to upend the comfortable state of Canada’s telecom regime. Until liberalization, cablecos were also comfortable serving their own niche markets, with no desire to compete against the much larger telcos, and keen to be exempt from greater government regulation, including the mandate of universal service. See Babe 1975; 1990; Winseck 1998. Telecom Decision CRTC 77-16. In the case of government-owned telco networks, as in Manitoba and Alberta. Previously, while the federal government could have chosen to assert its authority over telecom, provincial telcos were left alone due to “sensitivities to provincial interests”; see Winseck 1998, 212. While scholars of neoliberalism typically agree that the extension of market logic to new domains is one of these essential characteristics, other contours of neoliberalism (such as the extension of individual responsibility or the shrinking of the state) are more controversial. The move followed a series of stories by the cbc and an order from the federal Cabinet directing the crtc to act; see Tunney 2018.

182

Notes to Pages 52–65

18 Specifically, the Order Issuing a Direction to the CRTC on Implementing the Canadian Telecommunications Policy Objectives; see Government of Canada 2006. 19 I am grateful to Tamara Shepherd for helping to crystallize this point. 20 This was not a dramatic departure following liberalization, as even during the monopoly era, state-owned utilities were promoted on the basis of their “business-like management” (Armstrong and Nelles 1986, 322). 21 For histories of the internet’s initial developments, see Abbate 1999; Hafner and Lyon 1996; Leiner et al. 2009. 22 According to a former president, at one point the bc Internet Association represented well over a hundred isps (M. Hrybyk, personal communication, 4 May 2016). Winseck (1998, 296–7) estimates between 15 and 20 isps existed in major cities such as Vancouver, and half a dozen in mid-sized cities such as Windsor. 23 See “isps Put the Boots to Telco” n.d.; Winseck 1998. 24 While Bell and telus have remained distinct corporate entities, they do share one another’s wireless facilities (Bell using telus facilities in Western Canada and telus using Bell’s in the east) – something that competitors have often pointed out in debates over wireless competition. See cpac 2012; 2013b. CHAP TER THREE

1 See Clement et al. 2012; Gurstein 2003. This argument has also been reflected in much of the scholarship on the digital divide since the late 1990s, which has emphasized skills and usage over physical access; see van Dijk 2006. 2 The most significant initiative through which incumbent isps (specifically Shaw, Rogers, telus, and Bell) promote elements of effective use is through their support of MediaSmarts, an organization devoted to promoting digital and media literacy in Canada, by helping “children and teens develop the critical thinking skills they need for interacting with the media they love”; see MediaSmarts n.d. 3 Rogers was the first company to take on a more socially responsible role in providing low-cost connectivity for low-income families. The practice was then adopted by telus and Bell, receiving public encouragement from ised Minister Navdeep Bains (ised 2017). ised has since transformed these efforts into a government-backed social assistance program being extended across incumbent companies; see Chhabra 2018b. 4 Vancouver Regional FreeNet Assn. v M.N.R., [1996] 3 fc 880; see Bodnar 2007.

Notes to Pages 66–74

183

5 The idea that something can both be a public good and provided by private industry is not unique to connectivity, and has been recognized by contemporary updates of public good theory; see Kaul and Mendoza 2003. 6 Also known as Free-Net, not to be confused with Freenet (which is a peer-to-peer platform designed to be resistant to censorship). 7 Bodnar (2007) counts 26 in Canada in 1995, with 67 “in various stages of development.” However, by 2007 this number had been reduced to approximately six. 8 Garth Graham, personal communication, 26 and 27 September 2015. 9 What one Executive Director of the ncf jokingly referred to as “mandatory donation”; see Hunt 2014, 61. 10 For example, ncf’s partnership with Ottawa Community Housing; see Schnurr 2016. 11 This help can extend to kinds of technical assistance that are not strictly an isp’s responsibility. N. Ouzas, personal communication, 8 October 2014; R. Kouhi, personal communication, 17 October 2014. 12 There were bold visions put forward for a national broadband plan from the mid-1990s to the early 2000s, the most notable of which was the National Broadband Task Force’s (2001) recommendation for a multi-billion-dollar investment in infrastructure to extend broadband to all communities. The 2019 budget included the most major federal commitment to telecom infrastructure, with a $1.7 billion commitment for the Connect to Innovate program; see Government of Canada 2019. 13 Municipally owned isps of various kinds exist in Cochrane (https://www.cochranetel.ca/), Granisle (https://villageofgranisle.ca/granisleinternet-service/), Kingston (http://www.utilitieskingston.com/ Networks), Kincardine (http://www.brucetelecom.com), Georgina (http://www.georgina.ca/broadband), Muskoka (http://www.lakeland networks.com), Prince Rupert (http://www.citywest.ca), Stratford and St. Marys (http://www.rhyzome.ca/), Sudbury (http://agilisnet .com/), and Thunder Bay (http://www.tbaytel.net). 14 Active regional networks include the Columbia Basin Broadband Corporation (https://ourtrust.org), Eastern Ontario Regional Network (http://www.eorn.ca), EIDNet (https://www.eidnet.ca/), the Niagara Regional Broadband Network (http://www.nrbn.ca), the London and Region Global Network (http://www.largnet.ca/), Connecting WindsorEssex (http://www.cw-e.ca/), YorkNet (http://york.ca/yorknet), Internet Papineau (http://www.ipapineau.net), and Réseau Picanoc.net (http://www.picanoc.net). The swift project provides public funding to new fibre networks in southwestern Ontario (https://swiftrural

184

15

16

17 18

19

Notes to Pages 75–7

broadband.ca/). There have also been a number of publicly funded regional projects to improve connectivity for First Nations communities (First Mile Connectivity Consortium 2018), and the Qiniq network (https://www.qiniq.com/) is a satellite network serving all communities in the territory of Nunavut. Numerous other broadband projects have been pursued through provincial and territorial governments; see ey 2016a; 2016b. The best-known Canadian example is the business/government cooperative of Fredericton; see Richard and Philpot 2014. Six other cooperatives act as isps in a region west of Kitchener and north of London, Ontario (http://www.cni.on.ca/). Other examples of co-operatives include csur (http://www.csur.ca/), ccap (https://www.ccapcable.com/), and CoopTel (http://www.cooptel.qc.ca) in Quebec, crrs (https://crrs .net/) in Labrador, clwc in Nova Scotia (http://cedarlakewireless.com), Access (http://www.myaccess.ca/) in Saskatchewan, and a group of natural gas co-ops that formed Corridor Communications (http://cciwireless.ca) in Alberta. As in Olds, the role of a long-term project champion was key in preparing a plan for such a unique business venture, and in gaining and maintaining the support of local leadership. However, instead of being spearheaded by members of a not-for-profit community organization such as oicrd, the project was championed by the manager of the City’s ict department. Specifically, Calgary, Campbell River, Kamloops, Kelowna, Nelson, New Westminster, and Penticton. There were a number of contractual disputes between Axia and Bell (Axia 2011; 2015; Gignac 2003) as well as disputes with the provincial government (Auditor General of Alberta 2018). While many of Alberta’s rural public facilities have benefited from connections to SuperNet, a persistent problem has been connecting other kinds of users to the network; in many rural communities there was simply no business case for private investment by isps in the last mile. See Bull 2017; Gignac 2010. At the 2013 Digital Futures Symposium (see R. McMahon, McNally, and Joseph 2020), one former chairman described how the oicrd was meant to get the city away from the political cycles of small-town governance and the “fickle” leadership that can result. The oicrd has depended on volunteer support and is not part of the city administration, but maintained close ties to the municipal government. It has therefore been able to shepherd the fibre project over the long term, while taking advantage of some of the funding opportunities (provincial loans) available to

Notes to Pages 78–84

20

21

22 23 24

25

185

municipalities. The Town of Olds has recently sought repayment of these loans and greater oversight over the oicrd (Collie 2020). Project champions in Olds have discussed their experiences with numerous other municipal representatives considering investing in broadband. O-Net also took steps to develop into a regional isp, offering services to any network that can connect to an ixp in Calgary. See Collie 2015; Dobson 2017. Chris Litschko (personal communication, 28 March 2014), the manager of a municipally owned isp (Lakeland Networks, which grew out of a power utility), attributed the decision to expand into internet services to an “entrepreneurial point of view,” and described the choice of whether or not to extend services to particular customers as ultimately a “business decision.” See CBC Radio 2019. This is essentially the argument made by the City of Calgary before the crtc in 2014; see City of Calgary 2014. This can mean being selective about what parties the city does business with, aside from practical or business considerations. For instance, the City of Kamloops (2015) states that dark fibre from its network “is available by lease to customers who can demonstrate economic or other benefits to Kamloops.” A particularly important factor is geographic location and proximity to affordable upstream connectivity. Locally available expertise is also a major asset, as are stable and supportive governments with access to funding; funding and government support continues to be an issue for O-Net. See Collie 2020. CHAP TER FOUR

Note to the epigraph: Speaking at the Canadian Telecom Summit; see cpac 2012. Ken Engelhart was Senior Vice-President, Regulatory and Chief Privacy Officer for Rogers Communications. 1 Some, such as SaskTel, Manitoba Telephone System, and Alberta Government Telephones, were provincial Crown corporations. Others (such as Bell) operated under government charters that granted them exclusive rights in exchange for certain obligations. 2 The terms “market failure” and “market power” (and whether these states accurately describe the industry) are highly contentious in Canadian telecom. In policy debates iisps often point to the market power of incumbents, while incumbents tend to minimize their own power by emphasizing the competitiveness of the existing market.

186

Notes to Pages 85–90

3 In Telecom Decision 2005-20, the crtc created an incentive structure for telco incumbents (incumbent local exchange carriers, or ilecs) to provide a high quality of service for the smaller competitors that depend on using the incumbents’ infrastructure. The incentives are meant to ensure that all competitors are treated equally by the incumbents (thus levelling the playing field between these competitors), as well as compensating for the power of the incumbents (levelling the playing field between incumbents and smaller competitors). See crtc 2005, para. 31. 4 The ruling in crtc Telecom Decision 2006-14 was intended as a fair way for smaller telcos to obtain broadcast licenses (as Broadcasting Distribution Undertakings, or bdus), thereby allowing them to compete in the “core business” of cablecos. The crtc decided that in exchange, the cablecos would require lowered “barriers to entry” to compete in the telcos’ core business, which meant greater access to telco facilities. 5 In 2016, crtc Chairman Blais responded to a Bell representative’s repeated calls for a level playing field by describing the idea as “elusive.” See crtc 2016b. 6 Other agencies governing competitiveness in the telecom industry include Innovation, Science and Economic Development Canada (formerly Industry Canada), the Competition Bureau, and ultimately the federal Cabinet. 7 Before the mid-1990s, internet service could often only be obtained through an ren linked to CA*net. While the focus of these networks was research and education, bcnet ended up connecting a growing number of interested institutions, and Nova Scotia’s network (nstn) distinguished itself by taking a more commercial, for-profit route. On 31 March 1997 CA*net “passed the torch ... to the private sector” (CA*net Institute 2001, 10) when the ren backbone was sold to Bell. Some rens were subsequently bought out by private industry, while other public networks changed their focus to become non-competitors. Today, rens such as Cybera serve innovation- and entrepreneurshipenabling functions, but these efforts must be carefully directed to avoid competing with private industry. 8 See crtc 1994; 1995; Government of Canada 1996; Information Highway Advisory Council 1995; 1996; 1997. 9 Internationally, Canada is unusual in the reach of its cable network and because of its regulatory regime mandating wholesale access to both telco and cableco infrastructure; see Van Gorp and Middleton 2010. 10 As of 2020, the parameters of this micro-regime are still contested and undetermined; see Hathout 2018.

Notes to Pages 91–9

187

11 ised administers a regime for mandated access to wireless towers (Industry Canada 2013), which is intended to increase competition and limit the number of towers, but this enables multiple companies to share space on a single tower, rather than letting one company make use of another’s network. 12 However, there is a requirement for wholesale providers to isolate wholesale customer information from their retail operations through an internal carrier services group (csg; see crtc 2011b). This is meant to prevent incumbents from trying to retain or reclaim retail customers who have transferred to an iisp being served through an incumbent’s wholesale division. 13 For instance, see crtc 2013a. Ken Engelhart of Rogers characterized this process as “a bit of a dark art” that can result in some baffling crtc judgments. Engelhart also criticized how the regulatory regime rewards those who work it most effectively, stating: “I don’t think costing can ever be good enough or accurate enough to really sustain a competitive system like this. The whole system depends, not on how good your sales is, not on how good your marketing is, not on how good your engineering is, but how good your costing people are” (cpac 2013b). 14 According to Bram Abramson, TekSavvy’s former chief legal and regulatory officer (personal communication, 27 January 2017), the most important advantage incumbents have (in addition to greater regulatory resources) “is that when it comes to wholesale, it’s their network, not ours. They have an inside view that we do not – on what is possible or not, what works or not, what costs what, and so forth.” 15 See comments from Bell (crtc 2014b) and Cogeco (crtc 2014c) in the crtc’s 2014 wholesale review. 16 Near the end of 2016, crtc Chairman Blais raised structural separation as a possibility for incumbents that “grumble about having to provide wholesale access to their competitors” (Blais 2016), but this seems to have been meant to remind the giant firms that their corporate personhood depended on sovereign authority, rather than as a serious threat. 17 The heart of the regulatory dispute concerned Bell’s attempt to charge iisps for the total amount of data used by their customers, and was tied to longstanding concerns among incumbents about “congestion” on their networks; see Geist 2011a. This congestion was attributed to “heavy users” (disproportionately iisp subscribers) whose bandwidth consumption had to be “subsidized” by lighter users. Hence, ubb would be a way to compensate incumbents for congestion caused by iisps and to adjust the behaviour of some heavy users by linking usage with willingness to pay. Following

188

18

19

20

21

Notes to Pages 99–127

Cabinet’s intervention, the crtc reversed its original decision and ruled more favourably for iisps, but details arising from aspects of the decision would continue to be disputed before the regulator for years to come. See Chase and Marlow 2011; Krashinsky 2011. Interestingly, both ministers chose to do so through their Twitter accounts, making this the first time that a change in Canada’s telecom regulations was effected by a tweet. As characterized by Michael Geist (personal communication, 10 July 2014). Former crtc Commissioner Timothy Denton (personal communication, 12 November 2014) described it as “the first time a government realized it could lose an election on an apparently obscure issue of telecommunications policy.” Specifically, the crtc’s rejection of mandated wholesale in wireless and the crtc’s reluctance to investigate abusive sales practices; see Jackson 2018. In the announcement of a new review of wireless services just two days after the policy direction, the crtc (2019a) signalled that it favoured mandated access to these networks, but only on a temporary basis until new entrants established themselves and “market forces take hold.” CHAP TER F IVE

1 Bell Media Inc. v GoldTV.Biz, [2019] fc 1432. A voluntary websiteblocking regime to combat child abuse was established in the mid2000s, and is discussed in chapter 7. CHAP TER SIX

1 R v Spencer, [2014] 2 scr 212 at para. 46. 2 R. v TELUS, [2013] 2 scr 3. 3 Bell successfully opposed an Ontario Provincial Police warrant in 1947 and argued against a Criminal Code wiretap amendment in 1948. In the absence of legislation, police concluded that they themselves had the lawful authority to authorize a wiretap; see Cornfield 1967, 113–14. 4 From 1951 into the 1970s the rcmp maintained an extensive but highly secret wiretapping program in cooperation with Bell and the British Columbia Telephone Company for national security purposes (Molinaro 2017). In 1969 the Chief of the Toronto Police stated that while his department had been tapping phones since 1966, they had “no formal, or informal, liaison with the telephone company on wiretapping, but added: ‘The company does assist in other ways’” (Burns 1969, 2). Before the Privacy Act, wiretaps were sometimes installed without the phone

