Changing Things: The Future of Objects in a Digital World 9781350004351, 9781350004368, 9781350004337

Citation preview

Changing Things

Changing Things The Future of Objects in a Digital World Johan Redström and Heather Wiltse

BLOOMSBURY VISUAL ARTS Bloomsbury Publishing Plc 50 Bedford Square, London, WC1B 3DP, UK 1385 Broadway, New York, NY 10018, USA BLOOMSBURY, BLOOMSBURY VISUAL ARTS and the Diana logo are trademarks of Bloomsbury Publishing Plc First published in Great Britain 2019 Copyright: © Johan Redström and Heather Wiltse, 2019 Johan Redström and Heather Wiltse have asserted their right under the Copyright, Designs and Patents Act, 1988, to be identified as Authors of this work. Cover design: Louise Dugdale Cover image © Mina De La O / Getty images All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. Bloomsbury Publishing Plc does not have any control over, or responsibility for, any third-party websites referred to or in this book. All internet addresses given in this book were correct at the time of going to press. The author and publisher regret any inconvenience caused if addresses have changed or sites have ceased to exist, but can accept no responsibility for any such changes. A catalogue record for this book is available from the British Library. A catalogue record for this book is available from the Library of Congress. ISBN: HB: ePDF: eBook:

978-1-3500-0435-1 978-1-3500-0433-7 978-1-3500-0434-4

Typeset by Integra Software Services Pvt. Ltd. To find out more about our authors and books visit www.bloomsbury.com and sign up for our newsletters.

Contents 1 Introduction 2 What Is Going On with Things? 3 Just Press Play, Please 4 Fluid Assemblages 5 Things for Us 6 Things in Themselves 7 A Conceptual Toolkit 8 Assembling an Analytic Playlist 9 Making Concepts References Index

1 9 21 33 43 61 77 127 145 165 174

1

Introduction We are surrounded by things. It is hard to think of any moment of life in which we are not using or at least around artifacts designed and produced by humans. Acquiring, configuring, arranging, repurposing, and eventually disposing of things constitute a cycle that provides a kind of ongoing material narrative that continues throughout life. In at least industrialized regions, most humans arrive in the world to find a full assortment of things already present around them; and going through people’s things, deciding what to keep and what to dispose, often marks the last phase of settling their affairs after they have passed. Since its beginnings in mid-eighteenth-century Great Britain, the Industrial Revolution and corresponding development of consumer culture shifted the design, production, and acquisition of things into high gear and almost uninterrupted acceleration, such that some people and parts of the world are now often rather drowning in excess or discarded stuff, while many others suffer from the consequences. We are technological beings through and through and since our earliest beginnings, and gradually we have come to the point where it is now the artificial rather than what could be perhaps called “nature” that constitutes the ultimate horizon of our existence. And now, again, the ways things are made and used have fundamentally changed. Many of the things we now live with do not take only physical form: smartphones and watches, laptops and game consoles, wearable health and fitness trackers, and similar are different from our things of the past. For reasons to be explored in this book, we will call these things fluid assemblages. Whereas your mobile phone might look rather similar to any other sleek appliance at home, you only need to disconnect it from the network to realize just how different it is. The mobile phone, and what you do with it, is in no sense delimited or defined by its physical presence. Sure, we can touch it, feel

2

Changing Things: The Future of Objects in a Digital World

it, look at it—but it is what we do with it, or perhaps rather through it, that defines its role in our life: searching, talking, watching, browsing, sharing, shopping, listening, tracking, posting, liking, and so on. It is nearly impossible to assess and understand what these things do based on physical appearance alone. Unlike our old stable and predictable physical things, these digital things are networked, dynamic, and contextually configured. Indeed, they can at times also be changeable and unpredictable things, even inscrutable when it comes to what they actually do and whom they really serve. Just as we can recognize the very real actions and effects of these networked computational things, it is also in many cases becoming ever more difficult to clearly identify and pin them down. It is almost as if they emerge from some kind of virtual ether when activated, only to withdraw again when the applications are closed or the devices switched off. Moreover, they show up in different ways depending on the context, responding to variables such as user account, location, time, and so forth. They can also change over time as the code is updated, while the code that generates unique instances of things can travel across devices. But when they are up and running, these things—compositions of devices, code, and networked resources and connections—are remarkably solid in terms of providing certain kinds of functionality and enabling certain actions. In what follows, we will explore the idea that the changes in what things are, and how we live with them, currently underway are in some ways as drastic as the shift from handmade to industrially manufactured things. And while this time it is also a change driven by technological developments and economic interests, to think of it as a matter of the digital and computational would be much like reducing the Industrial Revolution to a matter of iron and steel, steam engines and electricity. Rather, it is in the reconfigurations of relations between people and things, between the design of and life with, that the most significant transformations are taking place. While the first decades of industrialized production were primarily about mimicking forms familiar from earlier forms of production, there was a gradual realization that this change actually called for a different understanding of how something becomes meaningful in someone’s life. This insight sparked the birth of industrial design as we now know it: the realization that the new industrial opportunities called for a new unity of art and technology, a new sociopolitical program, a new aesthetics, and above all a new vision of the

Introduction

3

“good life.” Or, as seen from another perspective, industrialized capitalism had to call a new discipline into being to realize its full potential—and so it did. We will argue that we now face a challenge somewhat similar to this situation: a fundamentally different way of “making things” made possible by rapid technological development and further accelerated by new economical ideas that are currently radically outperforming the pace in which design develops. This is causing an increasingly problematic gap between what these new “things” are and the ways we are meant to incorporate them into the fabric of our everyday lives. Of course, there is already massive benefit present in our lives, just as also the first years of industrially produced copies of previously manmade objects were of great value. Indeed, one might even be inclined to think we are already doing so well that this should not be of too much concern. But there is also a slightly uncanny feeling that these new things bring about something that is not necessarily what we first thought. The fact that the business model of some of the now largest corporations is built on trading detailed information about what we do, with whom, where, and how, is at least sometimes causing concern, leaving us with a feeling that perhaps we do not quite understand precisely what it is that these things do. And while we’re at it, why do we always have to make do with “beta” these days, why can’t things just work? At the same time, they are wonderful things. Things we want to have with us all the time. Things we want to keep in our hands. Some of them are the first things we pick up in the morning and the last things we put down when going to bed. They are things we use to stay in touch with loved ones. They offer safety, excitement, entertainment, inspiration, and much more. This complex situation cannot be reduced to a matter of good or bad. Instead we have to ask to what extent do our existing ways of relating to things, be it in use, design, practice, or theory, actually allow us to grasp what these new things are? And most important, what sense and sensibility do we now need to cultivate in order to design and live well with them? Our purpose with this book is to investigate and begin to articulate some of the key characteristics of these things—these fluid assemblages—so that we might be able to better understand and care for their character and consequences. In doing this, we are attuned to the constant and in many ways inevitable interplay between theoretical lenses and empirical observations, many of which have originated with our own experiences of interacting with and through

4

Changing Things: The Future of Objects in a Digital World

these things. The theoretical frameworks we are accustomed to thinking with, explicitly or implicitly, individually and within discursive communities, shape what we can see, connect, conceptualize, visualize, and articulate; and it is when these are no longer adequate to account for what we are seeing and break down that we must do the challenging and often distinctly uncomfortable work of pushing beyond them, even as the only way beyond is through. In this spirit, we have worked explicitly and intentionally with an interplay between case studies and theoretical frameworks, using each to push back against and expand the other. Although we at times foreground one or the other, the two are always intertwined. Although we do our best to respect the areas of scholarship with which we engage, our faithfulness is primarily to the account we try to develop. Our aim is to develop a perspective that is incisive, rich, and connected to key foundational issues in philosophy, design, and related areas, not least because of the many strains of thought we bring into conversation with each other. We do not claim to do this comprehensively or rigorously, but hope that we can through our (unfaithful) articulations at least sketch the outlines of territories worth investigating—with the hope that others may join us there and improve on our efforts. Even as we engage with things that currently exist, our goal is not merely to give an account of the present, but to identify the underlying dynamics, trajectories, and potentialities that may guide future developments. Moreover, while present configurations, technologies, and sociotechnical norms may provide the environment and resources that scaffold and shape ongoing processes of design and development, there is always the possibility to engage with these matters more explicitly and intentionally. This is the kind of reflective and critical engagement that we hope to enable and prototype here, inquiring into what is but always with an orientation to what things could (and perhaps should or should not) be assembled. In short, our project is about changing things.

About this book Over time, a variety of ways of thinking and talking about contemporary things have been developed in order to attempt to understand them and the roles they play in the world. New technological developments require new

Introduction

5

theory, while (at least ideally) incisive critical theory can inform responsible and innovative design practice and, thus, the building of the artificial world. This interplay between things and “thing theory” has been ongoing at least since the Industrial Revolution, and has been made explicit in design theory, philosophy of technology, and other fields concerned with the character of contemporary life and society. There are a few key assumptions and commitments that guide our investigation. The first is that when we refer to “things,” we mean designed technical artifacts, not things in general. Certainly, all the material things in the world are significant and worthy of philosophical consideration. But here we are more specifically concerned with things that are the objects of industrial design—things that are designed with at least some degree of professional design expertise and in service of particular client and user groups, goals, functions, and interests. The second is that we are interested in how these things actually exist in the world and what they do. In this sense, our approach could be considered materialist in that we believe, to use the overused expression, matter matters and is not reducible to experience, concepts, semantics, discourse, social constructions, subjectivities, and the like. While these aspects are absolutely significant, and in very concrete ways, they also provide only a partial perspective. This is especially important to recognize as we consider things that outstrip our usual sense-making capabilities, thus requiring us, somewhat paradoxically, to move outside the frame of human experience. At the same time, human experience is our motivation for addressing these things. We care about the ways in which things mediate human experience and engagement with the world, and how experience and engagement are shaped by the character of things and the variety of mediations they can enable. In the context of contemporary challenges such as sustainable development, such a human-centric view might seem quite insufficient as the things we are addressing, their production and their use, have an impact on the world far beyond human experience. The reason for starting with human experience instead lies with the need to understand and articulate how we perceive and act upon this new complexity. Or in other words, if the categories—such as “things”—we form and use as the basis for the way we understand and reason about our actions do not quite match what is actually

6

Changing Things: The Future of Objects in a Digital World

there, then chances are we will not be very successful in changing things in the direction we intend. Finally, while our focus here is on the things we call fluid assemblages, our investigation is in some ways broader. We find that in order to understand what is going on with fluid assemblages we need to bring in a number of different perspectives and work across a variety of scales, from the level of local interactions to that of larger social structures and dynamics. This is a matter of properly accounting for the different kinds of entities that are gathered together in things that guide their continual unfolding, modes of presence, appropriations, figurations, and configurations in social worlds in which they act in ways that are not neutral. As such, it turns out that a large extent of our investigation is as much about designed things in general as it is for fluid assemblages in particular, even as fluid assemblages represent gatherings of more striking dynamism and scale. While we point to fluid assemblages as a new and distinctive type of thing, we also recognize continuity with previous kinds of things rather than total rupture. In fact, many of the dynamics entailed in fluid assemblages are simply the realization of tendencies and forces at work long before technical developments enabled them to be expressed in this particular form.

Overview of the book This book has three main parts. The first serves to introduce the notion of fluid assemblages and the design issues they entail. Following this introductory chapter, we ask the question of “What is going on with things?” In Chapter 3, “Just Press Play, Please,” we move deeper into a specific example of how these developments unfold, and we then conclude this first part with our initial account of what is a fluid assemblage. The second part of the book engages more broadly in concepts and cases, taking a journey through a range of different domains and disciplines to enable a multifaceted view on things. In Chapter 4 we develop the concept of fluid assemblages in more detail. In the next two chapters we consider existing possibilities for considering things as they exist for us and things in themselves, respectively. Then in Chapter 7 we develop a conceptual toolkit

Introduction

7

for fluid assemblages, drawing on and applying concepts developed in other contexts. We try out the toolkit in Chapter 8, assembling an analytic playlist in an extended analysis of Spotify as one particular case. With a set of cases and associated conceptions in place, we are then better equipped to return to the main question raised here, namely that of how to design and live well with fluid assemblages. The third and final part of the book consists of the concluding chapter in which we further articulate issues in the designing of and the living with fluid assemblages. In particular, we will discuss how the notion of an assemblage helps us explain why current aesthetics and approaches to design are seemingly not able to address and resolve certain critical issues in how these things are made part of our lives and how we use them. This opens up, or so we will argue, for rethinking how industrial design has come to configure design and use in relation to each other, and how this now needs to be replaced by other configurations to form a different social contract (much like how mass manufactured goods could not rest on the same social contract between maker and user as was the case when making was bespoke and a craftsmanship). Indeed, whereas things used to be passive resources for us to make part of our lives, we are perhaps now closer to a more symmetric set-up: we are quickly becoming as much part of the doings of things as they are a part of ours. We may even ask: if the primary human “product” of the industrially manufactured object was the “consumer,” what would be the corresponding entity brought about by the fluid assemblage? This book is an attempt to find out.

2

What Is Going On with Things? To ask “what is a thing?” is to dwell on a question at least as old as philosophy. Our relation to things, in the wide sense, is so significant for our existence that perhaps only our relation to other people matters more to us. As such, these questions have formed the basis for much intellectual effort over the course of history. Even one of the central questions asked here—how things that change can be perceived as both the same and not the same at the same time—has fascinated people for thousands of years. In the story about the ship of Theseus, a thought experiment used by Heraclitus, Plato, and onward, the materials of a ship are gradually replaced as the wooden planks and other materials decay, eventually resulting in a ship that retains none of its original parts. Is it still the same ship? In the seventeenth century, the philosopher Thomas Hobbes would complicate matters further by asking us to consider what would happen if all the discarded pieces were used to build another ship, so that in the end we have two ships: one that has continuously retained a certain form over time and another that is made from the original materials. Which one is now Theseus’s ship? Are there now two of them? What is it that makes it possible for a thing to have an identity over time: form, material, or something else? Across a wide range of academic disciplines, there are enormous resources to draw upon if one wants to understand more about things and how we relate to them: with respect to material, aesthetics, culture, politics, economics, and so on. And yet, the reason we keep asking the same questions over and over again is that things change, urging us to add yet other pieces to the by now gigantic picture we collectively try to paint of what it means to make and live with things. And thus we are trying to add only a very small piece here.

10

Changing Things: The Future of Objects in a Digital World

In order to bring into focus what is now changing when it comes to our everyday things, it may be helpful to consider briefly a few aspects of things that have been possible to more or less take for granted up until now.

Taking things for granted We can start at the very beginning, with our perceptual experience of things. A “thing” is something quite extraordinary. It is something that stands out, something that in our experience emerges as a meaningful whole that can be understood as some form of unity distinct from its surrounding. We could think of perceptual phenomena such as the ability to distinguish figure– ground relationships. Precisely what it is that grounds this experience of a meaningful unity is far from trivial to understand, but we do know that it begins already in basic perception and that aspects of these abilities may even be hardwired in our perceptual systems. For instance, how an infant develops the perception of object permanence, such as being able to follow an object and to understand also that things that are out of sight can still exist, was called out by development psychologist Jean Piaget as critical to how our intellectual abilities develop. Not to mention it enables us to play Peekaboo. The fact that artifacts have fairly stable physical properties is what allows us to make sense of and relate to them in other ways as well. For example, it is a hammer’s solidity, in many senses, that allows for appropriating it for various purposes. Regardless of whether that use is the most typical one of striking another object or another perhaps more unusual one, such as using it for a doorstop, it is the hammer’s physical form and properties that allow for these actions. Ecological psychologist J. J. Gibson even argued that our perceptual systems have evolved to intimately connect the sensorial, and primarily visual, presence of things with an immediate perception of what we can do with them, and coined the term “affordance” to describe what it is that we directly see we can do with a thing (Gibson 1979). Indeed, Gibson’s theory of direct perception is partly grounded in an analysis of variants and invariants in the ambient optic array that the animal perceives as it moves around in an environment. For such perceptual systems to evolve, rich aspects of the animal–environment relationship must remain constant beyond more rudimentary forms of object consistency.

What Is Going On with Things?

11

Certainly, things also change, but they do so in certain ways and not others. A wooden handle may perhaps become worn and smooth over time and use, but even this kind of process of aging is itself fairly straightforward and predictable. Or, we can think of a raincoat that can be used for weather protection but also as a way to block the sun, a pillow, a cover, a container, a way to carry things by using the pockets, and certainly many other things as well. The capability of coming into multiple relations to things through use is based on creative appropriation of a generally stable physical form.1 Even less physically solid things can still be stable with respect to their character. Take, for example, the classic children’s toy “silly putty,” a pliable, somewhat sticky material that can be molded, stretched, torn apart, and smashed back together into a single piece again. It flows slowly if left to sit, and when pressed against newsprint will lift off the ink. It is a remarkable substance (that is great fun to play with), and this is largely due to its complex set of properties and resulting variety of possible behaviors. Yet even with this variety, these core properties themselves remain stable and thus, at least at a basic level, predictable and understandable. Most things are not only relatively stable over time, but also across instances of the same type of thing. Each individual artifact is in some ways unique due to its particular existence in time and space and its own history; yet at the same time, and especially when it comes to things that are mass-produced, it is generally reasonable to expect that similar types of things will be consistent across different specific instantiations. So different kinds of hammers tend to still be quite similar to each other, and hammers of the exact same model will likely be all but indistinguishable, especially when brand new. Indeed, this example would not work as an illustration if this was not the case: that one can evoke an experience of what a thing is and what can be done with it by just saying a word to someone. The same goes for toasters, chairs, lamps, dry erase markers, raincoats, flower pots, bicycles, and many other things that populate our world. In addition to stability, another factor that helps in making sense of things is the fact that the connections between forms and functions are often readily apparent. Indeed, making these connections clear has been considered a central objective in many industrial designs. For instance, the term “affordance” mentioned earlier has made its way into design as well, and although prevalent

12

Changing Things: The Future of Objects in a Digital World

interpretations of affordances in design differ substantially from Gibson’s initial theory, the term is still used to address how to make the intended use of something easy to perceive and understand. Thus, there are many reasons for us to take things for granted. Indeed, most of the time it makes sense to think of them as stuff, simply there to be used.2 While we recognize the difference between natural and artificial, there are actually quite a few aspects of experience suggesting that the two categories are not that different from each other: they are both there for us to be used when and how we want to. In theory, literally, this attitude translates into a treatment of things as already there, and it is only in certain forms of design theory and philosophy that we actually see an account of how they come into being (and continue to evolve). But, as we will argue, it is actually this “coming into being” that is the perhaps most critical aspect of contemporary things. Currently obscured by our persistence in taking things for granted, contemporary things have become dynamic and active, constantly responding and changing over time. While recent technological developments such as an “Internet of Things (IoT)” may have accelerated this development, this story begins much earlier.

What Is Going On with Things?

13

Things have changed The first significant milestone in the change in what things are (at least from the perspective of our present investigation) was the introduction of mechanics, and its miniaturization that eventually followed. While simple mechanical constructions certainly expand the complexity of what a “thing” can be, they maintain some kind of relation between internal and external complexity. That is, a more complex or powerful machine has also tended to present itself as more complex and powerful. The reason is simple: because of the way mechanical constructions work, more complicated ones typically require more space as well. The same holds for power; for instance, a huge steam engine is typically more powerful than a small one. Over time, something intriguing is starting to happen in these relations, between internal and external complexity. As mechanics began to be miniaturized, our immediate perception of external versus internal complexity started to evolve in new directions. An interesting illustration is the historical fascination with automata. In Europe, these evolved significantly during the renaissance, thanks to the development of more advanced and miniature mechanical constructions and clockworks. Often, the intention was to mimic living creatures, to give the illusion of something mechanical being alive; and although we would not mistake any such design for an actual living creature, some of these designs are distinctively uncanny. Still, although certainly very complex, key aspects of the basic relation between visible and actual complexity were still present. Digital technologies introduce a fundamental break with this condition as there is no longer any perceivable relation between internal and external structure, complexity included. Looking at an early digital device such as a pocket calculator, we can of course inspect its interior and, knowing something about electronics, we might be able to locate the different parts, trace connections between the keypad and analog/digital converters, processor, etc.—but we cannot really “see” anything of what it does when it is operating besides the information that has been designed to be presented on its screen. Nevertheless, the pocket calculator is still possible to understand as a “thing,” as perhaps especially its exterior possesses a certain degree of physical permanence. Although mechanics to some extent introduced a more

14

Changing Things: The Future of Objects in a Digital World

complex temporal complexity to what a “thing” can be, even the most complex automata are next to nothing when compared to the temporal flexibility opened up by computational technologies, and the gradual separation of machine and program. After computational technologies came networked computational technologies. While it might at first glance seem that this development amounted to more or less adding communication as another function, the significance of networking capabilities was much more profound in terms of changing what things are and what they do. This meant that the scope of possible actions and effects, as well as resources available, was no longer limited to the immediate physical vicinity of the thing but was rather determined by the topology of the network of which it was part. To be sure, many of the actions supported can be seen as acts of communication. Yet, there is also something else going on when it becomes possible to essentially change the state of another thing or cause it to do something remotely. This is what happens every time a sent message appears in an inbox, sets off a notification, or causes a smartphone to vibrate in a pocket, to take a simple example. We can also think here of all the many financial transactions that now take place via computational systems: everything from personal banking to automated high-frequency stock market trading that now plays a huge role in the world of global finance. We can see that (computational) things are involved here, but it is the connections between them that enable these sometimes extraordinarily complex actions that have very real consequences in the world. An even more extreme and clear example is that of drone warfare. In addition to dramatically expanding the scope of actions that can be accomplished through use of a networked thing, the addition of network capabilities also brought the ability for things to make use of networked resources during their operations. In other words, instead of being more or less self-contained and with functions determined by certain local resources, digital networked things can be composed by components brought together from elsewhere in the network. This can also happen in dynamic, contextually dependent, and evolving ways, such that a particular “thing” is constantly changing both over time and across instances in relation to specific contexts. Things are often now composed on the fly from a variety of networked resources, both physical and digital.

What Is Going On with Things?