Notes to Pages 129–42

5 6 7

8 9 10 11

12

13 14

15 16

17 18

189

company’s knowledge, and the information obtained was generally not used as evidence in court; see Globe and Mail 1977. BMG Canada Inc. v John Doe, [2004] 3 fcr 241. BMG Canada Inc. v John Doe, [2005] 4 fcr 81. Today, major copyright owners rely on Canadian isps to forward “notices” to subscribers suspected of piracy, using a gentler, governmentmandated regulatory regime that tries to nudge Canadians away from infringement. More broadly, copyright owners target large distributors, and promote structural or legal changes, such as the recent proposal for the crtc to administer website blocking; see Fair Play Canada 2018. Copyright owners who still threaten to sue individual infringers engaged in file sharing are often called “copyright trolls”; see Zajko 2015b. See Bell 2004; Kerr and Cameron 2006; Rogers 2004; Shaw 2004. Konrad von Finckenstein, who would go on to become Chairman of the crtc. Rogers v Voltage, [2018] 2 scr 643. TekSavvy fought for the right to notify the subscribers it was being asked to identify (Dobby 2012; Gaudrault 2012), and argued for ways to minimize the privacy impacts of any information disclosure; see TekSavvy 2015, 722–31. Distributel originally agreed to identify some subscribers, but then changed its position and opposed the copyright owners in court, leading them to drop the case; see Distributel 2013, 52–4; O’Brien 2013. ME2 Productions, Inc. v Doe, [2019] fc 214. And in particular, PIPEDA’s section 7(3)(c.1)(ii), which permits an organization to “disclose personal information without the knowledge or consent of the individual ... [if] the disclosure is requested for the purpose of enforcing any law of Canada, a province or a foreign jurisdiction, carrying out an investigation relating to the enforcement of any such law or gathering intelligence for the purpose of enforcing any such law.” R v Rogers Communications, [2016] onsc 70; R v TELUS, [2013] 2 scr 3. In 2014 some intermediaries expressed concerns that regulations prevented them from disclosing information about lawful access. While telus sought government clarification (Bronskill 2014), TekSavvy and Rogers published their reports without consulting government. Rogers’ Ken Engelhart remarked in an interview (Freeze, Dobby, and Wingrove 2014) that “there was too much sensitivity in the past about not wanting to upset law enforcement officials.” R v Rogers Communications, [2016] onsc 70. Privacy custodians could further distinguish themselves through policy advocacy, commitments about data routing, storage, and retention, and

190

Notes to Pages 144–51

more detailed information about their procedures for disclosing personal information; see Clement and Obar 2016. 19 The Office of the Privacy Commissioner (2018) of Canada has announced a shift towards a less complaints-based, more proactive approach to privacy regulation. This is an encouraging move, but it remains difficult to evaluate how much of a departure it represents. CHAP TER SEVEN

1 In a document from 2012, the example used by Public Safety Canada (2014, 1152) to justify the need for interception capabilities involved a Hell’s Angels case in which it took the rcmp six weeks to develop a “solution” for an intermediary that was unequipped for interception. 2 Specific concerns included questions around the sorts of equipment intermediaries would be required to purchase, install, and operate, what sorts of information they would be required to collect about their subscribers, and how quickly they would be forced to provide such information upon request. For example, Industry Canada 2014a; Nevis Consulting Group 2003; Paperny 2012; Public Safety Canada 2012; 2014. 3 Court orders take time to obtain and utilize police resources. In some cases, it is simply impossible to obtain a warrant at the preliminary stages of an investigation, where there are insufficient grounds to believe that a criminal offence has been committed. Access to subscriber information has frequently been presented as being key to these “pre-warrant” stages of an investigation, but as long as intermediaries have no legal obligation to comply, “police depend on moral suasion and a service provider’s sense of civic duty to obtain their cooperation” (Public Safety Canada 2012, 190). 4 In 2009, Bell Canada reportedly had twelve employees dedicated to supporting law enforcement, processing an average of 250,000 law enforcement requests a year. See J. Brown 2009. 5 While in many cases intermediaries disclose information without charge, rcmp documents (Ling 2015) indicate that millions of dollars must have been paid by police to telecom companies since the early 2000s, with address information (name, street address, email) costing between $1 and $15 per disclosure, call record disclosures running into the hundreds of dollars, and wiretaps costing in excess of a thousand dollars apiece. 6 Bill C-30 would have empowered government-appointed “inspectors” who had the authority to examine the facilities and equipment of a telecommunications service provider, including any relevant docu-

Notes to Pages 152–7

7 8

9 10

11

12

191

ments, to verify compliance with the Act. See House of Commons of Canada 2012, secs. 33–8. R v Rogers Communications, [2016] onsc 70. Whereas previously, subscriber information could typically be obtained the same day as requested, police now estimated that applying for a court order could add days (Payton 2014) or weeks (Bronskill 2015) to the process. According to one internal rcmp survey, the effect this has had on police investigations has not been significantly negative, mostly resulting “in investigative delays, but not necessarily derailing investigations or changing ultimate outcomes” (Boutilier 2015). Publicly, police claimed the Spencer decision has meant that some criminal investigations are not being pursued, with the head of the Toronto Police Sex Crimes Unit claiming that the number of children rescued has been “cut in half” as a result of the decision (Cribb and Greenblatt 2015). There are ways to improve information sharing while respecting Spencer obligations, and telus (2016, 132) has “collaborated with law enforcement agencies to establish a new process to expedite the sharing of this critical information through court orders where children are in danger.” An exception being Bill S-4 in 2014, which imposed certain obligations for organization that had experienced a data breach; see Lithwick 2014a. This unifying discourse is exemplified by the presentation given by the Deputy Minister of Public Safety to representatives of the private sector at the first meeting of cstac in 2010. Competing telecom companies participating in the cyber security partnership were told they had a “shared interest ... in ensuring that individual Canadians and Canadian businesses trust our infrastructure, our products, services and institutions” (Industry Canada 2014b, 61). These include the Canadian Telecommunications Emergency Preparedness Association (ctepa), the National Cyber-Forensics and Training Alliance (ncfta: http://www.ncfta.ca/) and the Cyber Security Threat Exchange (cctx: https://cctx.ca/), as well as the Information Technology Association of Canada’s (itac) Cyber Security Forum (https://itac.ca/csf/). For example, the release of the Best Practices prepared by ctcp and cstac (described in this chapter) was initially meant to be publicized during Cyber Security Awareness Month in October 2013 (see Industry Canada 2014b). In the end, government opted for a “low-profile” approach to the release of the document, which included the “public acknowledgement” of cstac’s existence (ised 2016, 228), albeit with only “cursory information” about the group (250). Apparently, it is

192

13 14

15

16

17

18

19

Notes to Pages 158–75

important that intermediaries are aware of the existence of such collaborative relationships and that these efforts also undergo “public disclosure” (251), but generally the public visibility of these security partnerships has been carefully limited. Bill C-59, An Act respecting national security matters. 1st session, 42nd Parliament, 2019. A notable exception is cse’s Security Review Program, which intermediaries must comply with in order to be eligible for the crtc’s “Broadband Fund” (crtc 2019d). In recent years there has been a drawn-out controversy over a possible ban of Huawei equipment in Canadian networks, made more difficult by the fact that Canadian companies have been major adopters of the China-based company’s products; see Wingrove 2019. In 2011 and 2012, lawful access requirements for telecom equipment were pursued through updates to the Solicitor General’s Enforcement Standards (sges). Telecom companies successfully opposed changes to the sges, primarily on the basis of the costs of meeting the proposed lawful interception requirements; see Freeze and Trichur 2013. Specifically, the use of “upstream intelligence” to “recover” up to 30% of bandwidth in Bell’s core network, which would otherwise be taken up with traffic considered “illicit use and abuse” (Macaulay 2010). Most recently, ised and cse collaborated on a voluntary security certification for businesses by establishing the CyberSecure Canada program, for which Bell is one of the certification bodies; see ised 2019. The crtc has also granted very broad latitude for traffic monitoring and management as long as these are “employed to protect users from network threats ... as a necessary part of an isp’s network operations” (crtc 2009, sec. 44). isps can accommodate privacy custodian responsibilities by limiting surveillance and the collection of information to what is necessary to provide cyber security. The ceo of Cogent characterized himself as being in the “‘not only dumb pipe, but dumb network’ camp,” arguing that with enough bandwidth, “there’s not a whole lot of intelligence you need in your network” (Duckett 2018). See also Harvey 2018; Morris 2018. CONCLUSION

1 See ch. 1, note 5. 2 For recordings and summary of the meeting, see R.L. Hales 2019.

References

Abbate, J. 1999. Inventing the Internet. Cambridge, ma: mit Press. Abdelnour, S., H. Hasselbladh, and J. Kallinikos. 2017. “Agency and Institutions in Organization Studies.” Organization Studies 38, no. 12: 1775–92. Anderson, S. 2011a. “Net Neutrality in Canada: Closing in on Internet Openness.” In The Internet Tree: The State of Telecom Policy in Canada 3.0, edited by M. Moll and L.R. Shade, 113–21. Ottawa: Canadian Centre for Policy Alternatives. http://www.policyalternatives.ca/sites/default/files /uploads/publications/National%20Office/2011/06/Internet_Tree_0.pdf. – 2011b. “The Great Internet Billing Debate: Stop the Meter.” Financial Post, 28 March. http://business.financialpost.com/fp-comment/the-greatinternet-billing-debate-stop-the-meter. Antoniadis, P. 2018. “The Organic Internet: Building Communications Networks from the Grassroots.” In Co-Designing Economies in Transition: Radical Approaches in Dialogue with Contemplative Social Sciences, edited by V.M.B. Giorgino and Z. Walsh, 235–72. Cham, Switzerland: Palgrave Macmillan. Archer, M.S. 2000. Being Human: The Problem of Agency. Cambridge: Cambridge University Press. Arellano, N. 2011. “Small isps Foresee Cost Burden in ‘Lawful Access’ Bills.” ITBusiness.ca, 27 June. http://www.itbusiness.ca/news/small-isps-foreseecost-burden-in-lawful-access-bills/16419. Armstrong, C., and H.V. Nelles. 1986. Monopoly’s Moment: The Organization and Regulation of Canadian Utilities, 1830–1930. Philadelphia, pa: Temple University Press. Auditor General of Alberta. 2018. “Report of the Auditor General.” Office of the Auditor General of Alberta. November. https://www.oag.ab.ca /documents/678/2018-11-07-_November_2018_Report.pdf (accessed 16 September 2020).

194

References

Austen, I. 2005. “A Canadian Telecom’s Labor Dispute Leads to Blocked Web Sites and Questions of Censorship.” New York Times, 1 August. Avis, A.W. 1995. “Public Spaces on the Information Highway: The Role of Community Networks.” M.A. thesis. Calgary, ab: University of Calgary. Axia. 2011. “Axia NetMedia Corporation Management’s Discussion & Analysis for the Year Ended June 30, 2011.” 20 September. https://web.archive.org/web/20150509182453/http://ir.axia.com/files/doc_ financials/2011/MDA%20YE%20FY11%202011-9-20%20FINAL.pdf. – 2015. “Axia NetMedia Corporation Management’s Discussion & Analysis for the Year Ended December 31, 2014.” 5 March. https://web.archive.org/web/20150319040804/http://ir.axia.com:80/files/do c_financials/2014/AXX-Q4-2014-MDAFS.pdf. Babe, R.E. 1975. Cable Television and Telecommunications in Canada: An Economic Analysis. East Lansing, mi: Division of Research, Graduate School of Business Administration, Michigan State University. – 1990. Telecommunications in Canada. Toronto: University of Toronto Press. Bailey, S. 2007. “‘Up to Our Knees in Evil’: Police Fight Growing Demand for Web Child Porn.” Canadian Press, 7 February. Barnard-Wills, D., and D. Ashenden. 2012. “Securing Virtual Space: Cyber War, Cyber Terror, and Risk.” Space and Culture 15, no. 2: 110–23. Barnett, M. 1993. “Institutions, Roles, and Disorder: The Case of the Arab States System.” International Studies Quarterly 37, no. 3: 271–96. Barrett, T. 2005. “To Censor Pro-Union Web Site, Telus Blocked 766 Others.” Tyee, 4 August. http://thetyee.ca/News/2005/08/04/Telus Censor/. Barth, B. 2018. “The Fight against Google’s Smart City.” Washington Post, 8 August. https://www.washingtonpost.com/news/theworldpost/wp /2018/08/08/sidewalk-labs/. Beaudry, P. 2010. “Wireline Deregulation: The Canadian Experience.” Telecommunications Policy 34, no. 10: 606–15. Beer, J.F. de, and C.D. Clemmer. 2009. “Global Trends in Online Copyright Enforcement: A Non-Neutral Role for Network Intermediaries?” Jurimetrics 49, no. 4: 375–409. Behar, R. 2018. “Here’s How Bell, Rogers and Telus Profit from Your Mobile Subscriber Data.” MobileSyrup, 18 May. https://mobilesyrup.com/2018/05/18/bell-telus-rogers-sell-location-datathird-parties-enstream/. Bell. 2004. “Written Representations.” 11 March. https://cippic.ca/sites/ default/files/file-sharing-lawsuits/Bell_Written_Submissions.pdf.

References

195

Bell Canada. 2010. “pstp 08-0107esec: Combating Robot Networks and Their Controllers: A Study for the Public Security and Technical Program (pstp).” 6 May. http://www.scribd.com/doc/51938416/Botnet-Analysis-Report-Final-Unclassified-v2-0. Bendrath, R., and M. Mueller. 2011. “The End of the Net as We Know It? Deep Packet Inspection and Internet Governance.” New Media & Society 13, no. 7: 1142–60. Bevir, M. 1999. “Foucault and Critique: Deploying Agency against Autonomy.” Political Theory 27, no. 1: 65–84. – ed. 2011. The Sage Handbook of Governance. London: sage. Birdsall, W.F. 2000. “The Digital Divide in the Liberal State: A Canadian Perspective.” First Monday 5, no. 12. https://firstmonday.org/ojs/index.php/fm/article/view/820/729. Birnhack, M.D., and N. Elkin-Koren. 2003. “The Invisible Handshake: The Reemergence of the State in the Digital Environment.” Virginia Journal of Law and Technology 8, no. 2: 1–57. Blackburn, S. 2013. “Proud of Our Commitment to Privacy in Canada.” TELUS Blog, 7 August. https://web.archive.org/web/20160827024537/http://blog.telus.com/public -policy/proud-of-our-commitment-to-privacy-in-canada/. Blais, J.-P. 2016. “Getting Ahead of the Curve – Jean-Pierre Blais to the Canadian Chapter of the International Institute of Communications.” 16 November. https://web.archive.org/web/20161117154401/http://news.gc .ca/web/article-en.do?nid=1154239. Bodnar, C. 2007. “The Vancouver Community Network, Social Investing and Public Good Models of ict Development.” The Journal of Community Informatics 3, no. 4. http://ci-journal.net/index.php/ciej/article/view/307. Boutilier, A. 2015. “Internal rcmp Documents Show No Significant Problems with Court Decision Requiring Police to Get a Warrant to Obtain Telecom Customers’ Data.” Toronto Star, 12 July. http://www.thestar.com/news/canada/2015/07/12/halt-to-warrantless-disclosures-not-hindering-rcmp-say-documents.html. – 2019. “Class-Action Lawsuit Alleging Bell Privacy Violations Moves Ahead.” Toronto Star, 14 May. https://www.thestar.com/politics/federal/2019/05/14/class-action-lawsuitalleging-bell-privacy-violations-moves-ahead.html. Bowker, G.C., and S.L. Star. 1999. Sorting Things Out: Classification and Its Consequences. Cambridge, ma: mit Press.