15

However, even if they seem quite straightforward in terms of user experience, it is not possible to see how they really function beneath their userfacing surfaces. This is due to both the fact that they are constituted largely by code and also that even the code is generally not accessible to end users. The notable exception here is open-source software; yet even in this case, seeing the underlying code is different from actually watching the processes that are in operation, connections made, data sent, and so on as a programmed thing runs. We can see these dynamics at work when we consider something like a smartphone. It is in a certain sense clearly an object-like, perhaps even beautiful, product of industrial design that we can hold in our hands. But what we actually do with the device relies only minimally on its physical presence as a hunk of metal and glass. Rather, in using a smartphone we make use of the apps and services that it runs locally or, probably more often than not, that connect to other systems and platforms. Also, although smartphones of the same model are all but identical out of the box, customization for a particular user begins as soon as one is turned on for the first time. This involves setting language and location, and creating or signing in to an account that is used across devices. If a person already has such an account that has been used on another similar device part of the same “ecology” (such as Apple’s iCloud), the settings and apps can be carried over directly, essentially replicating the forms and functions of an old device on a new one (even as the new one no doubt has more advanced technical capabilities that provide new kinds of functionality as well). The device will also evolve over time as new software updates become available; apps are added, updated, and deleted; and media content is generated, modified, and removed. Moreover, many of these changes are much more than cosmetic, since they can substantively change what a smartphone can do. It is also important to note that much of this functionality relies on network connections (cellular, wireless internet, GPS) that are activated as soon as the device is turned on. The extent of functionality that is enabled by these connections can be brought into sharp relief when, for example, flying or traveling in a country where one does not have cellular service. What we think of as this particular smartphone, then, in terms of how it is present to us as a thing available for use is dynamic and evolving, and relies on resources not contained within the device itself. And unlike our earlier things, these

16

Changing Things: The Future of Objects in a Digital World

things are made to “know” who is using them and to make intensive use of this information. Taking a step back, it seems that the basic ways things are made have now fundamentally changed. This has happened before. Indeed, entirely new areas, such as industrial design, were called into being to care for the new relations between design and use that became necessary to develop as industrial production drove a change from everyday things being handmade (typically by a craftsman for a known customer—as in how a tailor would make clothes to measure) to being mass-produced for mass consumption. And while aiming for the one optimal design that would be the best for all seemed to work well during the first decades of industrial design, not the least also commercially, the need to cater for individual identity increased in importance; and thus phenomena such as “streamlining,” “branding,” and so on that applied varying cultural and aesthetic expressions to functionally nearly identical products were inevitable. What is important here, however, is that while such moves certainly changed the ways we relate to things, and to consumption as an expression of identity, the basic forms of production still produced reasonably stable objects. Certainly, industry has changed significantly over the past century in all sorts of ways, globalization included; but much of this has not necessarily been that present at the surface level of interacting with products. Taking a new piece of clothing made by a brand we have been using for a long time, we might notice that while it used to be produced in a certain place it is now made somewhere else, that the tactile quality is slightly different as certain materials have been changed, etc.; but typically we see little that challenges our perception of this new thing as yet another instance of, for example, a certain kind of raincoat. At least we do not see any change amounting to the shift from made to measure by someone for someone, to the industrial mass production for mass consumption. And so although production has changed in a number of ways, it has not changed in ways that fundamentally challenge how we basically perceive what the thing is and how it is made and used the way that the shift from craft to industrial production once did. Networked computational things, however, constitute a radical shift from this condition. During the investigations and explorations that led to this book, we have come to think that the change from things to fluid assemblages actually is

What Is Going On with Things?

17

a shift in some ways as fundamental as industrialization when it comes to challenging the way we understand what things are. And again, it is how the things are made—the assembling of things—that is transforming the relations between production and consumption, between design and use. Just as radical is the shift from things that are more or less defined by their physical presence to things that are dynamically defined on the basis of both physical and digital components and in relation to similarly dynamic and interlocking networked infrastructures. What we experience as things are therefore now not so much like the metaphorical tips of icebergs with underlying structures that we cannot see, as might have been said of something like an early pocket calculator, but rather like the top layers of lava flows that span continents and are constantly moving, changing shape, connecting to other flows, and eventually creating new formations. We need to account for the assembling of things in order to understand their temporal forms and their associated relations and consequences, and we need to account for the structure of the resulting assemblages in order to understand their networked, spatial, topological forms and their associated behaviors and capabilities. There are, therefore, compelling reasons not to take things for granted. While prevalent dichotomies between production and consumption, between design and use, enforce perspectives that as a person you are either making a thing or experiencing something already made, the character of contemporary things cut across such simplifications. If we remain with such divisions between making and using, design theory will stay with the former, philosophy of technology with the latter. But as we will argue in this book, these basic dichotomies need revisiting if we are to design and live well with these things. And so, this is where we begin: at the intersections between making and using, trying to articulate and understand things that are unfolding, assembled, and dynamic.

Toward a new account of things Starting to articulate this shift, we suggest that the technical artifacts that now exist in our world are more like fluid assemblages than what we traditionally

18

Changing Things: The Future of Objects in a Digital World

think of as things: assemblages because they are made out of a diverse range of material and immaterial resources both contained within the object as it appears in front of us as well as located elsewhere in the network; fluid because their precise forms are assembled dynamically and thus change continuously. These are quite different from the stable, predictable things that are the traditional objects of industrial design and focus of theorizing about the artificial. In the ways they are manifested, the functions they serve, and the “user experiences” they support, these fluid assemblages can appear to be quite thing-like. And yet, the interfaces that are engaged in use are only the surfaces that conceal the many layers of code, platforms, systems, and interconnections between them that allow specific instances to come into being as things available for use. Moreover, in many cases it is not at all straightforward to determine what constitutes “use” of a thing, and by whom. We can see this, for example, when we consider the many popular web applications in which people essentially pay for the service through having their personal data and attention sold to advertisers. In fact, one of the more notable shifts in moving from stable things to fluid assemblages is that, rather than ending at a single, fixed point of sale, relations between producers and consumers often continue for as long as use of the product continues. In fact, it now seems like this relationship is even more stable than the things themselves, as terms of service often state that these may change over time. From a commercial perspective, considerable effort is put into telling compelling stories about shifts from “products” to “services” to help customers understand the new forms of business. Just think about what the notion of a personal “music collection” refers to now, with services such as Spotify and Apple Music, in contrast to earlier practices of collecting records. Overall, we can begin to see a range of new practices with quite different dynamics than those that have generally governed design and use so far. We can see, among other things, use feeding into continual design and dynamic customization; multiple stakeholders extracting different kinds of value through ongoing use (e.g., value for both users and those who aggregate data regarding use for various purposes); and different arrangements around ownership, with differences also in associated rights, regulations, and restrictions regarding access and use. To give an example: few people anticipated that access to, and aggregation of, people’s internet searches would become the perhaps most important asset in the advertisement business,

What Is Going On with Things?

19

and that a business based on a “search engine” thus would reshape the economic reality of newspapers and by extension that of journalism too. In parallel to these technological developments, there has been across a number of discourses a “return to things.” With the earlier emergence of the post-industrial society and the introduction of the new technologies we associate with it in everyday life, there came an interest in and also significant orientation toward information and communication, the immaterial, virtual and symbolic. Perhaps it is therefore no surprise that it was followed by a “return to things” that displaces the postmodern emphasis on subjectivities and semiotics with attention to objects and materiality. While to some extent certainly a backlash against the general excesses and limitations of the linguistic turn in some areas, and the visions of technological futures turning to tunnel visions of immaterial information societies in others, these new perspectives in many cases also have deeply political and practical motivations. Ranging from pressing issues of sustainability and how to live within the constraints of finite resources to the realization that also computation and communication are things literally taking place, the “real” world continuously calls for our attention, as does the very “real” experience of being a body in the world. In many ways, the synthesis of the thesis and its antithesis in ideas such as that of “an Internet of Things” was therefore inevitable. And so we return to “things,” this stuff that we create and that according to some is what defines us and the time we live in: the artificial as the effective horizon of human existence, the Anthropocene. And yet, as we can see by even the brief survey of technological developments earlier, we are now dealing with a rather new set of dynamics and logics around things and their production, use, and functions. This state of affairs leaves us with a complexity and a rupture that we do not yet fully understand but that we, as a start, need to try to articulate. Or in other words, what is now needed is to move beyond thinking about the present discourse as a “return to things,” since in fact what we now encounter is something we have never lived together with before. This is certainly not a return to something we were once comfortable with. We need a new vocabulary and theoretical/analytical approach that will allow us to speak to the multifaceted, dynamic, constantly emerging and evolving forms and functions of contemporary things. The aim of this book is to sketch out such an approach, and in doing so to prototype a discourse around things as fluid assemblages.

20

Changing Things: The Future of Objects in a Digital World

Notes 1

This ability to come into multiple distinct relations to things is referred to as multistability within the subfield of philosophy of technology known as postphenomenology (e.g., Ihde 1990). In this context, it serves to counter technological determinism and more substantive theories of technology by highlighting the capacity for human agency and creativity in relation to things.

2

This also aligns with philosopher Martin Heidegger’s diagnosis of how our relation to technology causes us to see the world (including possibly other people and even ourselves) as bestand, or standing reserve, simply there as a resource to be used (Heidegger 1977).

3

Just Press Play, Please In what follows we will take another look at the historical developments outlined in the previous chapter, but this time as seen through an example: the very simple and mundane act of pressing a play button to listen to music.1 While techniques for music playback have quite a long history, there has been in just the past decade or so an explosion in the development of technologies and services for listening to music. Yet, even with such a diverse array of music-playing technologies historically and in terms of currently available options, some things remain constant. We approach a music-playing thing because we want to hear music; and whether we find it on the plastic button of a tape or CD player, the click wheel of a classic iPod, or in the interface of a digital app, we know to look for the familiar right-facing triangle icon. To hear music, we press play. This simple act of pressing play has remained quite consistent, even as the complexity of the underlying systems that make the playing possible has increased tremendously. Indeed, many digital music players are now only one component of vast ecosystems including digital service providers, musicians, record labels, advertisers, and other digital platforms. Moreover, they participate in these ecosystems in much more complex and dynamic ways than their simpler historical predecessors. We will here consider simply the assembly or activation of a music-playing thing such that its functionality of pressing play is made available to us. We will trace the nature of this assembly through a variety of cases, beginning with older analog technologies and watching for significant changes as we move on to address more contemporary technologies and (eco)systems. Through these illustrations, we will introduce another key idea in this book: while our new massively data-driven and networked computational things are inherently different from previous “things,” it seems that what designers

22

Changing Things: The Future of Objects in a Digital World

aim to express and users expect to experience in use are not at all this radical disruption but rather something familiar and in line with what we have lived with before. In other words, what we tend to search for and expect in use is something of a paradox, namely that the complexity of technology is resolved through a simplicity of expression. This is a design paradox we need to unpack as it sits at the center of what designing and living with fluid assemblages is like at the moment. To understand the background of this tension between technical complexity and keeping things simple, we need to turn back to the developments in industrial design in the early 1950s and the search for clear and functional design of (technical) objects on one hand, and a growing interest in increasingly complex products and systems on the other hand. At the design school HfG Ulm, we see early examples of both the kind of industrial design that is still today highly influential in the technical domain (e.g., the HfG– Braun collaboration about the design of radios and other music systems for the home) and the first steps toward making design an interdisciplinary project set up to deal with increasingly “wicked problems” (e.g., diagrams and graphs inspired by mathematics used to visualize design and its processes). While the orientation toward simplicity stems from a concern for usefulness and utility, it is more than anything also an aesthetic orientation. Indeed, if there is one well-known phrase that captures the Modernist aesthetic, it would probably be “form follows function.” To say that it is an aesthetic orientation is not to diminish its scope, but to suggest that it is a central part of a “design worldview” inherently tied to a certain way of thinking and creating design expression. Of course, there are many other aesthetic orientations, but in industrial design as practiced in the parts of the world where these new devices are often conceived, this worldview is still something much deeper than just one out of many orientations one may choose between while designing. And while it largely has given way to other orientations, and especially more diversity, within many domains of design, it still exercises a massive influence in the domain of tools and technology. Indeed, notions such as “ease of use” and “usability” are, within the domain of computational things, still central qualities sought. The concern for complexity also has its roots in early Modernism, and the idea that there are important relations between design and society, between individual objects and industrial systems. Even in early examples of industrial

Just Press Play, Please

23

design, we see an explicit interest in how objects are related to each other in systematic ways, and how design can help address and make sense of the resulting complexity. Indeed, it is almost as if the orientation toward simplicity in expression coevolves with the increasing complexity of function and the systems the things exist within: the more complex the functions and the relations between things, the more important to keep things simple. From a design point of view, this may seem like an, or even the most, obvious response, which makes it even more important to unpack it as the aesthetic orientation it actually is. Because there is nothing given, necessary or “natural” in the domain of design. The artificial is, through and through, inherently something made by someone—and thus something that always could be made differently. Now, let us trace the simple act of pressing play to see how things have indeed changed.

A short history of music players Beginning with pre-digital examples of music-playing technology, we can think of classic record and tape players. These devices are quite respectable design objects in the classic sense: things that are mass-produced in factories and then purchased by users who then own and can do whatever they like with them. Playing music on them requires plugging in the power cord and perhaps pushing a power button, loading a record or tape, and then pressing play. Especially in the case of a record player, it is typically possible to manually intervene at more or less all stages of the mechanical process as well, such as manually lifting and positioning the arm with the needle onto the record, stopping or slowing down the rotation with the hand, etc. All of the assembling of these things occurs on the manufacturer’s side before they reach the end users, and unless he or she decides to physically modify the device it will remain the same. Although the mechanics involved are somewhat harder to inspect, this logic also applies in the case of stand-alone CD players. However, when we move to considering a CD played in a computer rather than a stand-alone CD player, we notice some different dynamics emerging. First, and most obviously, computers do much more than play CDs; this is only one of many functions they have which are managed by the underlying

24

Changing Things: The Future of Objects in a Digital World

operating system and installed software. And software is in fact needed to play the CD. The physical CD drive may not seem much different than that of a classic CD player, but the fact that it is now operated by software marks a key shift. There are now multiple software options that can be used in conjunction with the same CD drive—ones which can be updated and configured independently of the underlying hardware, and thereby change overall functionality. Also, and in contrast to the elegant simplicity and functional transparency of a record player arm lifting up and over a record, loading a CD in a computer launches countless computational processes that are not generally visible (although it might be possible to use system monitoring tools in order to see some of what goes on). This would often include a query to a music database service such as Gracenote in order to retrieve the track names for the CD, revealing another (networked) component of the assemblage. CDs can also be ripped and stored in a computer hard drive, marking another key development when music can be stored in digital formats and played without the need to load an external storage device. And of course they can also just start in digital format and be distributed without the need to ever involve physical storage media other than computer hard drives. As music-playing things, digital music player applications appeared as a rather new kind of animal. They are “assembled” from a variety of components when the application is launched. These include the code for the application itself and the underlying operating system that manages its processes, including sound output (which might be internal or peripheral speakers). The computer itself is the component that seems most object-like in a traditional sense; yet even with this simple example we can see that only a small part of its functionality is determined through its assembly in a factory. Much of what a computer does is rather determined by its operating system and applications, which can be updated and configured in ways that can greatly change functionality without changing the underlying hardware. In addition to the assembling that happens through basic software, it is also possible to further modify an application’s “assembling” as a thing available for use through configuring its settings. These can change how it behaves and how it looks. One such example are various forms of automatically generated playlists, ranging from “shuffle” functions first known as “random” playback order in CD players that mix up the predefined playback order to

Just Press Play, Please

25

more elaborate algorithms based on categorizations, tags, and other kinds of metadata attached to the song. Another example might be how the classic WinAmp player allowed for customization through “skins.” Moreover, anyone could develop these skins and make them available for others; and branding them with one’s logo could become a point of pride for their creators, serving as a visual reminder that it was another person (and not WinAmp) who created this particular component of the user’s personal WinAmp assemblage. At the same time that the MP3 file format and players such as WinAmp gave people much more freedom in terms of how they could play and distribute music, other trajectories sought to restrict the ways in which people could acquire, listen to, and distribute music, even as they also capitalized on the possibilities of the digital. The most significant player in this regard is arguably iTunes, with its “walled garden” approach to providing a coherent and seamless user experience while also ensuring that only certain kinds of “acceptable” use are possible. It is well known that the possibility to make infinite duplicate copies of music files without loss of quality and to easily and widely distribute them via the internet posed a significant challenge to existing structures in the music industry, and led to the emergence of new sociotechnical configurations that is still ongoing. However, even as these dynamics have driven the development of many more contemporary music-playing systems, our concern here is with the ways in which these are assembled and appear as things available for use.

Music and metadata A key development associated with iTunes and the iPod music player was in structured metadata associated with media files. This was clearly visible in the iPod in particular, where music could be accessed in multiple ways, through artist, album, genre, playlist, etc. Significantly, iTunes also included metadata reflecting usage, such as play count, skip count, and date last played, as well as data about when the file was added to the iTunes library and last modified. This arguably marks the beginning of the evolution of music players in which usage affects the future constitution and behavior of the system. This can be seen in a single track itself that has updated metadata, and in the resulting

26

Changing Things: The Future of Objects in a Digital World

ways in which tracks are displayed when sorting according to these variables. However, it also works in a more subtle way by affecting the frequency with which tracks are played on “shuffle” mode in both iTunes and synchronized devices (such as the various iPod models and now the iPhone). Both this personalization of the iTunes data and experience and the enforcement of certain usage restrictions are accomplished through accounts. Accounts have now become quite common and effectively extend the relationship between producer and consumer for as long as use of the product continues. Even web-based music players that do not require accounts track users and customize the offerings in fairly sophisticated ways. For example, the Last.fm music player web page (http://www.last.fm/listen) loads a variety of trackers, beacons, and analytics that, as of this writing and as revealed by the Ghostery browser extension, include ones for Audience Science, BlueKai, ClickTale, DoubleClick, Google Analytics, Omniture (Adobe Analytics), Qualtrics, Spotify Embed, and Yahoo Analytics. Refreshing the page or connecting from different locations also updates the musical suggestions provided. However, the extensive and fluid assemblage of Last.fm is rather disguised by an interface that invites the user to simply “type in an artist or genre and press play.”

Just Press Play, Please

27

From playback to runtime assembly A significant aspect of this continuing relationship between providers and users is that “use” can be precisely scripted and either enabled or limited in dynamic ways. For example, use can be customized or restricted based on location. On a basic level, detecting the country from which a person is connecting to a web-based service allows for language customization and for presenting what is most popular in that country. But it can also be used to restrict access to certain content or prevent access entirely, as in the case of Songza that could not be accessed through internet connections coming from outside the United States or Canada (and has now as of this writing been incorporated into Google Play Music). These restrictions can sometimes, however, be bypassed by connecting through a VPN service—another component that can be brought into the assemblage on the side of “use.” It is interesting to note that when we reach this situation of dynamic customization there is no longer any single, stable “object” that can be viewed “objectively.” Instead, what is stable across users is the set of rules and processes governing the ways in which the product is constituted at runtime for specific accounts connecting from certain locations at certain times—although even these rules themselves change over time. Indeed, one of the most prominent aspects of modern web-based music players, such as Deezer, Slacker, and Pandora, is how they adapt their music recommendations over time based on what individuals listen to and indicate that they like. Importantly, these “things” can also be continuously disassembled. For instance, streaming content providers may stop making certain content available. Starting to use Spotify on one device will stop playback on another device. Content may also stop being available because of changes in the governing legal contracts, as when the music of an artist from one day to another is no longer available as a new commercial agreement could not be reached. One can also experience the geographical specificity of such legal agreements when traveling, as some content is available in some countries but not in others. A major dynamic in this runtime production and customization of musicplaying things is that not only are they assembled dynamically, but the components assembled come from a variety of sources. One way this can be

28

Changing Things: The Future of Objects in a Digital World

seen is in the many examples of services that load ads in conjunction with the application. These ads themselves represent the complex and extensive assemblages of advertising services, such as Google ads or Apple’s iAd program. From a slightly different angle, many services now allow for authentication through social media accounts (such as those of Facebook or Twitter) and also connect to the functionality of these accounts in other ways (e.g., loading Facebook friends into a “friends” list, or enabling the sharing of one’s activity). The assembling of the music-playing things and their functionality in these instances is enabled and constituted partially through these other services. Indeed, quite a few reasons behind the particular design of some of these assemblages are related to the shift from a focus on consumer purchases to selling user data in many business models: since what is “sold” is not a “thing,” but data about the user that can be used to, for instance, customize advertisement and direct users to certain other services, gathering as much such data as possible becomes a key driver. This is a major reason for the increasing importance of accounts to access music, but it can also be seen in the extensive user profiling and tracking in services not necessarily requiring a login. For example, SoundCloud’s cookie policy (https://soundcloud. com/pages/cookies) describes how, in addition to their own cookies used for managing sessions, they use a number of third party services (from Google, AdsWizz, Global Radio Services Limited, Quantcast, and Scorecard Research) that provide analytic and advertising functionality. They also use the “similar technologies” of Clear GIFs, Flash cookies, HTML5 local storage, activity tracking (“in-app analytics and messaging service provided by Appboy, Inc.”), app performance tracking (“adjust” service provided by Adjust GmbH), and bug reporting (“Crashlytics” service provided by Crashlytics Inc).2 Another general source of input for runtime customization is users themselves. One way this works is through application settings, but there are also a number of other means by which use of a thing can later feed back into how it is assembled. As previously noted, simply recording which music tracks are listened to can affect how music can be sorted and displayed. This allows for features that display the artists, tracks, etc. that a person has listened to the most. Recent listening was placed front and center in the Rdio3 online application, turning activity into the main content of the site in the form of a

Just Press Play, Please

29

collection of album artwork representing a timeline of recent listens. And of course listening activity also feeds back into the recommendations provided later on. Finally, it is interesting to note the extent of the shift from buying something and then really and truly owning it to using things to which one has access only provisionally. Systems can be upgraded or, more neutrally, modified without users’ consent. Content and features can be added and removed. Use is regulated through a mutually reinforcing combination of system architecture and law (Lessig 2006) such that, for example, customization and restriction of a web-based service based on location are reinforced by stipulations that users must not try to circumvent them. Instead of operating manuals, users are now faced with sometimes staggeringly extensive terms of service (which are perhaps even less likely to be read); and they must accept these before gaining access to the system, thereby entering into standing legal agreements of which most typical users have only the faintest understanding. Yet, these terms of service sometimes contain dire warnings and regulations regarding use, such as the Google Play terms of service that states in part (https://play.google.com/ intl/en/about/play-terms.html): NONE OF THE PRODUCTS ARE INTENDED FOR USE IN THE OPERATION OF NUCLEAR FACILITIES, LIFE SUPPORT SYSTEMS, EMERGENCY COMMUNICATIONS, AIRCRAFT NAVIGATION OR COMMUNICATION SYSTEMS, AIR TRAFFIC CONTROL SYSTEMS, OR ANY OTHER SUCH ACTIVITIES IN WHICH CASE THE FAILURE OF THE PRODUCTS COULD LEAD TO DEATH, PERSONAL INJURY, OR SEVERE PHYSICAL OR ENVIRONMENTAL DAMAGE.4

— Pressing play has become serious business indeed.