196

References

Bradshaw, J., and C. Dobby. 2015. “Bell Media President Kevin Crull Ousted over Journalistic Meddling.” Globe and Mail, 9 April. https://www.the globeandmail.com/report-on-business/bell-announces-departure-ofmedia-head-kevin-crull-over-journalistic-meddling/article23864190/. Braga, M. 2016. “Canadian Telecoms Push Back on Proposed Police Powers.” CBC News, 20 December. http://www.cbc.ca/news/technology/rogers-teksavvy-itac-cwta-bill-c51-national-security-1.3903930. – 2018. “How and When Rogers, Telus and Bell Sell Your Location to Third-Party Companies.” CBC News, 18 May. http://www.cbc.ca/news/technology/rogers-bell-telus-enstream-locationdata-sharing-securus-1.4666739. Braithwaite, J. 2008. Regulatory Capitalism: How It Works, Ideas for Making It Work Better. Cheltenham, uk: Edward Elgar Publishing. Braithwaite, J., and P. Drahos. 2000. Global Business Regulation. Cambridge: Cambridge University Press. Bratton, B.H. 2015. The Stack: On Software and Sovereignty. Cambridge, ma: mit Press. Brodkin, J. 2017. “1 Million nyc Homes Can’t Get Verizon FiOS, so the City Just Sued Verizon.” Ars Technica, 13 March. https://arstechnica.com/techpolicy/2017/03/nyc-sues-verizon-alleges-failure-to-complete-citywide-fiberrollout/. Bronskill, J. 2014. “Feds Were Worried as Telecom Firm Planned to Go Public on Police Access to Canadians’ Phone Calls and Emails, Memo Shows.” National Post, 1 December. http://news.nationalpost.com/news/canada/feds-were-worried-as-telecomfirms-planned-to-go-public-on-police-access-to-canadians-phone-callsand-emails-memo-shows. – 2015. “Feds Ponder Warrantless Police Access to Internet Subscriber Info: Chiefs.” Canadian Press, 24 August. http://www.nationalnewswatch.com/2015/08/24/feds-ponder-warrantlesspolice-access-to-internet-subscriber-info-chiefs/#.Vds3XqNza73. Brousseau, E., M. Marzouki, and C. Méadel, eds. 2012. Governance, Regulation and Powers on the Internet. Cambridge: Cambridge University Press. Brown, J. 2009. “How Bell Canada Is Keeping Tabs on Telecom Fraud.” Canadian Security, 10 November. Brown, W. 2015. Undoing the Demos: Neoliberalism’s Stealth Revolution. New York: Zone Books. btlr. 2020. “Canada’s Communications Future: Time to Act (Final Report of the Broadcasting and Telecommunications Legislative Review Panel).”

References

197

29 January. http://www.ic.gc.ca/eic/site/110.nsf/eng/00012.html #Toc26977844. Bull, S. 2017. “Addressing the Myths: SuperNet & Rural Broadband Internet.” 16 March. http://www.vanhorneinstitute.com/wp-content/uploads /2017/04/Stephen-Bull-2017_03_16_SuperNet_DigitalFutures_Final.pdf. Burdick, C.K. 1911. “The Origin of the Peculiar Duties of Public Service Companies. Part I.” Columbia Law Review 11, no. 6: 514–31. Burns, J. 1969. “58 Wiretaps Last Year, Mackey Says.” Globe and Mail, 27 June. Burris, S., P. Drahos, and C. Shearing. 2005. “Nodal Governance.” Australian Journal of Legal Philosophy 30: 30–58. Caldwell, R. 2007. “Agency and Change: Re-Evaluating Foucault’s Legacy.” Organization 14, no. 6: 769–91. Calhoun, C. 1998. “The Public Good as a Social and Cultural Project.” In Private Action and the Public Good, edited by W.W. Powell and E.S. Clemens, 20–35. New Haven, ct: Yale University Press. Canada NewsWire. 2008. “Bell E-Learning Centre Officially Opens – Connecting Rural Albertans to e-Learning Opportunities through the SuperNet.” 9 October. Canadian Committee on Corrections. 1969. Recommendations of the Canadian Committee on Corrections [Ouimet Report]. Ottawa: The Queen’s Printer. http://johnhoward.ca/wp-content/uploads/2016/12/1969-HV-8395-A6C33-1969-Ouimet.pdf (accessed 16 September 2020). Canadian Independent Telephone Association. 1911. “Sixth Annual Convention – Official Report.” Canadian Municipal Journal 7, no. 12: 484–7. CA*net Institute. 2001. “A Nation Goes Online: Canada’s Internet History.” https://www.canarie.ca/wpdm-package/publication-a-nation-goesonline/?wpdmdl=6973 (accessed 16 September 2020). Carpenter, B.E., ed. 1996. “rfc 1958: Architectural Principles of the Internet.” https://tools.ietf.org/html/rfc1958 (accessed 16 September 2020). Cashman, T. 1972. Singing Wires: The Telephone in Alberta. Edmonton: Alberta Government Telephone. Castells, M. 2000. The Rise of the Network Society. 2nd ed. Oxford, uk: Blackwell. Cave, M. 2006. “Encouraging Infrastructure Competition via the Ladder of Investment.” Telecommunications Policy 30, no. 3: 223–37. Cavoukian, A. 2011. “Lawful Access.” Globe and Mail, 6 December. https://www .theglobeandmail.com/opinion/dec-6-letters-to-the-editor/article4180180/.

198

References

News. 2005. “Telus Cuts Subscriber Access to Pro-Union Website.” CBC News, 24 July. https://www.cbc.ca/news/canada/telus-cuts-subscriberaccess-to-pro-union-website-1.531166. – 2009. “isps Must Help Police Snoop on Internet under New Bill.” CBC News, 18 June. http://www.cbc.ca/news/technology/isps-must-help-policesnoop-on-internet-under-new-bill-1.817756. – 2013. “Some TekSavvy Internet Customers Upset by Long Service Outages.” CBC News, 31 August. http://www.cbc.ca/1.1309647. CBC Radio. 2019. “Tech, High-Speed Internet Help Today’s Ranchers Thrive.” Spark, 1 February. https://www.cbc.ca/radio/spark/spark-424-1.4995529/ tech-high-speed-internet-help-today-s-ranchers-thrive-1.4995544. ccaice. 2005. “Canadian Coalition Against Internet Child Exploitation Releases National Action Plan.” 11 May. https://www.cybertip.ca/app/en /media_release_ccaice_action_plan. Centre for Public Involvement. 2014. “Citizen Planning Circle: Effective Use of Rural Broadband – Final Report.” http://www.oldsinstitute.com/wpcontent/uploads/2015/02/CPC-final-report-Final-compressed.pdf (accessed 16 September 2020). Chase, S., and I. Marlow. 2011. “Harper Steps into Web Dispute.” Globe and Mail, 1 February. http://www.theglobeandmail.com/technology/technews/harper-steps-into-web-dispute/article565219/. Chhabra, S. 2018a. “ised Minister Responds to Anti-Piracy Group, Defends Net Neutrality.” MobileSyrup, 29 January. https://mobilesyrup.com/2018/01/29/ised-minister-responds-anti-piracygroup-defends-net-neutrality/. – 2018b. “Federal Government Launches ‘Connecting Families’ Internet Program.” MobileSyrup, 16 November. https://mobilesyrup.com/2018/11 /16/ised-connecting-familis-rogers-low-cost-internet/. cira. n.d. “Canada’s Internet Infrastructure: Made-in-Canada Internet Exchange Points (ixps).” Canadian Internet Registration Authority (cira). https://cira.ca/canada%E2%80%99s-internet-infrastructure-madecanada-internet-exchange-points-ixps (accessed 16 September 2020). City of Calgary. 2014. “Final Submissions of the City of Calgary.” Telecom Notice of Consultation crtc 2013-551: Review of Wholesale Services and Associated Policies. 19 December. https://services.crtc.gc.ca/pub/Liste InterventionList/Documents.aspx?ID=212360&Lang=e. City of Kamloops. 2015. “Kamloops Community Network.” https://web.archive.org/web/20170719084853/http://www.kamloops.ca/it/ kcn.shtml. Clement, A., M. Gurstein, G. Longford, M. Moll, and L.R. Shade, eds. 2012. CBC

References

199

Connecting Canadians: Investigations in Community Informatics. Edmonton: au Press. https://www.aupress.ca/books/120193-connecting-canadians/ (accessed 16 September 2020). Clement, A., and J.A. Obar. 2016. “Keeping Internet Users in the Know or in the Dark: An Analysis of the Data Privacy Transparency of Canadian Internet Carriers.” Journal of Information Policy 6: 294–331. cmcrp. 2019. “Media and Internet Concentration in Canada Report, 1984–2018.” Canadian Media Concentration Research Project. http://www.cmcrp.org/media-and-internet-concentration-in-canada-19842018/ (accessed 16 September 2020). cnoc. 2013. “Part 1 Application Requesting Relief to Improve the Quality of Wholesale High-Speed Access Services Provided by Cable Carriers.” crtc. https://services.crtc.gc.ca/pub/TransferToWeb/2013/8660-C182201313113.zip. Collie, D. 2015. “o-net Could Become a Nationwide Internet, Phone, TV Service Provider.” Mountain View Gazette, 10 November. https://web .archive.org/web/20160114105637/http://www.mountainviewgazette.ca /article/ONET-could-become-nationwide-20151110. – 2020. “Town Calls on Olds Institute to Repay $14 Million Loan.” Mountain View Today, 27 May. https://www.mountainviewtoday.ca/oldsnews/town-calls-on-olds-institute-to-repay-14-million-loan-2375717. Collison, M. 2003. “Alberta SuperNet Connects: Big Sky Broadband.” Summit, Fall. http://www.summitconnects.com/Articles_Columns/PDF _Documents/200310_01.pdf. Competition Bureau. 2019. “Delivering Choice: A Study of Competition in Canada’s Broadband Industry.” https://www.competitionbureau.gc.ca /eic/site/cb-bc.nsf/eng/04470.html (accessed 16 September 2020). Copyright Board of Canada. 1999. “Phase I – Legal Issues, Board File: Public Performance of Musical Works 1996, 1997, 1998 (the ‘Tariff 22 Decision’).” https://web.archive.org/web/20200220014747/http://www.cbcda.gc.ca/decisions/1999/19991027-m-b.pdf. Cornfield, D.A. 1967. “The Right to Privacy in Canada.” Faculty of Law Review (University of Toronto) 25: 103–20. Cox, J. 2019. “Hundreds of Bounty Hunters Had Access to AT&T, T-Mobile, and Sprint Customer Location Data for Years.” Vice: Motherboard, 6 February. https://motherboard.vice.com/en_us/article/43z3dn/hundreds-bountyhunters-att-tmobile-sprint-customer-location-data-years. cpac. 2010. 2010 Canadian Telecom Summit – Regulatory Blockbuster. https://www.cpac.ca/en/programs/public-record/episodes/16021459/ (accessed 16 September 2020).

200

References

– 2012. 2012 Canadian Telecom Summit – Regulatory Blockbuster. https://www.cpac.ca/en/programs/public-record/episodes/18503327/ (accessed 16 September 2020). – 2013a. IIC Conference: Consumer Policy Meets Consumer Populism. http://www.cpac.ca/en/programs/public-record/episodes/28823095/ (accessed 16 September 2020). – 2013b. 2013 Canadian Telecom Summit – The Regulatory Blockbuster. http://www.cpac.ca/en/programs/public-record/episodes/24774817 (accessed 16 September 2020). Crang, M., T. Crosbie, and S. Graham. 2006. “Variable Geometries of Connection: Urban Digital Divides and the Uses of Information Technology.” Urban Studies 43, no. 13: 2551–70. Crawford, S. 2013. Captive Audience: The Telecom Industry and Monopoly Power in the New Gilded Age. New Haven, ct: Yale University Press. – 2019. Fiber: The Coming Tech Revolution — And Why America Might Miss It. New Haven, ct: Yale University Press. Cribb, R., and M. Greenblatt. 2015. “Advanced Encryption and SearchWarrant Requirements Have Allowed Child Molesters, Drug Dealers and Organized Crime to Hide Their Crimes from Police.” Toronto Star, 5 November. http://www.thestar.com/news/world/2015/11/05/advancedencryption-privacy-measures-hinder-police-in-digital-age.html. Crouch, C. 2011. The Strange Non-Death of Neo-Liberalism. Cambridge: Polity Press. crtc. 1994. “Telecom Decision crtc 94-19.” 16 September. http://www.crtc .gc.ca/eng/archive/1994/dt94-19.htm. – 1995. Competition and Culture on Canada’s Information Highway: Managing the Realities of Transition. Ottawa: Canadian Radio-Television and Telecommunication Commission. https://web.archive.org/web /20010920181413/http://www.crtc.gc.ca/eng/HIGHWAY/HWY9505.HTM. – 1997. “Telecom Decision crtc 97-8.” 1 May. http://www.crtc.gc.ca/eng/archive/1997/dt97-8.htm. – 1999. “Telecom Decision crtc 99-11.” 14 September. http://www.crtc.gc.ca/eng/archive/1999/DT99-11.htm. – 2001. “Order 2001-184.” 1 March. http://www.crtc.gc.ca/eng/archive/2001/O2001-184.htm. – 2002. “Telecom Decision 2002-34.” 30 May. http://www.crtc.gc.ca/eng/archive/2002/dt2002-34.htm. – 2005. “Telecom Decision crtc 2005-20.” 31 March. http://www.crtc.gc.ca/eng/archive/2005/dt2005-20.htm. – 2006. “Telecom Decision crtc 2006-14.” 29 March. http://www.crtc.gc.ca/eng/archive/2006/dt2006-14.htm.

References

201

– 2008. “Telecom Decision crtc 2008-108.” 20 November. https://crtc.gc.ca/eng/archive/2008/dt2008-108.htm. – 2009. “Telecom Regulatory Policy crtc 2009-657.” 21 October. http://www.crtc.gc.ca/eng/archive/2009/2009-657.htm. – 2010a. “Telecom Regulatory Policy crtc 2010-632.” 30 August. http://www.crtc.gc.ca/eng/archive/2010/2010-632.htm. – 2010b. “Telecom Decision crtc 2010-900.” 2 December. https://crtc.gc.ca/eng/archive/2010/2010-900.htm. – 2011a. “Telecom Decision crtc 2011-44.” 25 January. http://www.crtc.gc.ca/eng/archive/2011/2011-44.htm. – 2011b. “Broadcasting and Telecom Regulatory Policy crtc 2011-512.” 19 August. http://www.crtc.gc.ca/eng/archive/2011/2011-512.htm#b1. – 2014a. “Telecom Decision crtc 2014-398.” 31 July. http://www.crtc.gc.ca/eng/archive/2014/2014-398.htm. – 2014b. “Transcript, Hearing 26 November 2014, Review of Wholesale Service and Associated Policies.” 26 November. http://www.crtc.gc.ca/eng/transcripts/2014/tt1126.htm. – 2014c. “Transcript, Hearing 27 November 2014, Review of Wholesale Service and Associated Policies.” 27 November. http://www.crtc.gc.ca/eng/transcripts/2014/tt1127.htm. – 2015a. “Broadcasting and Telecom Decision crtc 2015-26.” 29 January. http://www.crtc.gc.ca/eng/archive/2015/2015-26.htm. – 2015b. “Telecom Decision crtc 2015-40.” 12 February. http://www.crtc.gc.ca/eng/archive/2015/2015-40.htm. – 2015c. “Telecom Notice of Consultation crtc 2015-134.” 9 April. http://www.crtc.gc.ca/eng/archive/2015/2015-134.htm. – 2015d. “Telecom Regulatory Policy crtc 2015-177.” 5 May. http://www.crtc.gc.ca/eng/archive/2015/2015-177.htm. – 2015e. “Telecom Regulatory Policy 2015-326.” 22 July. http://www.crtc.gc.ca/eng/archive/2015/2015-326.htm. – 2015f. “Broadcasting Regulatory Policy crtc 2015-355 and Broadcasting Order crtc 2015-356.” 6 August. https://crtc.gc.ca/eng/archive/2015/2015355.htm. – 2016a. “Transcript, Hearing April 18, 2016.” 18 April. http://www.crtc.gc.ca/eng/transcripts/2016/tt0418.htm. – 2016b. “Transcript, Hearing November 1, 2016.” 1 November. http://www.crtc.gc.ca/eng/transcripts/2016/tt1101.htm. – 2017a. “Strengthening Net Neutrality in Canada.” 20 April. https://web .archive.org/web/20170505160442/http://crtc.gc.ca/eng/internet/diff.htm. – 2017b. “Telecom Regulatory Policy crtc 2017-104.” 20 April. http://crtc.gc.ca/eng/archive/2017/2017-104.htm.

202

References

– 2018. “Telecom Regulatory Policy crtc 2018-123.” 13 April. https://crtc.gc.ca/eng/archive/2018/2018-123.htm. – 2019a. “Telecom Notice of Consultation crtc 2019-57.” 28 February. https://crtc.gc.ca/eng/archive/2019/2019-57.htm. – 2019b. “Telecom Regulatory Policy 2019-269.” 31 July. https://crtc.gc.ca/eng/archive/2019/2019-269.htm. – 2019c. “Telecom Notice of Consultation crtc 2019-420.” 16 December. https://crtc.gc.ca/eng/archive/2019/2019-420.htm. – 2019d. “Telecom Regulatory Policy crtc 2019-190.” 3 June. https://crtc.gc.ca/eng/archive/2019/2019-190.htm. – 2020. “Communications Monitoring Report 2019.” https://crtc.gc.ca/pubs/cmr2019-en.pdf (accessed 16 September 2020). csec. 2011. “cascade: Joint Cyber Sensor Architecture.” https://www.documentcloud.org/documents/1690204-cascade2011.html#document/p2. csis. 2012. “All Briefing Notes to the Director and/or to the Minister Concerning ‘Lawful Access’ Legislation for the Period September 2011 to the Present. Cabinet Confidences Should Be Excluded.” Access to Information Act Request A-2011-114. cstac. 2013. “Security Best Practices for Canadian Telecommunications Service Providers (tsps).” 6 November. http://www.ic.gc.ca/eic/site/smtgst.nsf/eng/sf10719.html. Cybertip.ca. 2006. “isps and Tipline Step Up Battle Against Internet Child Exploitation.” 17 December. http://web.archive.org/web/20061217184200 /http://www.cybertip.ca/en/cybertip/cleanfeed_canada. Dagger, F. 1910. “Is the Telephone a Natural Monopoly?” Canadian Municipal Journal 6, no. 1: 21–2. – 1915. “Public Ownership of Telephones in Canada.” Canadian Municipal Journal 11, no. 9: 308–9. Davies, W. 2014. The Limits of Neoliberalism: Authority, Sovereignty and the Logic of Competition. Thousand Oaks, ca: sage. Dean, M. 2007. Governing Societies: Political Perspectives on Domestic and International Rule. New York: Open University Press. Dean, M., and K. Villadsen. 2016. State Phobia and Civil Society: The Political Legacy of Michel Foucault. Stanford, ca: Stanford University Press. DeNardis, L. 2009. Protocol Politics: The Globalization of Internet Governance. Cambridge, ma: The mit Press. – 2014. The Global War for Internet Governance. New Haven, ct: Yale University Press. Denton, T. 2013. “Evolution of Canadian Telcom Policy 2008-2013.” Timothy

References

203

Denton’s Blog, 1 October. http://www.tmdenton.com/index.php/easyblog /entry/evolution-of-canadian-telcom-policy-2008-2013. – 2015. “The Klass Decision: The crtc Finally Adopts Internet Thinking.” Timothy Denton’s Blog, 19 February. http://www.tmdenton.com/index.php /easyblog/entry/the-klass-decision-the-crtc-finally-adopts-internet-thinking. – 2017. “Telecom Regulatory Policy crtc 2017-104 Framework for Assessing the Differential Pricing Practices of Internet Service Providers.” Timothy Denton’s Blog, 22 April. https://tmdenton.com/index.php/easyblog /entry/telecom-regulatory-policy-crtc-2017-104-framework-for-assessingthe-differential-pricing-practices-of-internet-service-providers. Dewey, J. 1927. The Public and Its Problems. New York: Holt. Dickson, F. 2015. “Secure Pipes: Changing the Expectation of Your Internet Service Providers.” Level 3, January. https://web.archive.org/web /20160322005112/http://securitysolutions.level3demandcenter.com/img/se curity_secure_pipes_frost_whitepaper.pdf. Dijk, Jan A.G.M. van. 2006. “Digital Divide Research, Achievements and Shortcomings.” Poetics 34, no. 4–5: 221–35. Distributel. 2013. “Motion Record of Distributel Communications Limited.” http://www.scribd.com/doc/124826637/Motion-Record-of-DistributelCommunications-Limited (accessed 16 September 2020). Dobby, C. 2012. “TekSavvy Illegal Downloading: Judge Awards More Time to Warn Clients.” Financial Post, 17 December. http://business.financial post.com/2012/12/17/judge-gives-teksavvy-more-time-to-warn-customersof-illegal-downloading-copyright-case/. – 2014. “Telus Warns against ‘Artificial’ Competition within Industry.” Globe and Mail, 30 September. http://www.theglobeandmail.com/report-onbusiness/telus-warns-against-artificial-competition-within-industry /article20871170/. – 2016. “Ontario Court Rules Police Orders Breached Cellphone Users’ Charter Rights.” Globe and Mail, 14 January. http://www.theglobeandmail .com/report-on-business/industry-news/the-law-page/court-sides-withtelecoms-in-landmark-cellphone-privacy-case/article28180968/. Dobson, C. 2017. “Regional Broadband Preparedness Study: Municipal & Regional Opportunities & Options in the caep Region.” 31 August. http://clearwater.municipalwebsites.ca/Editor/images/Documents /COUNCIL/Regional_BB_CAEP_Study_Sep_2017.pdf. Domingues, J.M. 1995. Sociological Theory and Collective Subjectivity. London: Macmillan. Douglas, M. 1966. Purity and Danger: An Analysis of the Concepts of Pollution and Taboo. London: Routledge and Kegan Paul.

204

References

Duckett, C. 2018. “Telcos Flailing against Tide to Turn into Dumb Pipes.” ZDNet, 21 October. https://www.zdnet.com/article/telcos-flailing-againsttide-to-turn-into-dumb-pipes/. Edwards, L. 2011. “Role and Responsibility of the Internet Intermediaries in the Field of Copyright and Related Rights.” WIPO. http://www.wipo.int /export/sites/www/copyright/en/doc/role_and_responsibility_of_the _internet_intermediaries_final.pdf (accessed 16 September 2020). eorn. n.d. “Open Access: A Competitive Internet Marketplace.” https://web.archive.org/web/20160421053853/https://www.eorn.ca/en/reso urces/fact-sheets/20130513EN_EORN_FactSheet_EN_OpenAccess _Web.pdf. Everest, P. 2014. “Mapping the Future of Rural Broadband.” Olds Albertan, 16 April. https://www.mountainviewtoday.ca/olds-news/mapping-the-futureof-rural-broadband-1810052. ey. 2016a. “Jurisdictional Review of Plans in Select Emerging and Developed Markets.” Nova Scotia Department of Business. https://beta.nova scotia.ca/sites/default/files/documents/1-697/nova-scotia-rural-internetjurisdictional-review-en.pdf (accessed 16 September 2020). – 2016b. “Review of Alternatives for Rural High Speed Internet.” Nova Scotia Department of Business. https://beta.novascotia.ca/sites/default /files/documents/1-696/review-alternatives-rural-high-speed-interneten.pdf (accessed 16 September 2020). Fair Play Canada. 2018. “Fair Play Canada.” https://www.fairplaycanada.com/. fcc. 2015. “Open Internet.” Federal Communications Commission. https://web.archive.org/web/20160118173655/https://www.fcc.gov/general /open-internet. – 2020. “Notice of Apparent Liability for Forfeiture and Admonishment in the Matter of T-Mobile usa, Inc.” Federal Communications Commission. https://docs.fcc.gov/public/attachments/FCC-20-27A1.pdf (accessed 16 September 2020). Federation of Canadian Municipalities. 2009. “Dealing with Telecom Companies: Protecting Municipal Rights of Way.” https://www.fcm.ca /Documents/reports/Dealing_with_Telecom_Companies_Protecting _Municipal_Rights_of_Way_EN.pdf (accessed 16 September 2020). First Mile Connectivity Consortium. 2018. Stories from the First Mile: Digital Technologies in Remote and Rural Indigenous Communities. Fredericton: fmcc. http://firstmile.ca/wp-content/uploads/Stories-from-the-First-MIle2018.pdf (accessed 16 September 2020). Flichy, P. 2007. The Internet Imaginaire. Translated by L. Carey-Libbrecht. Cambridge, ma: mit Press.

References

205

Foucault, M. 1982. “The Subject and Power.” Critical Inquiry 8, no. 4: 777–95. – 1995. Discipline & Punish: The Birth of the Prison. Translated by A. Sheridan. Second Vintage Books Edition. New York: Vintage Books. – 2007. Security, Territory, Population: Lectures at the College de France, 1977–78. Translated by G. Burchell. New York: Palgrave Macmillan. Fraser, D. 2016. “Tower Dump Case Raises Troubling Questions about Law Enforcement and Privacy.” Canadian Privacy Law Blog, 18 January. http://blog.privacylawyer.ca/2016/01/tower-dump-case-raises-troubling.html. Freeze, C., C. Dobby, and J. Wingrove. 2014. “TekSavvy, Rogers Break Silence over Government Requests for Data.” Globe and Mail, 5 June. http://www.theglobeandmail.com/technology/tech-news/teksavvy-opensbooks-on-government-data-requests/article18999107/. Freeze, C., and R. Trichur. 2013. “Wireless Firms Rejected Ottawa’s Changes to Surveillance Rules over Cost Concerns.” Globe and Mail, 16 September. http://www.theglobeandmail.com/technology/mobile/wireless-firms-rejectottawas-changes-to-surveillance-rules-over-cost-concerns/article14363379/. French, R.D. 2007. “Net Neutrality 101.” University of Ottawa Law and Technology Journal 4, no. 1–2: 109–33. Galloway, A.R. 2004. Protocol: How Control Exists after Decentralization. Cambridge, ma: mit Press. Gane, N. 2020. “Competition: A Critical History of a Concept.” Theory, Culture & Society 37, no. 2: 31–59. Gaudrault, M. 2012. “Why We Are Not Opposing Motion on Monday.” DSLReports Forums: TekSavvy, 15 December. http://www.dslreports.com/forum/r27824891-Why-we-are-not-opposingmotion-on-Monday. Geist, M., ed. 2005. Internet and E-Commerce Law in Canada 6, no. 5. https://web.archive.org/web/20180320230726/http://www.macerajarzyna.c om/pages/publications/BMG%20Case%20-%20E-Commerce.pdf. – 2011a. “Canada’s Usage Based Billing Controversy: How to Address the Wholesale and Retail Issues.” Queen’s Law Journal 37, no. 1: 221–56. – 2011b. “Canada’s Net Neutrality Enforcement Failure.” Michael Geist Blog, 8 July. http://www.michaelgeist.ca/content/view/5918/159/. – 2012a. “How the crtc Helped to Put an End to Internet Throttling.” Michael Geist Blog, 2 March. http://www.michaelgeist.ca/2012/03/netthrottling-column/. – 2012b. “How Canada’s Telecom Companies Have Secretly Supported Internet Surveillance Legislation.” Michael Geist Blog, 22 May. http://www.michaelgeist.ca/content/view/6505/135/.

206

References

– 2016a. “The Policy Battle over Information and Digital Policy Regulation: A Canadian Perspective.” Theoretical Inquiries in Law 17, no. 2: 415–49. – 2016b. “Why Your Telecom Must Defend Your Right to Privacy.” The Toronto Star, 25 January. http://www.thestar.com/business/2016/01/25 /why-your-telecom-must-defend-your-right-to-privacy-geist.html. – 2017. “Why Warrantless Access to Internet Information Is Back on the Lawmaking Agenda.” Globe and Mail, 3 April. http://www.theglobeand mail.com/report-on-business/rob-commentary/why-warrantless-access-tointernet-information-is-back-on-the-lawmaking-agenda/article34567424/. – 2019. “Canadian Copyright Website Blocking Underway As TekSavvy Appeals Federal Court Ruling.” Michael Geist Blog, 26 November. http://www.michaelgeist.ca/2019/11/canadian-copyright-website-blockingunderway-as-teksavvy-appeals-federal-court-ruling/. Geser, H. 1992. “Towards an Interaction Theory of Organizational Actors.” Organization Studies 13, no. 3: 429–51. Geuss, R. 2009. Public Goods, Private Goods. Princeton, nj: Princeton University Press. Giddens, A. 1984. The Constitution of Society: Outline of the Theory of Structuration. Cambridge: Polity Press. Gignac, T. 2003. “Learning Live: The SuperNet’s High-Speed Promise.” Calgary Herald, 10 March. – 2010. “SuperNet’s Digital Dream Still out of Reach for Rural Alberta.” Calgary Herald, 22 August. https://web.archive.org/web/20131113082105 /http://www2.canada.com/calgaryherald/news/story.html?id=4bac9195673f-4dbf-b817-6fc5750496de&p=1. Gilbert, M. 2014. Joint Commitment: How We Make the Social World. Oxford, uk: Oxford University Press. Gillespie, T. 2006. “Engineering a Principle: ‘End-to-End’ in the Design of the Internet.” Social Studies of Science 36, no. 3: 427–57. Globe and Mail. 1977. “rcmp Bypassed Bell, Pried Open Terminal Box to Tap Apartment Phone.” Globe and Mail, 19 February. Goldsmith, J.L., and T. Wu. 2006. Who Controls the Internet?: Illusions of a Borderless World. Oxford, uk: Oxford University Press. Gómez-Barroso, J.L., and C. Feijóo. 2010. “A Conceptual Framework for Public-Private Interplay in the Telecommunications Sector.” Telecommunications Policy 34, no. 9: 487–95. Government of Alberta, Bell Canada, and Axia SuperNet Ltd. 2005. “Amended and Restated SuperNet Master Agreement.” Government of Canada. 1880. Bell Telephone Company of Canada Act. – 1882. An Act to Amend Act Incorporating “The Bell Telephone Company of Canada.”