Compositions and breach of contract In earlier work, acts of defining what a given thing is were discussed based on a distinction between acts of design and acts of use (Redström 2008). Consider a glass bottle for instance. Acts of designing—of making as craft—a glass bottle would be acts such as preparing the material, heating the glass, blowing and shaping it, and cooling it. Acts of designing a bottle for industrial production would instead entail acts of producing a prototype that can be mass-produced,

30

Changing Things: The Future of Objects in a Digital World

through sketches, models, etc. While the process of making the bottle as such can differ, there is still a clear distinction between such acts of defining what the “bottle” is, and what then happens as we use it. Use is still a matter of defining what the bottle “is” (to us), but these acts will be based on the fact that the bottle is there for us in its physical form. And so I may use it to contain fluid that I can drink, thus defining it as a drinking vessel, but I can also use it to express my feelings by throwing it to the wall, thus (re)defining it as a kind of prop in a performance of sorts. It can be used as a small window in a cottage I am building, thus defining the bottle as a kind of building material. Eventually, it will perhaps be used as material for making a new bottle, thus closing the loop. In general, we might say that there are potentially a range of different acts defining what this thing is, but that they basically fall into two categories: one of “design” causing the thing to come into being, and one of “use” bringing the thing into a practice for some purpose. This distinction has been the stable basis for a kind of social contract established between design and use that, on one hand, allows “designers” to create objects for intended forms of use and intended users, and for “users” to acquire, interpret, and make use of these objects for their own purposes based on the typically predictable and stable properties of the objects as present physical things in their lives. When it comes to the fluid assemblages we now design and use, this basic picture is breaking down. And importantly, the basic social contract between design and use is becoming increasingly problematic as the underlying premises for that contract are being replaced by new forms of making and using. This causes a rupture that we do not yet know how to address, but that we, as a start, need to try to articulate. The basic cause for this change is that the fluid assemblage is never really made, at least not in the common sense that a bottle is made. The fluid assemblage is continuously in the making, intertwining acts of design and acts of use over time in ways that traditional mechanical objects certainly cannot. Yet, as we tried to show with the examples earlier, many “things” do their best in keeping up appearances, maintaining that the basic contract is still valid, and that the basic relations between designing and using are still in place. Looking at contemporary design, there is in many cases no single, uniform, consistent, stable thing when it comes to design objects (Wiltse, Stolterman, and Redström 2015). Rather, as we use computational and other digital materials,

Just Press Play, Please

31

the composition of things is determined on the fly according to a potentially infinite array of constantly shifting parameters and operations, many of which are hidden. This new kind of complexity we are now facing is something rather different, even as it recasts a classic and related tension between simplicity and complexity in new ways. The complexity that stems from dynamics of use in a social context was in a way external to the things themselves. The composition of a thing and the composition of the systems in which it was embedded were closely related, but also possible to separate. Now we are in a situation in which the composition of a thing at any given moment is determined in non-trivial ways by variables that are external to the thing itself. It appears almost as if we are heading toward a blind spot, where certain issues are occluded by our prevalent perspectives, and where we therefore need to develop new accounts of the basic “what” it is that we design. This leaves us with a problematic gap between existing frameworks and emerging design issues.

Notes 1

This chapter is based on the paper “Press Play: Acts of Defining (in) Fluid Assemblages” presented by the authors at the Nordes 2015 conference.

2

https://soundcloud.com/pages/cookies, accessed December 20, 2017.

3

Rdio has now been acquired by Pandora, and is no longer available.

4

The all-caps formatting has been removed since we first noticed this text, but the content remains otherwise the same.

4

Fluid Assemblages In the previous two chapters, we have explored aspects of why and how our computational networked artifacts are different from what we have encountered in the past, although they still appear to be “things” to us. Indeed, we have argued that how we currently design and live with them is something of a paradox, as we on one hand persist in calling these out as “things” while at the same time acknowledging that they are not at all like our things of the past. In what follows, we will try to unpack this seemingly contradictory position and how it might have come about.

Assemblages The notion of assemblage used here stems from the work of Gilles Deleuze and Félix Guattari (Deleuze and Guattari 1987). This is a difficult concept to capture (and it was probably never meant to live a life in intellectual captivity). Addressing the agency of assemblages, Jane Bennett offers a very good first presentation of what these things are like: Assemblages are ad hoc groupings of diverse elements, of vibrant materials of all sorts. Assemblages are living, throbbing confederations that are able to function despite the persistent presence of energies that confound them from within. They have uneven topographies, because some of the points at which the various affects and bodies cross paths are more heavily trafficked than others, and so power is not distributed equally across its surface. Assemblages are not governed by any central head: no one materiality or type of material has sufficient competence to determine consistently the trajectory or impact of the group. … And precisely because each memberactant maintains an energetic pulse slightly “off ” from that of the assemblage,

34

Changing Things: The Future of Objects in a Digital World

an assemblage is never a stolid block but an open-ended collective, a “nontotalizable sum.” (Bennett 2010, 23–24)

There are a couple of key points to make before turning to our things that change. Looking at different ways that components can be drawn together, the notion of assemblage sits between collection and totality. In contrast to how a collection is a gathering of objects that does not have properties beyond the parts included in the collection, an assemblage is characterized by emergent properties. An assemblage cannot be reduced to its parts. But whereas a totality also is irreducible, the assemblage differs from it in that it can still be taken apart. An assemblage is decomposable. In other words, whereas the collection does not achieve any additional emergent properties or capacities, and a totality cannot be taken apart, an assemblage is characterized by both emergent properties and that its components retain their identity. To better understand what makes emergent properties come about, and thus what creates the difference between totality and assemblage, we need to look into how the components are related to each other. In Deleuze and Guattari’s account, a totality is characterized by interiority relations; that is, the components change as they become part of the totality. This implies that they, typically, cannot turn back into what they were before becoming part of the whole. The emergent properties of the totality are therefore “transcendent”: a totality moves above and beyond its constituent parts. An assemblage, in contrast, is made out of exteriority relations between components, that is, relations between the components that do not change their individual properties or otherwise transform them. This makes it possible for an assemblage to be taken apart (or otherwise transformed), and for the components to retain their properties afterward. Thus, the emergent properties of assemblages are “immanent,” a result of continuously interacting components that will disappear should those interactions come to an end. If we turn back to the things addressed in this book, even this very brief introduction of assemblages illustrates the effectiveness of the concept in this context. The examples of what happens as components start to interact with each other have shown that we are talking about much more than collections here. The smartphone is not just a collection of parts; it is an assemblage of apps, hardware, network connectivity and so on and so forth with capacities

Fluid Assemblages

35

way beyond the sum of its parts. At the same time, however, it is clear that we can take this assemblage apart, at least to some extent. If our phone had been a “seamless totality,” for instance, removing an app or changing networks would not have been possible as there is no reverse. Indeed, the notion of immanence is a most present aspect of their character: as soon as the interactions between components do not work, the emergent properties of the assemblage change or disappear completely. Just consider what happens when connectivity drops, the battery runs out or the device makes itself unavailable for further interaction because of a software update. From a more philosophical point of view, the addition of “fluid” to assemblage is completely superfluous. An assemblage is in its very nature fluid in one way or another; this is just another way of saying that its emergent properties are immanent. The reason for talking about “fluidity” here is to put an emphasis on how speed has become such a significant factor. While assemblages such as a community, an organization, or a machine are also constantly in becoming, they typically change at the pace of human perception or in many cases even slower. Manually building a machine takes time, as does taking it apart. These computational networked machines, however, create their particular instantiations in runtime. The conditions for their coming into being have of course been created over longer periods of time, but the assembling as such is literally happening during use—and in response to how use unfolds down to events happening at the level of microseconds. They have a fluidity that extends way beyond what we have seen in terms of speed before—and not only in terms of processor clock speed or network transfer capacity but increasingly also in terms of “time to market,” frequency of updates, design methodology, and so on and so forth.

Design methodology The fact that these things are already a significant part of our everyday lives seems to suggest that design (in the wide sense of word) has already resolved much of what is involved in making them. Indeed, in terms of technology their coming into being has been underway for quite some time, with inventions such as object-oriented programming and massively networked computing providing not only the means but importantly also the conceptual

36

Changing Things: The Future of Objects in a Digital World

frameworks necessary to conceive them. As industrial design started out with an ambition to create a new unity between art and technology, we could perhaps think of design as engaged in a spectrum ranging from the more technical (toward engineering) to the experiential (toward the artistic). If we look at how design has responded to the condition brought about by fluid assemblages across this spectrum, we can see that we are pushing toward development at one end, whereas we seem to be holding back at the other end. If we start in the more technical end of this spectrum, we find several rather recent developments that respond to the design condition posed by fluid assemblages. Relevant examples include the methodologies for “lean” and “agile” development. While their origins can be found in the lean engineering pioneered in Japan by Toyota and frameworks such as Kansei engineering, these ideas have more recently come to a different expression in software development and associated industries. These methodologies depart dramatically from earlier models for product and systems development in which things would only come together toward the end, after often long processes of development and optimization. Instead, these approaches rely on short sprints where each and every effort is meant to produce results that can be shipped as products. Thus, instead of producing one major “new” product every year or so, the product is updated continuously, sometimes even on a daily basis. Design methodology has relied on iterative prototyping to resolve complexity and uncertainty since its very beginning. To make prototypes is for design a central part of most learning processes, not just in the general sense of learning by doing, but more specifically because there are very few ways of learning how a certain “whole” comes together without actually working its tensions and conflicting elements out. In this respect, lean and agile are not that different from earlier practice, and can even be celebrated as a learning process that allows for failing early and often. However, they also differ substantially from how design has been done before. Traditional design methodology is completely conditioned by the requirements of mass production, and as such its learning process is directed toward optimizsation, toward understanding what needs to be done, and then resolving all the issues involved in getting there.

Fluid Assemblages

37

The logic behind this process is quite obvious: when you are about to massproduce something, you better make sure it is the right thing you replicate over and over. The logic of lean and agile processes is quite different. While they too are about resolving uncertainty by means of iteration, they are not at all oriented toward this kind of optimization. Instead they orient toward notions such as the Minimum Viable Product. It is no longer about first finding the best possible solution and then moving to market, but rather to make the entire development process into a dialogue with market: by constantly releasing updates and making extensive use of user data, progress is tracked and measured, to form the basis for decisions about how to make the next move. On the user side, there is for instance frequent use of A/B testing: by releasing two different versions (hence the A and B) to different groups of users and then tracking which version yields the best response or result, design decisions can be based on actual data. And by rolling out such tests extremely frequently, perhaps even several times a day, the distance between designing and using can be shortened dramatically compared to when in the past it might be even years between when the designing first happened and people could actually use the thing. To see the vast difference between the design philosophy behind notions such as the Minimum Viable Product and what ideas have historically governed industrial design, compare the following two reflections: the first from Eric Ries, one of the main proponents of lean methods for start-ups; the second from Max Bill, first student at the Bauhaus and then later Headmaster at the HfG Ulm, the school that introduced the industrial design aesthetics that still dominates much technology design: At this point in our careers, my cofounders and I are determined to make new mistakes. We do everything wrong: instead of spending years perfecting our technology, we build a minimum viable product, an early product that is terrible, full of bugs and crash-your-computer-yes-really stability problems. Then we ship it to customers before it's ready. And we charge money for it. After securing initial customers, we change the product constantly–much too fast by traditional standards–shipping new versions of our product dozens of times every single day. We really did have customers in those early days–true visionary early adopters–and we often talked to them and asked for their feedback. But we