References

207

– 1993. Telecommunications Act. http://laws-lois.justice.gc.ca/eng/acts/T-3.4/. – 1996. “Convergence Policy Statement.” 6 August. https://web.archive.org/web /20130531101136/https://www.ic.gc.ca/eic/site/smt-gst.nsf/eng/sf05265.html. – 2002. “Lawful Access – Consultation Document.” 25 August. http://www.justice.gc.ca/eng/cons/la-al/consult.html. – 2006. “Order Issuing a Direction to the crtc on Implementing the Canadian Telecommunications Policy Objectives.” 14 December. http://lawslois.justice.gc.ca/eng/regulations/SOR-2006-355/page-1.html. – 2013. “Harper Government Puts Consumers First in Telecommunications Plan.” 7 March. https://www.canada.ca/en/news/archive/2013/03/harpergovernment-puts-consumers-first-telecommunications-plan.html. – 2016. “Statement by the Government of Canada on Bell Canada Petition of crtc Wholesale Decision.” 11 May. https://www.canada.ca/en /innovation-science-economic-development/news/2016/05/statement-bythe-government-of-canada-on-bell-canada-petition-of-crtc-wholesaledecision.html. – 2018. “Cyber Security.” 4 October. https://web.archive.org/web /20190201153616/https://www.canada.ca/en/services/defence/cyber security.html. – 2019. “Building a Better Canada: Universal High-Speed Internet.” 19 March. https://web.archive.org/web/20190324220447/https://www.budget .gc.ca/2019/docs/nrc/infrastructure-infrastructures-internet-en.html. – 2020. “Order Issuing a Direction to the crtc on Implementing the Canadian Telecommunications Policy Objectives to Promote Competition, Affordability, Consumer Interests and Innovation.” 4 May. https://lawslois.justice.gc.ca/PDF/SOR-2019-227.pdf. Graham, G. 2011. “Towards a National Strategy for Digital Inclusion: Addressing Social and Economic Disadvantage in an Internet Economy.” In The Internet Tree: The State of Telecom Policy in Canada 3.0, edited by M. Moll and L.R. Shade, 29–36. Ottawa: Canadian Centre for Policy Alternatives. http://www.policyalternatives.ca/sites/default/files/uploads /publications/National%20Office/2011/06/Internet_Tree_0.pdf. Gurstein, M. 2003. “Effective Use: A Community Informatics Strategy beyond the Digital Divide.” First Monday 8, no. 12. http://firstmonday.org /ojs/index.php/fm/article/view/1107. Guttentag, M. 2018. “A Light in Digital Darkness: Public Broadband after Tennessee v. FCC.” Yale Journal of Law & Technology 20, no. 1: 311–75. Hafner, K., and M. Lyon. 1996. Where Wizards Stay up Late. New York: Simon & Schuster. Haggart, B. 2014. Copyfight: The Global Politics of Digital Copyright Reform. Toronto: University of Toronto Press.

208

References

Haggerty, K.D. 2006. “Tear down the Walls: On Demolishing the Panopticon.” In Theorizing Surveillance: The Panopticon and Beyond, edited by D. Lyon, 23–45. London: Routledge. Hales, D. 2014. “ESecurity Portfolio: Overview, Analysis of Value Added, and Way Ahead.” Defence Research and Development Canada. https://apps .dtic.mil/dtic/tr/fulltext/u2/1003640.pdf (accessed 16 September 2020). Hales, R.L. 2019. “Connected Coast Comes To Cortes.” Cortes Radio, 22 June. https://web.archive.org/web/20190626224703/https://cortesradio.ca /connected-coast-comes-to-cortes/. Hansen, L., and H. Nissenbaum. 2009. “Digital Disaster, Cyber Security, and the Copenhagen School.” International Studies Quarterly 53, no. 4: 1155–75. Harvey, P. 2018. “Dumb Networks, Not Just Dumb Pipes.” Light Reading, 9 August. https://www.lightreading.com/optical-ip/100g-ethernet/dumbnetworks-not-just-dumb-pipes/a/d-id/745234. Hathout, A. 2018. “‘An Existential Crisis’: cnoc Files Part 1 on Disaggregated Criteria.” Wire Report, 7 November. https://thewirereport.ca/2018/11/07 /an-existential-crisis-cnoc-files-part-1-on-disaggregated-criteria/. Hecht, J. 2015. “Net Neutrality’s Technical Troubles.” IEEE Spectrum, 12 February. http://spectrum.ieee.org/telecom/internet/net-neutralitys-technicaltroubles/. Higginbotham, S. 2014a. “There’s Something Rotten in the State of Online Video Streaming, and the Data Is Starting to Emerge.” Gigaom (blog), 6 February. http://gigaom.com/2014/02/06/theres-something-rotten-in-thestate-of-online-video-streaming-and-the-data-is-starting-to-emerge/. – 2014b. “Hey fcc, Netflix Thinks Peering Should Be a Net Neutrality Issue Too.” Gigaom (blog), 20 March. http://gigaom.com/2014/03/20/heyfcc-netflix-things-peering-should-be-a-net-neutrality-issue-too/. Hirsh, J. 2020. “Future Fibre: Municipal Internet.” Metaviews, 25 February. https://news.metaviews.ca/future-fibre-municipal-internet/. Horten, Monica. 2013. A Copyright Masquerade: How Copyright Lobbying Threatens Online Freedoms. London: Zed Books. House of Commons of Canada. 2012. “Bill C-30, First Reading (41-1).” 14 February. http://www.parl.gc.ca/HousePublications/Publication.aspx ?DocId=5380965&File=4. House of Commons Standing Committee on Industry, Science and Technology. 2018. “Transcript, Standing Committee on Industry, Science and Technology, Wednesday, February 26, 2018.” 26 February. https://www.our commons.ca/DocumentViewer/en/42-1/INDU/meeting-128/evidence. Howard, P.N., L. Busch, and P. Sheets. 2010. “Comparing Digital Divides: Internet Access and Social Inequality in Canada and the United States.”

References

209

Canadian Journal of Communication 35, no. 1. http://www.cjc-online.ca /index.php/journal/article/view/2192. Hu, T.-H. 2015. A Prehistory of the Cloud. Cambridge, ma: mit Press. Hunt, A., ed. 2014. In Their Own Words: The Story of National Capital FreeNet, 3rd ed. http://web.ncf.ca/fn352/InTheirOwnWords.pdf (accessed 16 September 2020). Hyslop, J. 1984. The Inka Road System. Orlando, fl: Academic Press. Ibbitson, J. 2012. “How the Toews-Sponsored Internet Surveillance Bill Quietly Died.” Globe and Mail, 15 May. http://www.theglobeandmail.com /news/politics/how-the-toews-sponsored-internet-surveillance-bill-quietlydied/article4179310/. Industry Canada. 1994. “The Canadian Information Highway.” Cat. No. C2229/1994E. – 2013. “Framework for Mandatory Roaming and Antenna Tower and Site Sharing – Spectrum Management and Telecommunications.” https://www.ic .gc.ca/eic/site/smt-gst.nsf/eng/h_sf10290.html (accessed 16 September 2020). – 2014a. “Lawful Intercept Alterations Re: Impacts to Wireless and Internet Providers – January 1, 2010 to March 1, 2013.” Access to Information Act Request A-2012-00175. – 2014b. “Documents and Correspondence Pertaining to the Founding of the Canadian Security Telecommunications Advisory Committee (cstac), as Well as Minutes, Meeting Summaries, and Presentations from cstac Meetings, from September 1, 2010 to the Present (July 10, 2013).” Access to Information Act Request A-2013-00188. – 2015. “Transparency Reporting Guidelines.” 30 June. http://www.ic.gc.ca /eic/site/smt-gst.nsf/eng/sf11057.html. Information Highway Advisory Council. 1995. “Connection, Community, Content: The Challenge of the Information Highway.” Ottawa: Industry Canada. http://itlaw.wikia.com/wiki/Connection,_Community,_Content: _The_Challenge_of_the_Information_Highway (accessed 16 September 2020). – 1996. “Building the Information Society: Moving Canada in the 21st Century.” Ottawa: Government of Canada. – 1997. “Preparing Canada for a Digital World.” Ottawa: Industry Canada. ised. 2016. “Any Documents Pertaining to Telecommunication Services Provider Security Best Practices, as Well as Minutes, Meeting Summaries, and Presentations from cstac Meetings, from July, 2013 to the Present (December 2, 2014).” Access to Information Act Request A-2014-00421. – 2019. “CyberSecure Canada.” 22 November. https://www.ic.gc.ca/eic/site /137.nsf/eng/home.

210

References

isp Summit. 2013. Regulatory Panel: What’s Cooking in Ottawa. https://www.you tube.com/watch?v=RN_yB7HAAPg (accessed 16 September 2020). “isps Put the Boots to Telco.” n.d. https://web.archive.org/web /20180119181043/https://www.hackcanada.com/telco/canuck/agt.html (accessed 27 May 2020). Jackson, E. 2016. “crtc Declares High-Speed Internet a Basic Service, Creates $750-Million Fund.” Financial Post, 22 December. https://financialpost .com/technology/crtc-declares-high-speed-internet-a-basic-service-creates750-million-fund. – 2017a. “bce Inc. Receives Final Approval to Buy mts in $3.9-Billion Deal, Xplornet Enters Manitoba Market.” Financial Post, 15 February. http://business.financialpost.com/fp-tech-desk/bce-inc-receives-finalapproval-to-buy-manitoba-telecom-services-in-3-9-billion-deal. – 2017b. “Left in the Digital Dark Ages, Small Town Canada Chases Its Own Gigabit Dream.” Financial Post, 3 March. http://business.financial post.com/fp-tech-desk/left-in-the-digital-dark-ages-small-town-canadachases-its-own-gigabit-dream. – 2018. “Ottawa Orders Regulator to Investigate Sales Practices by Big Telecom Companies.” Financial Post, 14 June. https://business.financialpost .com/telecom/ottawa-orders-crtc-to-investigate-telecoms-for-clearlyinappropriate-high-pressure-sales-tactics. John, R.R. 2010. Network Nation: Inventing American Telecommunications. Cambridge, ma: Belknap Press. Johnston, L., and C.D. Shearing. 2003. Governing Security: Explorations of Policing and Justice: Explorations in Policing and Justice. New York: Routledge. Kaiser, G. 1981. “Competition in Telecommunications: Refusal to Supply Facilities by Regulated Common Carriers.” Ottawa Law Review no. 13: 95–122. Karadeglija, A. 2018. “crtc Rural Broadband Fund Speed Targets Draw Criticism.” Wire Report, 27 September. https://thewirereport.ca/2018/09/27 /crtc-rural-broadband-fund-speed-targets-draw-criticism/. – 2019. “Helix Will Keep Cord-Cutters in Videotron ‘Ecosystem,’ ceo Says.” Wire Report, 27 August. http://thewirereport.ca/2019/08/27/helix-willkeep-cord-cutters-in-videotron-ecosystem-ceo-says/. Katz, M.L., and C. Shapiro. 1994. “Systems Competition and Network Effects.” Journal of Economic Perspectives 8, no. 2: 93–115. Kaul, I., and R.U. Mendoza. 2003. “Advancing the Concept of Public Goods.” In Providing Global Public Goods: Managing Globalization, edited by I. Kaul, P. Conceição, K. Le Goulven, and R.U. Mendoza, 78–111. Oxford, uk: Oxford University Press.

References

211

Kerr, I., and A. Cameron. 2006. “nymity, p2p & isps: Lessons from bmg Canada Inc. v. John Doe.” In Privacy and Technologies of Identity: A CrossDisciplinary Conversation, edited by K.J. Strandburg and D.S. Raicu, 269–94. New York: Springer. Kerr, I., and D. Gilbert. 2004. “The Role of isps in the Investigation of Cybercrime.” In Information Ethics in the Electronic Age: Current Issues in Africa and the World, edited by T. Mendina and J.J. Britz, 163–72. Jefferson, nc: Mcfarland & Company. Klass, B. 2015. “Mobile Wireless in Canada: Policy, Problems, and Progress.” Winnipeg: University of Manitoba. https://mspace.lib.umanitoba.ca /xmlui/handle/1993/30704 (accessed 16 September 2020). Klinenberg, E. 2007. Fighting for Air: The Battle to Control America’s Media. New York: Henry Holt and Company. Knopf, H. 2013. “Mass Copyright Litigation in Canada: Some Observations on the Roles and Responsibilities of isps and Their Customers.” Excess Copyright (blog), 8 January. http://excesscopyright.blogspot.ca/2013/01 /mass-copyright-litigation-in-canada.html. Kolb, A. 2001. “Transport and Communication in the Roman State: The Cursus Publicus.” In Travel and Geography in the Roman Empire, edited by C. Adams and R. Laurence, 95–105. London: Routledge. Kozak, N.I. 2010. “On the Last Mile: The Effects of Telecommunications Regulation and Deregulation in the Rural Western United States and Canada.” San Diego, ca: uc San Diego. http://escholarship.org/uc/item /8nv621sn (accessed 16 September 2020). Krashinsky, S. 2011. “crtc’s Internet Decision ‘Simply Wrong,’ Clement Says.” Globe and Mail, 1 March. http://www.theglobeandmail.com /technology/tech-news/crtcs-internet-decision-simply-wrong-clementsays/article568826/. Kushida, K.E. 2015. “The Politics of Commoditization in Global ict Industries: A Political Economy Explanation of the Rise of Apple, Google, and Industry Disruptors.” Journal of Industry, Competition and Trade 15: 49–67. Kyonka, N. 2012. “cira Calls for More Exchanges to Improve Internet Speed, Costs.” Wire Report, 28 September. http://www.thewirereport.ca /news/2012/09/28/cira-calls-for-more-exchanges-to-improve-internetspeed-costs/25789. Lawford, J., J. Lo, and M. De Santis. 2009. “Staying Neutral: Canadian Consumers and the Fight for Net Neutrality.” Ottawa: Public Interest Advocacy Centre. https://web.archive.org/web/20191008174129 /https://www.piac.ca/wp-content/uploads/2014/11/NN_Report _MASTER_FINAL_website.pdf.

212

References

Lawson, P. 2011. “Bill C-12 and ‘Lawful Authority’ under pipeda.” Slaw, 23 November. http://www.slaw.ca/2011/11/23/bill-c-12-and-lawfulauthority/. Lee-Murphy, M. 2020. “Court Docs Detail Expansion of Canadian WebsiteBlocking Order.” Wire Report, 8 January. https://thewirereport.ca/2020/01 /08/court-docs-detail-expansion-of-canadian-website-blocking-order/. Leiner, B.M., V.G. Cerf, D.D. Clark, R.E. Kahn, L. Kleinrock, D.C. Lynch, J. Postel, L.G. Roberts, and S. Wolff. 2009. “A Brief History of the Internet.” SIGCOMM Computer Communication Review 39, no. 5: 22–31. Lemley, M.A., and L. Lessig. 2001. “The End of End-to-End: Preserving the Architecture of the Internet in the Broadband Era.” UCLA Law Review 48, no. 4: 925–72. Levi-Faur, D. 1999. “The Governance of Competition: The Interplay of Technology, Economics, and Politics in European Union Electricity and Telecom Regimes.” Journal of Public Policy 19, no. 2: 175–207. – 2006. “Regulatory Capitalism: The Dynamics of Change beyond Telecoms and Electricity.” Governance 19, no. 3: 497–525. Lie, E. 2003. “Promoting Broadband: The Case of Canada.” International Telecommunication Union. https://www.itu.int/osg/spu/ni/promote broadband/casestudies/canada.pdf. Ling, J. 2014. “For Canada’s Spies, Your Data Is Just a Phone Call Away.” Vice: Motherboard, 15 May. http://motherboard.vice.com/read/canadian-spiescan-look-at-your-user-data-consequence-free. – 2015. “The rcmp Spent $1.6 Million to Run an Unconstitutional Spying Program.” Vice, 20 January. http://www.vice.com/en_ca/read/the-rcmpspent-16-million-to-run-an-unconstitutional-spying-program-239. Lithwick, D. 2014a. “Legislative Summary of Bill S-4: An Act to Amend the Personal Information Protection and Electronic Documents Act and to Make a Consequential Amendment to Another Act.” Library of Parliament, 11 June. https://lop.parl.ca/sites/PublicWebsite/default/en_CA /ResearchPublications/LegislativeSummaries/412S4E. – 2014b. “R. v. Spencer, Internet Privacy and Parliament.” HillNotes (blog), 25 November. https://hillnotes.wordpress.com/2014/11/25/r-v-spencerinternet-privacy-and-parliament/. Lombardi, R. 2006. “The Rollercoaster Ride of Public Telcos.” Wire Report, 1 December. http://www.thewirereport.ca/news/2006/12/01/the-rollercoasterride-of-public-telcos/11487. Macaulay, T. 2010. “Upstream Intelligence: A New Layer of Cybersecurity.” IAnewsletter. – 2015. “Risk Management and the Internet of Things.” Presented at the jiq

References

213

2015, Quebec City, 9 November. https://www.youtube.com/watch ?v=1H6PnxJ95jM. – 2017. RIoT Control: Understanding and Managing Risks and the Internet of Things. Cambridge, ma: Morgan Kaufmann. MacDonald, N. 1987. “Electronic Surveillance in Crime Detection: An Analysis of Canadian Wiretapping Law.” Dalhousie Law Journal 10: 141–66. MacDougall, R. 2013. The People’s Network: The Political Economy of the Telephone in the Gilded Age. Philadelphia, pa: University of Pennsylvania Press. Malik, A. 2019. “TekSavvy Appeals Federal Court Decision Ordering isps to Block GoldTV.” MobileSyrup, 26 November. https://mobilesyrup.com/2019 /11/26/teksavvy-appeals-federal-court-decision-ordering-isps-block-goldtv/. Mansbridge, J. 1998. “On the Contested Nature of the Public Good.” In Private Action and the Public Good, edited by W.W. Powell and E.S. Clemens, 3–19. New Haven, ct: Yale University Press. Marowits, R. 2007. “Telus Considers Bell Merger.” Toronto Star, 21 June. https://www.thestar.com/business/2007/06/21/telus_considers_bell _merger.html. Martin, M. 1991. Hello, Central?: Gender, Technology, and Culture in the Formation of Telephone Systems. Montreal, qc: McGill-Queen’s University Press. Mathew, A.J. 2014. “Where in the World Is the Internet? Locating Political Power in Internet Infrastructure.” Berkeley, ca: University of California, Berkeley. https://www.ischool.berkeley.edu/sites/default/files/ashwindissertation.pdf (accessed 16 September 2020). – 2016. “The Myth of the Decentralised Internet.” Internet Policy Review 5, no. 3. https://policyreview.info/articles/analysis/myth-decentralisedinternet. Mattern, S. 2015. “Deep Time of Media Infrastructure.” In Signal Traffic: Critical Studies of Media Infrastructures, edited by L. Parks and N. Starosielski, 94–112. Urbana: University of Illinois Press. McChesney, R.W. 2013. Digital Disconnect: How Capitalism Is Turning the Internet against Democracy. New York: The New Press. McDowell, S.D., and C.C. Buchwald. 1997. “Public Interest Groups and the Canadian Information Highway.” Telecommunications Policy 21, no. 8: 709–19. McKelvey, F. 2018. Internet Daemons: Digital Communications Possessed. Minneapolis: University of Minnesota Press. McKenna, A. 2011. “Téléchargement de The Hurt Locker: Des Internautes

214

References

Canadiens Poursuivis.” La Presse, 28 November. https://web.archive.org /web/20150121031546/http://techno.lapresse.ca/nouvelles/internet/201111 /28/01-4472247-telechargement-de-the-hurt-locker-des-internautescanadiens-poursuivis.php. McMahon, D. 2011. “The Canadian Cyber Security Situation in 2011.” https://citizenlab.org/cybernorms2012/cybersecurityfindings.pdf (accessed 16 September 2020). McMahon, D., R. Rohozinski, and Bell Canada. 2013. “The Dark Space Project.” https://web.archive.org/web/20190407223842/http://cradpdf.drdcrddc.gc.ca/PDFS/unc159/p537638_A1b.pdf. McMahon, R., M. McNally, and K. Joseph. 2020. “Shaping ‘Digital Futures’ in Alberta: Community Engagement for Rural Broadband Development.” Canadian Journal of Communication 45, no. 1. https://www.cjconline.ca/index.php/journal/article/view/3527. McNally, M.B., D. Rathi, K. Joseph, J. Evaniew, and A. Adkisson. 2018. “Ongoing Policy, Regulatory, and Competitive Challenges Facing Canada’s Small Internet Service Providers.” Journal of Information Policy 8: 167–98. McRae, B. 2019. “Report to Council.” Village of Gold River, 17 June. http://www.villageofgoldriver.com/municipal-office.html. McTaggart, Craig. 2006. “Was the Internet Ever Neutral?” Arlington, va: 34th Research Conference on Communication, Information and Internet Policy. http://papers.ssrn.com/abstract=2117601 (accessed 16 September 2020). Mediacaster. 2011. “Canadian isps to Deliver Customer Information in Hurt Locker Lawsuit.” Mediacaster Magazine, 9 September. https://web.archive .org/web/20111103084359/http://www.mediacastermagazine.com/news/ca nadian-isps-to-deliver-customer-information-in-hurt-locker-lawsuit /1000575650/. MediaSmarts. n.d. “What We Do.” http://mediasmarts.ca/about-us/what-wedo (accessed 28 May 2020). Meier-Hahn, U. 2016. “Exploring the Regulatory Conditions of Internet Interconnection – A Survey Among Internet Interconnection Professionals.” hiig Discussion Paper Series. Berlin: Alexander von Humboldt Institute for Internet and Society. https://papers.ssrn.com/abstract=2740312 (accessed 16 September 2020). – 2017a. “The Internet Was Built on Trust. But What Does It Run On?” ripe Labs, 5 July. https://labs.ripe.net/Members/uta_meier_hahn/theinternet-was-built-on-trust-but-what-does-it-run-on. – 2017b. “The Secrets of De-Peering.” ripe Labs, 6 August. https://labs.ripe .net/Members/uta_meier_hahn/the-secrets-of-de-peering.

References

215

Melody, W.H. 1987. “Information: An Emerging Dimension of Institutional Analysis.” Journal of Economic Issues 21, no. 3: 1313–39. Middleton, C. 2011. “Structural and Functional Separation in Broadband Networks: An Insufficient Remedy to Competitive Woes in the Candian Broadband Market.” In The Internet Tree: The State of Telecom Policy in Canada 3.0, edited by M. Moll and L.R. Shade, 61–72. Ottawa: Canadian Centre for Policy Alternatives. http://www.policyalternatives.ca/sites /default/files/uploads/publications/National%20Office/2011/06/Internet_ Tree_0.pdf. Miller, C.C. 2014. “Revelations of N.S.A. Spying Cost U.S. Tech Companies.” New York Times, 21 March. http://www.nytimes.com/2014/03/22 /business/fallout-from-snowden-hurting-bottom-line-of-tech-companies .html. Milleron, J.-C. 1972. “Theory of Value with Public Goods: A Survey Article.” Journal of Economic Theory 5, no. 3: 419–77. Mitchell, D. 2007. “Broadband at the Margins: Challenges to Supernet Deployment in Rural and Remote Albertan Communities.” In How Canadians Communicate II: Media, Globalization, and Identity, edited by D. Taras, M. Bakardjieva, and F. Pannekoek, 261–87. Calgary, ab: University of Calgary Press. Moglen, E. 2012. “Freedom to Connect 2012 Keynote: ‘Innovation under Austerity.’” Washington, dc, 22 May. https://www.softwarefreedom.org /events/2012/freedom-to-connect_moglen-keynote-2012.html. Molinaro, D. 2017. “‘In the Field of Espionage, There’s No Such Thing as Peacetime’: The Official Secrets Act and the Picnic Wiretapping Program.” Canadian Historical Review 93, no. 3: 457–82. Morales, F. 2011. “A Telecom Nerve Centre to a Nation.” Globe and Mail, 3 January. https://www.theglobeandmail.com/report-on-business/a-telecomnerve-centre-to-a-nation/article560097/. Morin, S. 2011. “Updated: Business Disclosure of Personal Information to Law Enforcement Agencies: pipeda and the cna Letter of Request Protocol.” Privacy Pages, November. https://web.archive.org/web/201306 10071851/http://www.cba.org/cba/newsletters-sections/pdf/2011-11privacy1.pdf. Morozov, E. 2011. The Net Delusion: The Dark Side of Internet Freedom. New York: PublicAffairs. Morris, I. 2018. “Robot Wars: Telecom’s Looming AI Tussle.” Light Reading, 14 March. https://www.lightreading.com/artificial-intelligence-machinelearning/robot-wars-telecoms-looming-ai-tussle/d/d-id/741400. Mosco, V. 2009. The Political Economy of Communication, 2nd ed. Los Angeles, ca: sage.

216

References

Mueller, M.L. 2004. Ruling the Root: Internet Governance and the Taming of Cyberspace. Cambridge, ma: mit Press. – 2010. Networks and States: The Global Politics of Internet Governance. Cambridge, ma: mit Press. – 2017. Will the Internet Fragment?: Sovereignty, Globalization and Cyberspace. Cambridge, ma: Polity. Murphy, D. 2018. “Cory Hutcheson, Embattled Missouri Sheriff, Pleads Guilty in Deal with Feds.” Riverfront Times, 20 November. https://www.riverfronttimes.com/newsblog/2018/11/20/cory-hutchesonembattled-missouri-sheriff-pleads-guilty-in-deal-with-feds. Mussio, L. 2001. Telecom Nation: Telecommunications, Computers, and Governments in Canada. Montreal, qc: McGill-Queen’s University Press. National Broadband Task Force. 2001. “The New National Dream: Networking the Nation for Broadband Access.” http://publications.gc.ca /collections/Collection/C2-574-2001E.pdf (accessed 16 September 2020). Neocleous, M. 2008. Critique of Security. Edinburgh: Edinburgh University Press. Nevis Consulting Group. 2003. “Summary of Submissions to the Lawful Access Consultation.” 6 August. http://canada.justice.gc.ca/eng/cons/laal/sum-res/index.html. New Age. 2015. “btrc to Take Action against Illegal isp Business.” New Age, 20 March. http://archive.newagebd.net/104521/btrc-to-take-action-againstillegal-isp-business/. Newman, R.A. 2019. The Paradoxes of Network Neutralities. Cambridge, ma: mit Press. Noam, E.M. 1994. “Beyond Liberalization II: The Impending Doom of Common Carriage.” Telecommunications Policy 18, no. 6: 435–52. – 2001. Interconnecting the Network of Networks. Cambridge, ma: mit Press. Norton, W.B. 2010. “Internet Service Providers and Peering.” DrPeering International. http://drpeering.net/white-papers/Internet-ServiceProviders-And-Peering.html (accessed 16 September 2020). Nowak, P. 2008. “ndp to Introduce ‘Net Neutrality’ Private Member’s Bill.” CBC News, 27 May. https://www.cbc.ca/news/technology/ndp-to-introducenet-neutrality-private-member-s-bill-1.701119. – 2009. “Bell and Telus Will Merge within 2 Years, rbc Predicts.” CBC News, 12 August. http://www.cbc.ca/news/technology/bell-and-telus-will-mergewithin-2-years-rbc-predicts-1.787085. – 2017. “Bell’s Fibe Alt TV Is Offside in Several Ways.” Peter Nowak Blog, 15 May. https://web.archive.org/web/20190201024834/http://peternowak.ca /bells-fibe-alt-tv-offside/.

References

217

nsa. 2009. “ST-09-0002 Working Draft.” http://www.guardian.co.uk/world /interactive/2013/jun/27/nsa-inspector-general-report-document-datacollection (accessed 16 September 2020). O’Brien, G. 2013. “Independent isp Distributel Fighting Moviemakers over Customer Privacy.” Cartt.ca, 14 February. https://cartt.ca/independent-ispdistributel-fighting-moviemakers-over-customer-privacy/. oecd. 1980. “oecd Guidelines on the Protection of Privacy and Transborder Flows of Personal Data.” http://www.oecd.org/sti/ieconomy/oecd guidelinesontheprotectionofprivacyandtransborderflowsofpersonal data.htm (accessed 16 September 2020). – 2011. The Role of Internet Intermediaries in Advancing Public Policy Objectives. http://www.oecd.org/sti/interneteconomy/theroleofinternet intermediariesinadvancingpublicpolicyobjectives.htm (accessed 16 September 2020). – 2012. “Report on Experiences with Structural Separation.” http://www.oecd .org/daf/competition/50056685.pdf (accessed 16 September 2020). Oever, Niels ten, and Kathleen Moriarty, eds. 2018. “The Tao of ietf: A Novice’s Guide to the Internet Engineering Task Force.” ietf. http://www.ietf.org/tao.html (accessed 16 September 2020). Office of the Privacy Commissioner. 2014. “Metadata and Privacy – A Technical and Legal Overview.” 30 October. https://www.priv.gc.ca/en/opcactions-and-decisions/research/explore-privacy-research/2014/md_201410/. – 2015a. “Any and All Records Including but Not Limited to Memos, Written Notes, Presentations, Emails, Reports & Correspondence Prepared for or by Privacy Commissioner Daniel Therrien Concerning the opc’s Investigation into the rcmp’s Use of Warrantless Requests for Canadians’ Personal Information and the opc’s Subsequent Report on Same from August 1, 2014 to November 1, 2014.” Access to Information Act Request A-2014-00158. – 2015b. “Results of Commissioner Initiated Investigation into Bell’s Relevant Ads Program.” 7 April. https://www.priv.gc.ca/cf-dc/2015/2015_001 _0407_e.asp. – 2017. “2016 Survey of Canadians on Privacy.” 26 January. https://www.priv .gc.ca/en/opc-actions-and-decisions/research/explore-privacy-research /2016/por_2016_12/. – 2018. “Announcement: opc’s More Proactive Approach to Privacy Protection Unveiled in Departmental Plan Tabled in Parliament Today.” 16 April. https://www.priv.gc.ca/en/opc-news/news-and-announcements/2018 /an_180416_dp/. oicrd. n.d. “Gigabit Broadband in Olds.” Olds Institute. https://www.olds

218

References

institute.com/choose-olds/gigabit-broadband-in-olds/ (accessed 26 January 2019). OpenMedia. 2011. “Stop the Meter on Your Internet Usage.” https://web .archive.org/web/20110217051644/http://www.stopthemeter.ca/. Oremus, W. 2018. “It’s Time to Stop Saying ‘If You’re Not Paying, You’re the Product.’” Slate Magazine, 27 April. https://slate.com/technology/2018/04 /are-you-really-facebooks-product-the-history-of-a-dangerous-idea.html. O’Rourke, P. 2017. “Bell Announces Plans to Acquire AlarmForce Industries.” MobileSyrup, 6 November. https://mobilesyrup.com/2017/11/06/bellacquiring-alarmforce-industries/. Ostrom, V., and E. Ostrom. 1999. “Public Goods and Public Choices.” In Polycentricity and Local Public Economies: Readings from the Workshop in Political Theory and Policy Analysis, edited by M.D. McGinnis, 75–103. Ann Arbor, mi: University of Michigan Press. Pai, A. 2017. “fcc Chairman Pai Addresses the 2017 Canadian Telecom Summit.” Telecom Trends (blog), 7 June. http://mhgoldberg.com/blog /?p=11380. Palmer, P. 2015. “Structural Separation: It’s Time to Be Impolite.” Philip Palmer Law (blog), 18 March. https://web.archive.org/web/201704 05203954/http://philippalmerlaw.ca/structural-separation-its-time-to-beimpolite/. Paperny, A.M. 2012. “Telcos in Talks with Ottawa to Shape Internet ‘Spy’ Bill: Documents.” Globe and Mail, 29 June. http://www.theglobeandmail .com/technology/tech-news/telcos-in-talks-with-ottawa-to-shape-internetspy-bill-documents/article4376958/. Parsons, C. 2013. “The Politics of Deep Packet Inspection: What Drives Surveillance by Internet Service Providers?” University of Victoria. https://christopher-parsons.com/wp-content/uploads/2017/08/Parsons _Christopher_PhD_2013.pdf (accessed 16 September 2020). – 2015. “Stuck on the Agenda: Drawing Lessons from the Stagnation of ‘Lawful Access’ Legislation in Canada.” In Law, Privacy and Surveillance in Canada in the Post-Snowden Era, edited by M. Geist, 257–83. Ottawa: University of Ottawa Press. http://www.ruor.uottawa.ca/bitstream/10393 /32424/1/9780776621838_WEB.pdf. – 2017. “The (In)Effectiveness of Voluntarily Produced Transparency Reports.” Business & Society 58, no. 1: 103–31. Patry, W. 2009. Moral Panics and the Copyright Wars. Oxford, uk: Oxford University Press. Payton, L. 2014. “Online Privacy Decision Means ‘Back to the Drawing Board’ for Tories.” CBC News, 13 June. http://www.cbc.ca/1.2674793.