Changing Things: The Future of Objects in a Digital World

38

emphatically did not do what they said. We viewed their input as only one source of information about our product and overall vision. In fact, we were much more likely to run experiments on our customers than we were to cater to their whims. (Ries 2011, 3–4) For around a hundred years now the call to action has sounded in successive ways throughout the world: we have a duty to make useful, ethical products that are true to materials and manufactured under socially responsible conditions, using the best means available to us. Inherent within this call is a sense of moral responsibility, a social understanding. /…/ This makes us realise that what we’re actually striving for is something quite different—namely, an extreme utilisation of materials, where the maximum effect is achieved with the minimum of materials. For example, we can construct a tower 300 metres high (the Eiffel Tower) and make it so light (as Eiffel did) that if its height were reduced by a factor of one thousandth, ie by 30 centimetres, then its weight would drop by just seven grams—the weight of a pencil. This is a shining exemplar of the extreme exploitation of materials, an emblem for the technical age and the rational use of materials, as well as the germ of a new ideal of beauty. (Bill 2010a, 32–33)

These reflections come from different places and points in time, and one should be careful when comparing them. Still, they illustrate a couple of key points made here regarding design’s relation to the technologies it is working with. While the notion of making the most out of existing resources is a central concern, how to actually achieve that has fundamentally changed as people are now trying to cope with a different kind of complexity. While actual practice may not be quite as extreme as Ries’s account mentioned previously (after all, many companies that work with lean processes also have to comply with other requirements and certifications that prevent them from releasing things that are not functional), it points to a significant shift from refining something until it becomes the optimal design solution to instead trying to expose minimal products to a market as quickly as possible. In a sense, this means that the former so important difference between prototype and product is disappearing. As such, the shifts in perspectives, not to mention values and objectives, also illustrate the radically different configuration of production and consumption that these new technologies and associated

Fluid Assemblages

39

industries have introduced. Compared to how things used to be, it may seem as if we live in a constant “beta,” but more than anything, what we see here is a completely different configuration of the relations between designing and using.

Design aesthetics As we turn to how design addresses human experience, to how these things present themselves to us and become part of our practices, it is less obvious that we have developed design to care for this condition. On the contrary, much of what is going on over here seems to be about reassuring us that things have not changed that much after all. Indeed, we may ask: how is it possible for these complex assemblages to be experienced as “things,” even as simple things that we do not hesitate to let into our lives? To get started, we have a working hypothesis regarding how fluid assemblages come to present themselves to us as stable, contained, and predictable things: Fluid assemblages make use of mimicry: to appear as “things,” they camouflage themselves as totalities. The distinction between exteriority and interiority relations allows us to probe deeper into how the “thingness” of these assemblages is achieved. To see this, we first need to attend to how design typically has come to approach the relation between intended use and aesthetics. The perhaps primary task of industrial design is to bring about a meaningful whole. Design is, in this sense, the opposite of analysis: whereas analysis treats complexity by means of taking apart into ever more manageable parts, design is fundamentally propositional in how it aims to resolve conflicts by iteratively prototyping and proposing how a unified whole could be brought together. Much of what is treated in design aesthetics (as distinct from in Aesthetics in the philosophical sense) is about how to create such totalities, often articulated in terms such as consistency and coherence, as gestalt or as a unitary and consistent use experience. While it may be obvious that design is about creating the conditions for emergence in that sense—of making the sum greater than the parts (and the greater the difference, the greater the value!)—we may still ask whether this is really about seeking totality. Certainly, this question is one of nuance and difference, but if we turn to the present and pressing issues of sustainable

40

Changing Things: The Future of Objects in a Digital World

development we get an indication. Had design’s notion of a whole been fundamentally oriented toward assemblages, then we would expect there to be an awareness of how the constituent components retain integrity after the assemblage is taken apart. Should this have been the case, new notions such as “recycling” would not have been needed, as taking things apart and using them for other assemblages would have been an obvious consideration.1 On the contrary, it seems that as technology matures, it moves toward totalities. Working with the first computers, users had to know how to replace a radio tube and other parts that might break; users of contemporary computers may not even be able to replace the battery. An old car will expose most parts of its engine, mechanics, etc. as you open the hood; in a contemporary car most of it is likely to be covered by plastic to indicate that this is the domain of the experts in the workshop (who in turn typically rely on complex computational diagnostic equipment to uncover what is not working). Certainly, there is modular design, open design, and a range of other approaches that genuinely engage with design as assemblage—but overall, it is probably fair to say that design aesthetics have a basic orientation toward meaningful wholes understood as totalities rather than assemblages. To a significant degree, this design principle put forward by Max Bill in “The basis and aim of aesthetics in the machine age” from 1953 (thus around the time when he took on the position as headmaster of HfG Ulm) still holds for much industrial design: The basis of all production should be to fulfil, as a unity, the totality of all functions, including the aesthetic functions of an object. (Bill 2010b, 70)

It is worth a short reflection upon why this was, for the industrial technology of the time, a very reasonable and effective principle. The emergence of industrial design is completely conditioned by the shift from manual to machine production. Giving form to things prior to this shift was a matter of craft, most of the time working with just one or a few materials, and with objects that typically would have one or maybe a few key functions. Importantly, these things were also most of the time made by just one or a few people who would be able to lead the process from material to finished object, thus able to make sure that the outcome was a “whole,” and not just a collection of parts coming out of otherwise separate processes. With industrial production, this changed.

Fluid Assemblages

41

Indeed, it is called assembly line for a reason. And further, industrial design involves working with many different materials, larger sets of functions, and over time therefore also increasingly complex things. In this context, to achieve something that truly comes together in the sense that the traditional object is given form and made by craft is far from trivial. Thus, how to achieve such a seamless totality became the perhaps most central aesthetical issue when trying to find ways for the “new” to not only (unsuccessfully) mimic the “old,” but to offer something that would, literally, transcend it. The once intended effect, but now increasingly problematic consequence, is therefore that industrial design typically lets interiority relations completely dominate over exteriority relations when thinking about how form defines the way material builds things. In this way, both designers and users become oriented toward things not only as “meaningful wholes,” but as totalities. Not only in the sense of how they are made and built, but perhaps even more so in terms of what kind of use experience designers aim for and what kind of experience users typically expect and appreciate. This disposition has three significant implications with respect to following established aesthetics when designing fluid assemblages. First, it means that we as designers will seek to, literally, transcend the experience of an assemblage and push it toward the experience of a totality. For instance, this means we will try to more or less mask exteriority relations, and instead emphasize how the components have fused into a new seamless whole. Second, it means that we as users feel most confident when we only need to attend to a whole. Most certainly, we do not want to pay too much attention to all the interactions taking place between various components. While we might realize it is an assemblage, we certainly do not want a huge part in its immanence. On the contrary, we prefer if it is stable and unified, and it is only rarely that we will be interested in committing to the effort required to maintain and manage complex interactions between constituent parts. These two implications bring about the third, and most critical one, namely that we therefore come unprepared and unaware of just how different a fluid assemblage is from a traditional object—and that our current ways of working with design expressions to a significant degree keeps this hidden from us. In other words, there is presently a rapidly growing design space that designers do not quite know how to deal with in terms of aesthetics, or perhaps even care about from an aesthetical point of view.

Changing Things: The Future of Objects in a Digital World

42

So, we find ourselves in a situation where we on one hand have technologies and the beginnings of new design methodologies for making fluid assemblages, but where we at the same time persist in making them appear as normal “things.” Like Theseus’s ship, our new things are both the same and something very different.

Note 1

Just consider open source software as a counterexample: certainly one could think of it as a large recycling center as pieces of code are made available to be used to other things, but you rarely see this articulated as a matter of “recycling”—it is quite simply about sharing and contributing code for also others to use, and this relation to the code produced is an inherent part of its making.

5

Things for Us There can be seen two main orientations to considering things.1 The first, and most common, is to look at them in relation to human activities in which they are involved. In other words, the basic concern here is what a particular thing is for humans as they use or otherwise relate to or are affected by it. The second is to inquire into what things are “in themselves.” Of course we can never really escape our own situated perspectives as humans, but we can try to get to the bottom of things through an investigation more oriented toward basic ontology. These orientations have through history generated a vast range of contributions that can be productively brought to bear in making sense of contemporary things. While we may realize that we need to build on work from both strands, as one of the things that we discover when exploring fluid assemblages is that a framing of human (or “user”) experience is not sufficient for understanding what they really are and do, we may still believe that such a combination of multiple perspectives will, in the end, result in a quite thorough coverage. Yet this is a problematic assumption. Different analytic frames have conceptions of their objects of study that can actually be incommensurable, resting on different foundational assumptions and orientations. As Deleuze and Guattari state: “A concept always has components that can prevent the appearance of another concept or, on the contrary, that can themselves appear only at the cost of the disappearance of other concepts” (Deleuze and Guattari 1994, 31). Foregrounding certain aspects inevitably means backgrounding others. Another concern is that perspectives honed on more traditional objects of study will miss or not properly account for much of what is relevant about contemporary ones. There are also more fundamental reasons to care about the consequences of the conceptual frames used. Building on the “philosophy-physics” of Niels

44

Changing Things: The Future of Objects in a Digital World

Bohr, Karen Barad goes even further in her agential realist account. As she states: Discursive practices are specific material (re)configurings of the world through which local determinations of boundaries, properties, and meanings are differentially enacted. That is, discursive practices are ongoing agential intraactions of the world through which local determinacy is enacted within the phenomena produced. (Barad 2007, 820–821)

In other words, discursive practices do not merely describe phenomena, but participate in their production. This foundational role of conceptual and discursive frameworks thus entails a corresponding accountability for the ways in which we conceptualize, analyze, and discuss the things and phenomena with which we engage. Proceeding with care, then, there are two main themes we need to pay attention to in considering existing and potential ways of trying to understand changing things. The first is to see what existing bodies of literature give us: tools, perspectives, sensitivities, and so on. The second is that we need to be careful to notice aspects of contemporary things that existing perspectives do not handle well: significant characteristics and dynamics that are not accounted for, conceptual tools that are now clunky rather than incisive, underlying assumptions that are no longer valid, black boxes that have become too large and conceal too much, and basic orientations that point away from where much of the action is. In this chapter, we will focus on general orientations to things as they are involved in human activities: action and perception, communication, and networked computation. The latter, however, starts to move into a more explicit focus on things, and is essential for making sense of fluid assemblages. This includes scaling, looking at the range of aspects that can be considered from the level of artifacts to those of systems and social implications. In the following chapter, we will then turn to issues of conceptualizing things in themselves that are relevant for understanding fluid assemblages, how they are both similar to and different from other kinds of things. The purpose of the overview presented here is twofold. First, we want to illuminate various aspects of things by reading them through activities in which they are implicated—in other words, what they are for humans in particular