References

219

Pickren, G. 2018. “‘The Global Assemblage of Digital Flow’: Critical Data Studies and the Infrastructures of Computing.” Progress in Human Geography 42, no. 2: 225–43. Poster, M. 2006. Information Please: Culture and Politics in the Age of Digital Machines. Durham, nc: Duke University Press. Power, E.M. 2013. The Law of Privacy. Markham, on: LexisNexis. Public Safety Canada. 2011a. “Canada’s Cyber Security Strategy.” https://web .archive.org/web/20120206023350/http://www.publicsafety.gc.ca/prg/ns /cbr/ccss-scc-eng.aspx. – 2011b. “Summary of Public Consultation on Access to Customer Name and Address Information for Public Safety Purposes.” 22 October. https://web.archive.org/web/20130731190554/http://www.publicsafety.gc .ca/prg/ns/sum-conslt-eng.aspx. – 2012. “Records on Government Consultations with Law Enforcement and Justice Officials Concerning the Negative Impact of a Warrant Requirement for Basic Subscriber Information.” Access to Information Act Request A-2011-00255. https://telecomtransparency.org/wp-content /uploads/2015/07/A-2011-00255.pdf (accessed 16 September 2020). – 2013a. “All Briefing Material from the adm Level and above Re: Cyber Security Capabilities in Government of Canada Systems from January 1 2010 to October 22 2012.” Access to Information Act Request A-201200264(2). – 2013b. “Correspondence from the Public on Bill C-30 from Feb 1 to May 10, 2012.” Access to Information Act Request A-2012-00301. – 2013c. “Records Concerning Bill C-30 Currently before Parliament, from Oct 1 to Dec 3, 2012.” Access to Information Act Request A-2012-00333. – 2014. “Lawful Access Legislation Produced between Jan 1, 2012 and July 31, 2012.” Access to Information Act Request A-2013-00234. – 2016. “Presentations Re: National Cyber Security Directorate under File ps ncsb 07 (January 1, 2014 to July 3, 2015).” Access to Information Act Request A-2015-00103. – 2017. “Horizontal Evaluation of Canada’s Cyber Security Strategy.” 29 September. https://www.publicsafety.gc.ca/cnt/rsrcs/pblctns/vltn-cnd-scrtstrtg/index-en.aspx. – 2018. “Briefing Notes from 2016: A Cyber Briefing Binder for a Jan 26 Meeting (PS-008194); A Briefing Binder for a Feb. 2, Meeting with the ceos of the Business Council of Canada (PS-008502); A July 7 Memo Setting an Agenda for a July 14 adm Meeting on Cyber Security (PS011345); A March 23 Briefing Note on Collaboration between ncsd and cse (PS-9344 and PS-9346); An April 4 Briefing to Parliamentarians on

220

References

the Anti-Terrorism Act (PS-9496); A June 10 Annotated Agenda for Meetings with Australia’s Counterterrorism Coordinator (PS-I0920) and cse (PS-9344 and PS-9346).” Access to Information Act Request A-201700279. Rabideau, M. 1991. “Duarte v. R.: In Fear of Big Brother.” University of Toronto Faculty of Law Review 49: 171–85. Rajabiun, R., and C. Middleton. 2013. “Rural Broadband Development in Canada’s Provinces: An Overview of Policy Approaches.” Journal of Rural and Community Development 8, no. 2: 7–22. – 2015. “Public Interest in the Regulation of Competition.” Journal of Information Policy 5: 32–66. Rebel Networks. n.d. “151 Front St.” Rebel Networks Inc. (blog). https://rebelnetworks.com/data-center-network/151-front-street-toronto/ (accessed 28 May 2020). Regonini, G. 2017. “Governmentalities without Policy Capacity.” Policy Sciences 50, no. 2: 163–78. Rens, J.-G. 2001. The Invisible Empire: A History of the Telecommunications Industry in Canada. Translated by K. Roth. Montreal, qc: McGill-Queen’s University Press. resa1983. 2011. “Rogers Answerable to crtc over itmp Again.” dsl Reports – Rogers. http://www.dslreports.com/forum/r25526726-Rogers-answerableto-CRTC-over-ITMP-again (accessed 16 September 2020). Richard, M., and D. Philpot. 2014. “GoFred: Municipally-Owned ict Utilities in Fredericton, New Brunswick.” The Journal of Community Informatics 10, no. 2. http://ci-journal.net/index.php/ciej/article/view/993. Rideout, V. 2003. Continentalizing Canadian Telecommunications: The Politics of Regulatory Reform. Montreal, qc: McGill-Queen’s University Press. Rogers. 2004. “Written Representations of Rogers Communications Inc.” 12 March. https://cippic.ca/sites/default/files/file-sharing-lawsuits/Rogers _Written_Reps_Mar12.pdf. Romaniuk, B.S., and H.N. Janisch. 1986. “Competition in Telecommunications: Who Polices the Transition.” Ottawa Law Review 18, no. 3: 561–661. Rosenblum, N.L., and R.C. Post, eds. 2002. Civil Society and Government. Princeton, nj: Princeton University Press. Sandvig, C. 2007. “Network Neutrality Is the New Common Carriage.” Info 9, no. 2/3: 136–47. Sandvine. 2014. “The Global Internet Phenomena Report: 1H 2014.” https://www.sandvine.com/hubfs/downloads/archive/2014-1h-globalinternet-phenomena-report.pdf (accessed 16 September 2020). SaveOurNet.ca. 2009. “Open Internet Coalition Launches Series of Events.”

References

221

31 May. https://web.archive.org/web/20090607234027/http://saveournet .ca/townhall. Scalet, S.D. 2007. “Pipe Cleaners: Telcos Offer Managed Security Services.” CSO Online, 12 July. http://www.csoonline.com/article/2121659/malwarecybercrime/pipe-cleaners—telcos-offer-managed-security-services.html. Schewick, B. van. 2010. Internet Architecture and Innovation. Cambridge, ma: The mit Press. Schneier, B. 2012. “When It Comes to Security, We’re Back to Feudalism.” Schneier on Security (blog), 26 November. http://www.schneier.com/essay406.html. Schnurr, J. 2016. “Low Cost Internet for Ottawa Community Housing Tenants.” CTV News, 21 December. http://ottawa.ctvnews.ca/low-cost-internetfor-ottawa-community-housing-tenants-1.3213142. Schultz, R.J. 1999. “Still Standing: The crtc, 1976–1996.” In Changing the Rules: Canadian Regulatory Regimes and Institutions, edited by G.B. Doern, M.M. Hill, M.J. Prince, and R.J. Schultz, 29–56. Toronto: University of Toronto Press. Schwartz, Z. 2020. “Suspending Data Caps, Offering Bill Relief: How 22 Canadian Internet Service Providers Are Responding to the Pandemic.” Logic, 18 March. https://thelogic.co/news/suspending-data-caps-offeringbill-relief-how-22-canadian-internet-service-providers-are-responding-tothe-pandemic/. Scott, Peter. 2001. “Free-Net Access Points.” 29 September. https://web.archive .org/web/20010929112249/http://www.lights.ca/freenet/. Scotton, G. 2000. “Rural Alberta Wired.” Calgary Herald, 3 November. Settles, C. 2012. “O Canada, Land of Community Broadband?” Broadband Communities, October. https://www.bbcmag.com/pub/doc/BBC_Oct12 _OCanada.pdf. Shade, L.R. 1994. “Computer Networking in Canada: From CA*net to canarie.” Canadian Journal of Communication 19, no. 1. http://www.cjconline.ca/index.php/journal/article/view/794. Shaw. 2004. “Written Representations of Shaw Communications Inc.” 10 March. https://cippic.ca/sites/default/files/file-sharing-lawsuits/Further WrittenSubmissionsShaw.pdf. Shaw, G. 2004. “Sharing Music over Internet Not Illegal, Federal Court Rules.” Vancouver Sun, 1 April. Shaw, R. 2012. “Competitors Slam ‘Unfair’ Government Deal with Telus; Companies Call 10-Year Contract Flawed, Suggesting Taxpayers Will Not Get the Best Value for Money as a Result.” Vancouver Sun, 14 March. Shearing, C., and L. Johnston. 2010. “Nodal Wars and Network Fallacies: A

222

References

Genealogical Analysis of Global Insecurities.” Theoretical Criminology 14, no. 4: 495–514. Shepherd, T., G. Taylor, and C. Middleton. 2014. “A Tale of Two Regulators: Telecom Policy Participation in Canada.” Journal of Information Policy 4 (February). Singer, L. 2012. “Unwarranted Access?” National (Canadian Bar Association), June. Singer, M.G. 1965. “Negative and Positive Duties.” The Philosophical Quarterly 15, no. 59: 97–103. Smith, G. 2017. “Despite Delays, Internet Co-Creator Vint Cerf Still Dreams of an Interplanetary Internet.” SiliconANGLE (blog), 1 October. https://siliconangle.com/2017/10/01/despite-delays-internet-co-creatorvint-cerf-still-dreams-interplanetary-internet/. Solomon, H. 2016. “Protect Privacy of Subscribers, but Don’t Stick Us with the Bill, cwta Tells Ottawa.” Cartt.ca, 6 December. https://cartt.ca/protectprivacy-of-subscribers-but-dont-stick-us-with-the-bill-cwta-tells-ottawa/. Standing Committee on Industry, Science and Technology. 2006. “Standing Committee on Industry, Science and Technology – Evidence: Tuesday November 7, 2006.” http://www.parl.gc.ca/HousePublications/Publication .aspx?DocId=2485667&Language=E&Mode=1. Standing Senate Committee on Legal and Constitutional Affairs. 2009. “Proceedings: Issue 8 – Evidence for May 14, 2009.” 14 May. http://www.parl.gc.ca/Content/SEN/Committee/402/lega/08evb-e.htm ?Language=E&Parl=40&Ses=2&comm_id=11. Starosielski, N. 2015. The Undersea Network. Durham, nc: Duke University Press. Stevens, G. 1976. “The ‘Consumer’s Empty Chair.’” Globe and Mail, 6 August. Stoddart, J., F. Work, E. Denham, I. Hamilton, A.E. Bertrand, E. Ring, E.K. Bengts, et al. 2011. “Letter to Public Safety Canada from Canada’s Privacy Commissioners and Ombudspersons on the Current ‘Lawful Access’ Proposals.” 9 March. https://www.priv.gc.ca/en/opc-news/news-andannouncements/2011/let_110309/. Stoker, G. 1998. “Governance as Theory: Five Propositions.” International Social Science Journal 50, no. 155: 17–28. Stone, A. 1991. Public Service Liberalism: Telecommunications and Transitions in Public Policy. Princeton, nj: Princeton University Press. Streeter, T. 2011. The Net Effect: Romanticism, Capitalism, and the Internet. New York: nyu Press. Stryker, S., and A.S. Macke. 1978. “Status Inconsistency and Role Conflict.” Annual Review of Sociology 4: 57–90.

References

223

swift. 2016. “swift Receives $180 Million in Provincial and Federal Finding through Small Communities Fund.” 26 July. https://web.archive.org /web/20160926164014/http://swiftnetwork.ca/SWIFT_Media_Release _July_26_2016.pdf. TekSavvy. 2015. “Transcript Brief.” 5 November. https://cippic.ca/sites/default /files/CF_No._T-2058-12_-_TekSavvy_ats_Voltage_-_Transcript_Brief.pdf. Telecommunities Canada, and Industry Canada. 1997. “Framework for CoOperation between Telecommunities Canada (tc) and Industry Canada to Enhance the Ability of Canadian Communities to Utilize Electronic Public Space.” 17 August. http://tc.ca/framework.html. telus. 2014. “telus Transparency Report 2013.” https://web.archive.org/web /20150501060132/http://about.telus.com/servlet/JiveServlet/download Body/5544-102-1-6081/TELUS%20Transparency%20Report%202013 %20-English.pdf. – 2016. “2015 telus Sustainability Report.” https://web.archive.org/web /20181214023858/https://assets.ctfassets.net/rz9m1rynx8pv/71afvhy QCs0waoWWOGEU2y/bdf7bfb5a3cd3aaaf6831f59c6c3dda9/2015 _Sustainability_Report-EN.pdf. – 2020. “telus Wise: About.” https://www.telus.com/en/wise. telus and Government of bc. 2011. Connecting British Columbia Agreement. https://www2.gov.bc.ca/assets/gov/british-columbians-our-governments /about-the-bc-government/strategic-partnerships/contract-administrationadministrator-s-office/cbca.pdf. Therrien, D. 2015. “Submission to the Standing Committee on Industry, Science and Technology.” 12 February. https://www.priv.gc.ca/parl/2015 /parl_sub_150212_e.asp. Thies, C.G., and M. Breuning. 2012. “Integrating Foreign Policy Analysis and International Relations through Role Theory.” Foreign Policy Analysis 8, no. 1: 1–4. Timmermans, S., and S. Epstein. 2010. “A World of Standards but Not a Standard World: Toward a Sociology of Standards and Standardization.” Annual Review of Sociology 36, no. 1: 69–89. Toews, V. 2012. “House of Commons Debates.” Openparliament.ca, 28 February. https://openparliament.ca/debates/2012/2/28/vic-toews-1/. Tunney, C. 2018. “Ottawa Orders crtc to Investigate Reports of ‘Aggressive’ Telecom Sales Practices.” CBC News, 14 June. https://www.cbc.ca/news /politics/crtc-telecom-sales-practices-investigation-1.4706260. Valentino-DeVries, J. 2020. “Cellphone Carriers Face $200 Million Fine for Not Protecting Location Data.” New York Times, 28 February. https://www.ny times.com/2020/02/28/technology/fcc-cellphones-location-data-fines.html.

224

References

Valiquet, D. 2011. “Legislative Summary of Bill C-22: An Act Respecting the Mandatory Reporting of Internet Child Pornography by Persons Who Provide an Internet Service.” Library of Parliament, 15 February. https://lop .parl.ca/sites/PublicWebsite/default/en_CA/ResearchPublications/Legislative Summaries/403C22E. Van Gorp, A., and C.A. Middleton. 2010. “The Impact of Facilities and ServiceBased Competition on Internet Services Provision in the Canadian Broadband Market.” Telematics and Informatics 27, no. 3: 217–30. Vancouver Community Network. 2015. “Street Messaging System (sms) Feasibility Report.” July. https://www.streetmessenger.io/report.pdf (accessed 16 September 2020). Vandenberghe, F. 2007. “Avatars of the Collective: A Realist Theory of Collective Subjectivities.” Sociological Theory 25, no. 4: 295–324. Venugopal, R. 2015. “Neoliberalism as Concept.” Economy and Society 44, no. 2: 165–87. Vogel, S.K. 1996. Freer Markets, More Rules: Regulatory Reform in Advanced Industrial Countries. Ithaca, ny: Cornell University Press. Wark, W. 2016. “cse and Lawful Access after Snowden.” cips Working Paper. http://www.cips-cepi.ca/wp-content/uploads/2011/09/WWPFINALfinal.pdf (accessed 16 September 2020). Wendt, A. 2004. “The State as Person in International Theory.” Review of International Studies 30, no. 2: 289–316. Whitaker, R. 2012. “The Curious Tale of the Dog that Hasn’t Barked (Yet).” Surveillance & Society 10, no. 3/4: 340–3. Wilson, K.G. 2000. Deregulating Telecommunications: U.S. and Canadian Telecommunications, 1840–1997. Lanham, md: Rowman & Littlefield. Wingrove, Josh. 2019. “Huawei Likely Faces 5G Ban in Canada, Security Experts Say.” Bloomberg, 6 February. https://www.bloomberg.com/news /articles/2019-02-06/huawei-likely-faces-5g-ban-in-canada-security-expertssay. Winseck, D. 1998. Reconvergence: A Political Economy of Telecommunications in Canada. Cresskill, nj: Hampton Press. – 2015a. “Intermediary Responsibility.” In International Encyclopedia of Digital Communication and Society, edited by R. Mansell, P.H. Ang, C. Steinfield, S. van der Graaf, P. Ballon, A. Kerr, J. D. Ivory, S. Braman, D. Kleine, and D.J. Grimshaw, 488–502. Boston, ma: Wiley-Blackwell. http://www.david ellis.ca/wp-content/uploads/2015/01/winseck-Intermediary-Responsibility _Final-APH.pdf. – 2015b. “At Bell Media, Editorial Meddling by Execs Appears to Be a Recurring Problem.” Policy Options, 28 March. https://policyoptions.irpp

References

225

.org/2015/03/28/at-bell-media-editorial-meddling-by-execs-appear-to-be-arecurring-problem/. Winsor, H. 1973. “Wiretap Bill: A Balance Remains Elusive.” Globe and Mail, 18 June. Wire Report. 2004. “Call-Net Could Be Next Telco to Enter Consolidation Game with Power Companies.” 31 March. http://www.thewirereport.ca /news/2004/03/31/call-net-could-be-next-telco-to-enter-consolidationgame-with-power-companies/13256. – 2019a. “Bell Expanding Targeted Ads Program to All Services.” 8 January. https://thewirereport.ca/2019/01/08/bell-expanding-targeted-ads-programto-all-services/. – 2019b. “Telus Purchase of Canadian adt Business a Good Move: Analysts.” 1 October. https://thewirereport.ca/2019/10/01/telus-purchase-ofcanadian-adt-business-a-good-move-analysts/. Wood, J. 2006. “Research and Innovation in the Field of Security: A Nodal Governance View.” In Democracy, Society and the Governance of Security, edited by J. Wood and B. Dupont, 217–48. Cambridge: Cambridge University Press. Wood, J., and C.D. Shearing. 2007. Imagining Security. Portland, or: Willan. Wu, T. 2003. “Network Neutrality, Broadband Discrimination.” Journal of Telecommunications and High Technology Law 2: 141–76. Zajko, M. 2015a. “Canada’s Cyber Security and the Changing Threat Landscape.” Critical Studies on Security 3, no. 2: 147–61. https://doi.org/10.1080 /21624887.2015.1071165. – 2015b. “The Copyright Surveillance Industry.” Media and Communication 3, no. 2: 42–52. Zhang, L. 2007. “A Retrospective View of nat.” IETF Journal (blog), 7 October. https://www.ietfjournal.org/a-retrospective-view-of-nat/. Zittrain, J. 2003. “Internet Points of Control.” Boston College Law Review 44, no. 2: 653–88.

226

References

Index

151 Front Street West, 31–3, 42 5G (fifth-generation wireless), 162 Abramson, Bram, 187n14 academic networks. See Research and Education Networks (RENs) advertising. See under Bell (Canada) agency (in social theory), 11–12 Alberta: Alberta Government Telephones, 50–1, 56, 185n1; and Bell, 43; cooperatives, 184n15. See also Olds; SuperNet Allstream, 56 Atlantic Canada. See Maritime provinces AT&T, 43, 46 Axia, 67–8, 77 Bains, Navdeep, 100, 182n3 BCNET, 21, 87, 186n7 Bell (Canada): and advertising, 140–1, 143; and CA*net, 186n7; connectivity for low-income families, 182n3; consolidation, 45–6, 56; and cyber security, 157, 159–60, 192n17; and GoldTV, 105–6; history of, 43–6, 83; and

media content, 106, 109; and net neutrality, 114, 118–19; and privacy, 131, 139–41; sharing facilities with TELUS, 182n2; and surveillance, 188n3–4, 190n4; vertical integration, 109. See also GoldTV; SuperNet; system integrity; usage-based billing Bell (United States), 43, 62. See also AT&T Blais, Jean-Pierre, 28–9, 186n5, 187n16 BMG Canada Inc. v John Doe, 130–1 Board of Railway Commissioners (BRC), 83, 109, 180n9–10 bottlenecks. See under internet British Columbia, 180n1, 182n22; British Columbia Telephone Company (BC Tel) 56, 188n4; provincial government, 87. See also BCNET; Connected Coast broadcasting, 106, 118–19, 186n4; Broadcasting Act, 118 Cabinet (federal). See federal government of Canada

228

Index

cable networks (CATV/cablecos), 20, 47–8, 84, 89–90, 95 Calgary, 37, 184n17, 185n23 Canada. See federal government of Canada Campbell River, BC, 184n17 CANARIE, 21 CA*net, 186n7 CCAICE (Canadian Coalition Against Internet Child Exploitation), 133–4, 150, 155 Centrex, 55 child abuse imagery / child pornography, 133–4, 154–5 Citizen Lab, 139, 160 CityWest, 78, 174, 183n13 clean pipes. See smart pipes CNOC (Canadian Network Operators Consortium), 95 Cogeco, 131 Cogent, 192n19 Columbia Basin Broadband Corporation, 76 common carriage, 45–6, 107–9, 118–21 Community Access Program (CAP), 70 community networks, 21, 68–71, 78, 172–3. See also FreeNets; municipalities: municipal networks competition, 82–103; and consolidation, 56, 169–70; facilities-based, 88–92, 97–103; and market failure, 60, 84, 91; and market power, 84, 91, 96, 103; and the public sector, 59, 66–7, 75, 80, 86–7; servicebased, 89, 97–8; and structural separation, 98; in telephony, 45, 83. See also competitive dependence; liberalization

competitive dependence, 56, 81–2, 91–6 conferences, 16, 30–1, 34, 95, 178n13 Connected Coast, 173–5 co-operatives, 43, 184n15 copyright, 105–7, 114, 129–32; Copyright Board of Canada, 22 Coquitlam, BC. See QNet Cortes Island, BC, 174–6 COVID-19 pandemic, 5, 27, 65, 119–20 Crouch, Colin, 97 CRTC (Canadian Radio-television and Telecommunications Commission), 21–2, 28–31, 48–52; and basic services, 29–30; and Broadcasting Act, 118; and Internet Code, 52; and Telecommunications Act, 50–1, 89, 109, 118–19. See also federal government of Canada; liberalization; mandated access; network neutrality CSE/CSEC (Communications Security Establishment), 156, 158, 192n14, 192n17 CSTAC (Canadian Security Telecommunications Advisory Committee), 157–60 CTC (Canadian Transport Commission), 47–8, 181n9 CTCP (Canadian Telecommunications Cyber Protection Working Group), 157 Cybera, 21, 186n7 cyber security, 145–7, 152–64 dark fibre, 76 Denton, Timothy, 188n19 deregulation. See liberalization Dewey, John, 63

Index

229

digital divide. See internet: inequalities in disintermediation, 9, 166 Distributel, 131 DPI (deep packet inspection), 113–15 dumb pipes, 47, 161–2

Gold River, BC, 173 GoldTV, 105–7 governance, 6. See also under internet governmentality, 7–8, 11 Granisle, BC, 183n13

end-to-end, 46–7, 112, 166, 168 Engelhart, Ken, 81, 187n13, 189n16 EnStream, 13–14, 141 Europe: copyright, 106; mandated access, 89, 98

IETF

facilities-based competition. See under competition FCC (Federal Communications Commission), 120, 141 federal government of Canada: and liberalization, 50; under Justin Trudeau, 99; under Stephen Harper, 99, 102; policy directions (of 2006 and 2019), 52–3, 57, 90, 100–1; telecommunications governance and jurisdiction, 21–2, 43, 50–1, 86 fibre-optics, 26–7, 56, 72–3. See also 151 Front Street West; Connected Coast; dark fibre; mandated access: to fibre; New York City; Olds; QNet Finckenstein, Kondrad von, 189n9 Five Eyes, 156 Fletcher, Graham, 3 Flichy, Patrice, 35 Foucault, Michel, 7–8, 11, 147 FreeNets, 20, 68–72 Fredericton, 184n15 Galloway, Alexander, 26 Geist, Michael, 140, 149, 188n19

(Internet Engineering Task Force), 17–18 IISPs (independent ISPs), 20, 56, 92–6, 169 incumbents, 20, 42, 49, 51, 56–7, 101–3; as Incumbent Local Exchange Carriers (ILECs), 21; and lawful access, 150–1. See also liberalization; mandated access: to incumbent networks Industry Canada, 70, 133, 139. See also ISED information highway, 51, 89, 168 infrastructure, 8, 14–16, 27–8, 35–8, 176; legacy (pre-internet), 23, 42, 59–60. See also mandated access interconnection: and liberalization, 48; peering, 16–17, 33–4; in telephony, 43; transit, 178n11 intermediaries, 4–5, 169–73; kinds of, 19–21; as instruments of public policy, 9–12, 71–4, 167–8, 171–2 intermediation, 9, 165–8; and competition, 82; and net neutrality, 111–12, 121–2 internet, 15–16, 26–7; backhaul, 36–7; bottlenecks, 24, 56, 84, 94, 103; governance, 6–10, 21–2; history and development, 51, 55, 70, 84; inequalities in, 59–61, 72–4; as a public good, 64–5; service provider types, 19–21; utopias, 35,

230

Index

37; as vital, 27–30. See also end-toend; infrastructure ISED (Innovation, Science and Economic Development Canada), 22, 182n3, 187n11, 192n17 ITMPs (Internet Traffic Management Practices), 115–17 IXPs (internet exchange points), 16, 32–4, 56 Janisch, Hudson, 85 Kamloops, BC, 184n17, 185n24 Kelowna, BC, 184n17 Koblovsky, Jason, 116 Klass, Ben, 117–18

market power. See under competition MediaSmarts, 138–9, 182n2 mere conduit, 45, 155, 166, 177n5 Mezei, Jean-François, 117–18 middle mile, 76 MTS (Manitoba Telephone System). See under Manitoba Mueller, Milton, 155 Mulock Committee, 44, 46 municipalities: municipal networks, 21, 64–5, 74–9; and policy, 36; and telephony, 43, 56, 83. See also Olds; rights-of-way Murphy, Teresa, 116–17 (North American Network Operators’ Group), 16, 18 National Capital FreeNet (NCF), 69, 71 natural monopoly, 46–7, 83 neoliberalism, 51–3, 82, 97 net neutrality (network neutrality), 104–7, 108–22 Nelson, BC, 109, 184n17 Netflix, 17, 116, 118 New Westminster, BC, 184n17 New York City: and Verizon, 67 nodal governance, 11, 170, 172 Norton, William, 33 Nova Scotia, 184n15, 186n7 Nunavut, 183–4n14 NANOG

Labrador: CRRS cooperative, 184n15 Lakeland Networks, 185n21, 183n14 last mile, 32 lawful access, 132–9, 145–55, 163–4 liberalization, 49–55, 57, 81–3, 103, 171–2; international context, 51, 89; and privacy, 150–1 Litschko, Chris, 185n21 location data, 13–14, 141 Macaulay, Tyson, 120, 161 mandated access: to fibre, 90, 100; to incumbent networks, 89–101; and net neutrality, 111, 113–14; in telephony, 45. See also competitive dependence Manitoba: and Bell, 43; Manitoba Telephone System (MTS), 50, 56, 185n1 Maritime provinces, 180n5. See also Nova Scotia market failure. See under competition

Office of the Privacy Commissioner of Canada (OPC). See under privacy Ontario: conferences, 31, 34; cooperatives, 184n15; municipal networks, 56, 183n13; privacy legislation, 127; regional networks, 183n14

Index

Olds, AB, 36–9, 77–8 O-Net, 37, 39, 77–8 open access, 75–6 OpenMedia, 99, 114 Pai, Ajit, 120 pandemic. See COVID-19 Parsons, Christopher, 139 peering. See under interconnection Penticton, BC, 184n17 PIPEDA (Personal Information Protection and Electronic Documents Act), 124, 132–6, 151–2 policy directions (of 2006 and 2019). See under federal government of Canada populism (in telecommunications policy), 98–9 Prince Rupert, BC. See CityWest privacy, 123–44, 149–52, 162; Office of the Privacy Commissioner of Canada (OPC), 140, 190n19; Protection of Privacy Act, 128; transparency reporting, 138–9. See also lawful access; location data; PIPEDA

public good, 58–9, 61–71, 80, 167, 171–2 public policy, 36, 61–2. See also intermediaries: as instruments of public policy; public good public-private partnerships, 67–8 QNet, 75–6, 79 Quebec: cooperatives, 184n15; privacy legislation, 127; regional networks, 183n14 Quebecor. See Vidéotron regional networks, 76–7, 183n14

231

regulatory capitalism, 53–4, 57, 82, 97–8, 165, 169–71 Research and Education Networks (RENs), 55, 66, 86–7 rights of way, 43, 83 Rogers Communications: connectivity for low-income families, 182n3; and net neutrality, 116, 118; and privacy, 131, 137, 139–40 role conflict, 10–11, 167; and competition, 59, 67, 87–8, 93 (see also competitive dependence); and cyber security, 155–6, 161–2 ; and net neutrality, 104, 106–7, 161; pre-internet, 41; and privacy, 14, 135–7, 141–4; and public networks, 59, 61, 66–8; and surveillance, 135–7, 139, 146–7 Romaniuk, Bohdan, 85 rural connectivity, 60, 74, 87, 179n2; in the telephone era, 43–6, 83. See also internet: inequalities in Saskatchewan: and Bell, 43; cooperatives in, 184n15; SaskTel, 185n1 Scott, Ian, 100, 179n2 service-based competition. See under competition Shaw Communications: and lawful access, 123; and privacy, 131 Sinclair, Don, 175 smart pipes, 156, 159–62 Smith, Victoria, 175 Snowden, Edward, 129, 138–9, 146, 156 Spencer, Matthew David, 123. See also Spencer decision Spencer decision (R v Spencer [2014]), 123–4, 126, 132, 135–7, 151–2

232

Index

structural separation. See under competition SuperNet, 37–8, 67–8, 76–8 surveillance, 121, 124–56. See also DPI; lawful access; smart pipes; wiretaps system integrity, 46–8 Tbaytel, 78, 183n13 Telecommunications Act. See under CRTC

Telecommunities Canada, 70 telegraph, 42, 45, 108–9 telephony: history of, 43–8, 83; and privacy/surveillance, 127–8. See also universal service TekSavvy: and competitive dependence, 95; and net neutrality, 107, 114; and privacy, 131–2, 189n16 TELUS: 2005 labour dispute, 110–11; in BC, 87, 173; connectivity for low-income families, 182n3; history, 56; and lawful access, 126–7, 137, 140, 189n16, 191n8; and media content, 121; merging with Bell, 56; and privacy, 126–7, 131, 140; sharing facilities with Bell, 182n2; Wise program, 138 throttling, 114–17 Thunder Bay, ON. See Tbaytel Toews, Vic, 155 Toronto: police, 188n4, 191n8; as “Silicon Valley North,” 35. See also

151 Front Street West; Ontario: conferences; TorIX TorIX (Toronto Internet Exchange), 33–4. See also 151 Front Street West transit. See under interconnection transparency reports. See under privacy United States: copyright, 106, 130; history (of telecommunications), 44, 46, 62, 70; interconnection, 17, 36; public networks, 67; net neutrality, 110; privacy, 14; surveillance, 156. See also FCC universal service, 46, 171, 181n12 usage-based billing (UBB), 99, 117 Vancouver: Vancouver FreeNet, 65; Vancouver Community Network, 71 Verizon, 67 vertical integration, 106, 109–10, 113, 118–21 Vidéotron: and copyright, 130–1; and net neutrality, 118–19 Weston, Jay, 69–70 wholesale access. See competitive dependence; mandated access wiretaps, 127–8, 148, 151, 190n5 zero rating, 117–19 Zittrain, Jonathan, 5