Things for Us

45

kinds of use. This also requires engaging with scholarship in different disciplines that investigate these domains of human activity. In doing so, there will always be a sort of tension or torque between the primary concerns of various bodies of scholarship and the ways in which we engage them, since we are interested in what they have to say about things in particular. For some areas, particularly those in the social sciences, this actually goes against some of their primary commitments related to respecting the primacy of lived human experience and praxis. While we are sympathetic to these commitments, and share them in many ways, we also see the need to get to a more neutral ground that does not rely on human experience and existing social practice as analytic frame, since there is much of concern that is not accessible experientially during normal use. Thus, while we recognize the fundamental entanglement of experience, practice, and things (or materiality), we approach this entanglement through the lens of things—what they are and what they do. The second purpose is in a sense the reverse: to point to the multiple complex dynamics in which things are implicated, and to expand the types of issues that are brought into focus when considering their design and configuration. The first purpose can be seen as a spotlight shone on particular aspects of things from perspectives that consider many more moving parts, such that they account for only a small part of what things are and, conversely, things are only a small part of what these perspectives consider. This second purpose can be seen as beginning with things as wholes, and recognizing the broader dynamics and issues in which they are implicated. Rather than reflecting limited applicability, these touchpoints provide openings onto much larger sets of concerns and serve as connections that can be followed to explore those vantage points. These are two sides of the same coin, and reflect the holistic designerly perspective we take here. Rather than reducing or selecting piecemeal from rich areas of scholarship, we mean to expand things (specifically fluid assemblages) as both a lens of its own and—crucially—a site for inquiry and intervention. Moreover, addressing multiple sets of dynamics and their interactions is necessary when they are all present in things. It should also be noted that this investigation in many ways reverses the usual order when it comes to considering things and human activities. Typically, a certain domain or aspect of human activity is under investigation, and things are considered in the sense and to the extent that they are involved

46

Changing Things: The Future of Objects in a Digital World

or implicated in that activity. The activity or matter of concern is in focus, and things are one of the moving parts that must be accounted for. This also means that things are accounted for in terms of the human activity that is in question. The illumination of things that is achieved comes from a spotlight shone from the perspective of one particular set of concerns. However, things clearly can have multiple roles, identities, and functions. Understanding their character and consequences in the world in general, then, and in a more holistic way requires considering the multiple roles and identities they can have. Thus, the notion of a “general” understanding of things that we aim for here is of the Aristotelian rather than Platonic kind: we aim to reach toward the general by considering many different particulars, like understanding what a chair in general is by encountering and sitting in many particular instances of chairs, as there is not one universal instance, expression, account, or articulation present. This survey will inevitably be reductive in some ways, and will not quite do justice to the rich areas of scholarship referenced. Yet it is also necessary to get out of the thick of that richness and zoom out to a higher level in order to see something else that is more to do with general orientations, assumptions, and perspectives. Another key aspect of our investigation is that it is about connections, synergies, and tensions among different conceptual frameworks. So, our often necessarily cursory overviews can be seen as pointers toward paths leading to other areas that are potentially worth exploring in more depth, while our primary mission here is to create a higher-level map of a larger territory and to trace a web of connections among some of its main features.

Tool—Action and perception Perhaps one of the most basic ways of relating intentionally to things is to see them as tools that can be used for some specific purpose. Tool use might even be one of the things that makes us human (Nelson and Stolterman 2012), although other animals have also been observed using tools. From the time a human picked up a bone and turned it into a weapon or other implement, humans have been appropriating things for their own ends and using the materials they can find to build better tools. In other words, they have been

Things for Us

47

designing. In the craft tradition that preceded professional industrial design, there was a close interrelationship of designing, building, and using, and organic connection among materials, places, and contexts of use. This can be seen, for example, in Jones’s (1992) discussion of craft methods, and example of the craft process in the evolution of wagon making. Fast forwarding from this early tool use, the Industrial Revolution called industrial design into being in service of mass production (Dilnot 2014). Whereas things had previously been crafted for particular uses, the challenge became one of designing for use in a more general way in order to support mass production of uniform products. The perspective of things as tools is, then, quite old and quite common, found in history, philosophies, and practical technological development. It also generally makes sense, resonating with many of our own ordinary experiences of relating to things. Tools operate in the domain of the practical. The activities most prominently associated with this perspective of thing as tools are action and perception: what people can do, achieve, and perceive. As Peter-Paul Verbeek, a philosopher working in the postphenomenology subfield of philosophy of technology, states: “Technology mediates our behavior and our perception, and thereby actively shapes subjectivity and objectivity: the ways in which we are present in our world and the world is present to us” (Verbeek 2005, 203). In Verbeek’s development of postphenomenology in particular there is an emphasis on not only how the world comes to us, but also how humans are themselves constituted through relations. Yet at the same time, in postphenomenology and more generally within what might be referred to broadly as the “tool” perspective, there tend to be basic assumptions about human agency. Humans are at the center of the picture, determining what matters in relation to their needs and desires and generally driving the action even as their relations with things affect them as well. It is still humans who choose whether or not to look at the ultrasound image (Verbeek 2008), to use the system, or to purchase the product and “domesticate” it such that it becomes a meaningful part of their lives. Corresponding to this focus on human agency is a parallel focus on human subjectivity. This is to say that it is human experience that matters, and the aspects of things that are considered relevant are those that enter into human experience. This orientation is indeed at the very foundations of phenomenological method, which considers the world as it appears to us.

48

Changing Things: The Future of Objects in a Digital World

Something that exists in the world but does not appear or relate to us does not enter the picture. This might seem logical in some ways, but when it comes to contemporary computational things it gets to be more problematic as there is no inherent or necessary connection between a thing’s function and how it appears during use. This is another issue to which we will return later on. Another basic orientation within this perspective is that technological tools somehow deal with the real. Of course this cannot be the real in any direct or unproblematic sense, especially from a philosophical perspective. Yet in postphenomenology in particular there is a sense, particularly through the connection to scientific instruments and praxis, that what is at stake has to do with how people can come to connect to and make sense of what is real. And how could the world and one’s connection to it feel any more real or substantive than it does when hitting something with a hammer? A hammer evokes a reassuring solidity and simplicity of action and reaction that helps anchor even Heidegger’s lofty ontological discussion (Heidegger [1927] 2010). Tools, at bottom, are imminently practical, handy, goal-oriented things, and this heritage can be seen in even much more sophisticated technological tools. While this might seem completely straightforward and unproblematic, it poses a contrast to some of the basic assumptions and orientations around things as media for communication that will be discussed later. Tools are used for specific purposes based on the particular functions that they have. To design things as tools means to think about intended use and purpose. It also involves considering people who will do the using— the “users”—and contexts of use. Indeed, understanding eventual users and contexts of use is the focus of user-centered design, which seeks to design products (and now also increasingly services) based on human needs, desires, preferences, and ways of working. In this perspective, things have their identity in relation to the goals they help humans to achieve. The field of human–computer interaction (HCI) began with attempts to configure the interfaces of computational systems such that it would be possible for humans (more specifically, human brains, as bodies were not so much in the picture at this stage) to understand their functions in order to interact with them successfully in order to accomplish given goals. Lucy Suchman’s (2007) groundbreaking work showed that actual in situ human use is typically

Things for Us

49

less logically planned than systems developers tended to assume, yet it also maintained a focus on instrumental use of technology. In short, there seems to be a common understanding that technological tools exist to help humans get stuff done. Ensuring a good fit between product and use practices means conceiving in quite fine detail and making assumptions about the scenarios in which it will be used. These intended scenarios of use have been referred to as “scripts.” However, these can be “de-scripted” in the real world of actual use, where people can use things in quite different ways than their designers originally intended (Akrich 1991). However, even as it has been recognized that users frequently do not behave in relation to products in ways that designers intend, the general framing of how people relate to things is still in terms of what they do with them. There are very significant semiotic aspects to be sure, and (especially high-end) products often try to both sell and signify a certain lifestyle more than the products themselves. Yet at the same time the reason this is possible is to a large extent that the product implies certain practices of use, which in turn are part of a certain kind of lifestyle. Technological tools might be commonly thought of as relatively straightforward,  with any notable complexity coming in the form of technological  advances that render them ever more sophisticated. Yet, as the more detailed concepts and examples in the following chapters will hopefully show, there is much more to tools than engineering, making, and use. Tools that help us perceive and act in the world inevitably shape the character of that action and perception. The ways in which we take up with the world are often mediated by technologies, a human–world interface at least as important as the interface that enables interaction with a device. Technologies are often manifestations of larger systems and ecosystems that govern the techniques used in administration in a society at larger-than-individual scales, leading to particular configurations of power and collective forms of life. While things that are tools operate in the domain of the practical, it is important to remember that the practical is about everyday praxis—about the ways in which things are actually done. And questioning regarding the ways in which things are—and perhaps should be—done cuts to the very heart of issues related to that most basic of political questions: what it means to live a good life, together.

50

Changing Things: The Future of Objects in a Digital World

Media—Communication Many of the contemporary digital networked things that now pervade everyday life can be seen as media technologies. That is to say that they are involved in the creation, modification, sharing, and distribution of information, content, and messages that are part of social practices, rituals of communication, and the many data-intensive and information-based processes that run and govern our contemporary societies from local to global levels. This lens of information and communication and perspectives from the disciplinary fields of information, communication, and media studies bring into focus important aspects of contemporary things. Although information and communication technologies can frequently be seen as tools as well, and certainly have a material dimension that can put them also in the category of stuff, this lens brings into focus their operation in the domain of the symbolic. Media are involved in processes of information and communication. Although closely related, there are, at least for our purposes, a couple of main foundational issues associated with each. Information deals with underlying representational practices and units of meaning. It is the lifeblood that flows through digital networked technologies that are used for information and communication. Thus, while issues of representation and information are relevant in consideration of any kind of communication and information processing activity, they are absolutely key when it comes to considering contemporary technologies and their uses. Different technological configurations represent, transmit, store, and process data and information in ways that vary, but share commonalities through underlying elements and infrastructures. We thus need to consider what information is and does in these systems.2 Communication is a social practice, and communication technologies with different capabilities enable new types of communication practices that develop around them and in relation to existing practices. It is here that we run squarely up against the social, and it is typically the perspective of social and cultural practices that is used when considering communication technologies. Thus, this is one of the places where the torque between our perspective and that of the work that we consider might be most keenly felt.

Things for Us

51

First, we need to consider what is involved in (mediated) communication in terms of its most basic components. Perhaps the most obvious place to start a consideration of things in relation to communication is to think of the invention of key media technologies—actual technical apparatuses. Beginning with language as the most fundamental communication technology, relying on the bodily apparatus of vocal chords and so on, we can then think of the inventions of written language and recording media (clay, papyrus, paper), printing press, telegraph, radio, phonograph, film, telephone, broadcast television, and others. More recently the internet and World Wide Web have brought many more possibilities, from email and chat to virtual environments to many forms of social and collaborative media. Now, each of these media technologies involves slightly different technological capabilities and communication protocols. Importantly, each involves some particular form of representational practice by which meaning takes on a symbolic and material form that requires interpretation on the part of the receiver in order to render it meaningful. This is a hermeneutic practice that involves making sense out of both the content and the technical and social protocols and norms involved with a particular medium. Furthermore, different media forms contain other media. Or, as media theorist Marshall McLuhan famously stated: “The ‘content’ of any medium is always another medium. The content of writing is speech, just as the written word is the content of print, and print is the content of the telegraph” (McLuhan 1964, 8). More recently, Bolter and Grusin have argued that the representation of one medium in another—what they call remediation—is a “defining characteristic of new digital media” (Bolter and Grusin 2000, 45). They also identify two associated characteristics: immediacy, or the desire for the medium to disappear altogether to allow for the feeling of direct access to what is mediated; and at the other extreme hypermediacy, which instead draws attention to the medium and its multiplicity. Indeed, especially at the time their book was written during the late 1990s, the fashion did seem to push toward the excesses of possibilities enabled by new media. This phenomenon is perhaps most easily called to mind through the form of a television news program in which one frame commonly contains multiple windows, scrolling tickers, animated graphics, and more.

52

Changing Things: The Future of Objects in a Digital World

Returning to communication as such, it can be said that there is never any such thing as completely direct, unproblematic, or unmediated communication. At a fundamental level, communication can be seen as an attempt at reconciling self and other (Peters 2012), a project that is never perfectly achievable. There will always, even in the “simplest” face-to-face interaction, be glitches, misunderstandings, and breakdowns that must be negotiated. Representation, interpretation, and the impossibility of perfect understanding are basic conditions of communication—even those that do not involve advanced technologies. And of course when technologies are involved there are additional related complexities, and sources of breakdown. There have been different positions regarding the role and significance of media technologies themselves. Outlining first two extremes, the mathematical model of communication developed by Claude Shannon and Warren Weaver in the heyday of cybernetics during the time of the Cold War saw technology as the channel for information transmission (Shannon and Weaver 1959). In their influential model, a signal is encoded by a sender into a message that is sent through a channel and decoded on the other end by a receiver. In this view the channel is relatively neutral as the message, at least ideally, simply passes through it, although it is always threatened by noise that could degrade it. On the other hand, Marshall McLuhan famously asserted that “the medium is the message” (McLuhan 1964); in other words, a medium is not a neutral channel that just conveys a message more or less effectively, but is rather itself a significant (or even the primary) message. Concerns about the role and effectiveness of the medium also dominated early work in the field of computer-mediated communication (CMC), which considered whether the text-heavy medium of the computer was “rich” enough to enable effective human (or, perhaps better stated, humane) communication. On the other hand, it was also the disconnect between information that could be transmitted over the internet and the “richness” of a person’s physical body and context that enabled the free play of identity performance that characterized much early online interaction (Turkle 1997; Hayles 2008). As is typically the case with sweeping claims tinged to greater or lesser degrees with technological determinism, the reality is more complex than any of the previously mentioned perspectives, even as they have each helped to articulate certain issues and dynamics. Media technologies and the

Things for Us

53

capabilities they provide play a significant role, but communicative practices in which they are used are always contingent, culturally and historically embedded affairs. Pointers to various dynamics involved can be found in Lisa Gitelman’s comprehensive definition of media as “socially realized structures of communication, where structures include both technological forms and their associated protocols, and where communication is a cultural practice, a ritualized collocation of different people on the same mental map, sharing or engaged with popular ontologies of representation” (Gitelman 2006, 7). One key aspect of communication as a cultural practice is that it involves performance of self. This is the case also in face-to-face communication, in which the persona that one presents to others is a performance that does not necessarily match what is going on “backstage,” to use Goffman’s (1959) influential dramaturgical metaphor. Different media technologies allow for different types of performances. The early online environments allowed for very carefully crafted performances of self that could be very different from those “in real life” (Turkle 1997). Now in the age of social media there are generally much stronger connections between “real life” and “virtual” interactions, with interactions crossing over between various media, and between online and offline. In fact, an analytic and practical distinction between online and offline has become untenable in a situation of constant crossover between online and offline, “virtual” and “real” interactions that Coleman (Coleman 2011) has termed “X-reality.” Connection between online profiles and offline identity has also come to be legally required through terms of service agreements, such as those of Facebook, that explicitly prohibit using a false identity. Or conversely, a verified connection to an offline identity can be rewarded, as in the case of Twitter’s “verified identity” system and associated badge prominently displayed along with verified accounts. Rather than playing with different identities online, it is now seemingly more common to curate one’s personal brand in a way that maintains some level of consistency across social media channels. And brands in the more traditional corporate sense have very carefully managed social media presence, and run sophisticated search engine and social media marketing campaigns. With this very broad background in place, we can now already point to a few key differences between the concepts of media and tools (discussed in the first section). Although these perspectives may be used in relation to the

54

Changing Things: The Future of Objects in a Digital World

same ostensive objects of study, they come with different kinds of underlying assumptions. While tools are generally considered in terms of more or less practical goals and effects in relation to practices of use, media are involved in culturally situated practices of communication and practices of making and interpreting meanings that can never be singular or conclusive. There are always gaps between sender and receiver, world and representation, reality and model that cannot be perfectly bridged. These gaps thus become key sites for investigation, particularly as more and more of our social realities and access to them are brought into the realm of the computational.

Computer—Data processing Computation can in one sense perhaps be seen as quite different from media, having more affinity with mathematical logic, science, and engineering than culturally situated practices of communication and meaning-making. Yet the ways in which computation exists in the world are very much based not only on these types of practices, involving issues of representation, interpretation, and culture in its many applications, but also on its most fundamental levels involving data, algorithms, and protocols. Computation is a tool for data processing, a medium of information. The roots of computation are contained in the word itself, from a time in which “computers” were people who performed computations. The use of computers to perform complex calculations for scientific purposes is certainly still present, as is their use in increasingly sophisticated military applications (think, for example, of drone warfare or battlefield robots). Another major theme of information management and access was eloquently prefigured by Vannevar Bush, who had in his capacity as director of the Office of Scientific Research and Development in the United States coordinated the effort to apply scientific knowledge to warfare during the Second World War. Writing in an article in The Atlantic magazine at the end of that war in 1945, he suggested that the energies and achievements that had been marshaled so effectively for the cause of war should now be channeled toward other worthy objectives during peacetime. The problem he identified was that of an overwhelming amount of information that was becoming increasingly difficult for scientists

Things for Us

55

to stay on top of, particularly as scientific specialization increased. And there was a need not only to store information, but also, importantly, to make it accessible in a way that supports how people think, namely, by association. What he envisioned as one possible solution seems remarkably prescient: Consider a future device for individual use, which is a sort of mechanized private file and library. It needs a name, and, to coin one at random, “memex” will do. A memex is a device in which an individual stores all his books, records, and communications, and which is mechanized so that it may be consulted with exceeding speed and flexibility. It is an enlarged intimate supplement to his memory. (Bush 1945)

Since storing files of all kinds on computers, external storage devices, and networked storage services is now a normal part of life, as is googling for all imaginable types of information, it is perhaps difficult for many of us to imagine a time in which these types of structures were not in place (or were much more rudimentary). Yet it is possible to see how computational systems are still being developed in order to serve this purpose, from automated organization techniques to ever “smarter” access to contextually relevant information. One particular arena in which information became increasingly important was in military intelligence. During the Cold War, information came to be seen as an end in itself (Ceruzzi 2003), foreshadowing more contemporary post-2001 programs of massive surveillance that have operated on the general principle of collecting as much data as possible. It is thus always worth being mindful of this when considering computational things that generate information that could be seen as “intelligence” and/or used for control in some context. Another trajectory has its roots in the American counterculture of the 1960s and 1970s that came to expression in the burgeoning cyberculture of the 1990s as a group of hippie entrepreneurs (including the founders of the Apple computer company and Wired magazine) reimagined computers as tools that could be used to serve the utopian ideals of personal liberation, expression, and community building rather than warfare (Turner 2010). The first commercial sites on the internet had some of the communal spirit of the early online communities, seen in features such as reviews (and reviews of reviews) that allowed for policing of sites (such as Amazon and eBay) that

56

Changing Things: The Future of Objects in a Digital World

would not have been possible in a top-down way; and this community element in fact became essential for such sites to survive (Ceruzzi 2003). Now, it seems that every computational device and application is in some way “social.” In this sense, the invitation or even imperative to share with others can be seen as a heritage of both the communal spirit of early cyberculture and the drive of both governments and businesses to be able to find out what people are up to with ever-increasing speed and precision. It is also worth considering some of the basic aspects of computational architecture, in which we can see that computers have always been assemblages in a fundamental sense. Early computer models consisted of a central mainframe computer with “dumb” terminals that were used to access it and perform computations using a time-sharing model. The processor and display were clearly separated. In fact, computer historian Paul Ceruzzi states that a “systems approach” to studying the history of computing is not particularly helpful because a modern computer “is a system: an arrangement of hardware and software in hierarchical layers” (Ceruzzi 2003, 4). And again: The word “software” suggests that there is a single entity, separate from the computer’s hardware, that works with the hardware to solve a problem. In fact, there is no such single entity. A computer system is like an onion, with many distinct layers of software over a hardware core. Even at the center— the level of the central processor—there is no clear distinction: computer chips carrying “microcode” direct other chips to perform the processor’s most basic operations. Engineers call these codes “firmware,” a term that suggests the blurred distinction. (Ceruzzi 2003, 80)

Another way in which the assemblage character of computers can be seen is in the very concept of an “assembler,” an early program that assembled another program into machine instructions that the computer could execute. It was also actually a milestone for computation when instructions and data were stored in the same place in stored programs (Ceruzzi 2003). This was also the milestone that turned computers into things with agency of their own, not just sitting waiting for instructions from elsewhere but potentially running programs that allowed them to respond to external events. The programmability of computers is what allows the flexibility for them to be turned into such a vast and diverse array of devices, and it is the characteristic that effectively defines modern computing (Ensmenger 2012). Yet even with

Things for Us

57

this flexibility and assemblage character, computational systems have also always needed to appear solid and unified during use. One basic example of this was file storage systems developed in early personal computers, where files were stored in fragments on the disk but represented as single files for the user; the operating system was responsible for storing the data and retrieving and reassembling the files when needed (Ceruzzi 2003). This was even to some extent visible in consumer defragmentation software, which provided a visual representation of system memory as it was optimized by the program. From a design perspective, working with what might be called digital materials implies a quite different set of actions and implications than working with and giving form to physical materials. Using digital materials does not use them up, as with physical ones, but rather establishes relations and conditions under which they are activated. These are relations that persist and need to be maintained throughout the life of a digital product. Starting with electricity, this extends to the connections between hardware and software components, pieces of code (particularly apparent in object-oriented programming languages), and network connections to other resources and devices through the internet and World Wide Web or other more near-range protocols such as Bluetooth or RFID. These relations can also extend to those between devices in what can be referred to as ubiquitous computing, leading to complexity and emergent effects of a different order (Coyne 2010; Ekman et al. 2015).

Roles, relations, agencies At this point we have covered quite a bit of ground in considering things as tools, media, and computers. Although this has already been at a quite high level, especially in relation to the depth at which all of the issues and dynamics we have referenced can be explored, we can already at this point extend the previous analysis by recognizing the possibilities for variation in the distribution of roles, relations, and agencies that are implied in relation to tools, media, and computers. To begin, it is possible to note a simple yet significant distinction between what might be called active and passive roles. When a thing is used as a tool to mediate action or perception, a person may be on the receiving end of that

58

Changing Things: The Future of Objects in a Digital World

action or perception just as well as at the active end. This can be referred to simply as a distinction between subject and object (with no intention whatsoever of returning to the tired distinction between these in an absolute sense!). A telescope or binoculars can be trained on a distant human target rather than the heavens; medical imaging equipment can be used to reveal the inside of a person’s body; a read receipt can reveal that someone else opened a message; and text created with technology (as in fact all are) can be interpreted in terms of what it says on behalf of and/or about its author. This mediational structure in which human activity is in focus as what is revealed in fact happens all the time whenever someone accesses a web page or uses really any kind of fluid assemblage. It is possible to make a similar distinction between sender and receiver of messages and content, or between producer and consumer, author or reader of media content. The leveling of opportunities to be in both of these roles is in fact one of the most hyped, discussed, and debated characteristics of media that have been variously described as new (Bolter and Grusin 2000; Manovich 2001), connective (van Dijck 2013), convergent (Jenkins 2006; Meikle and Young 2012), collaborative (Löwgren and Reimer 2013), social, and so on. Whereas a few decades ago the media landscape was dominated by a few big industry players that controlled the means of production and distribution, and were thus the big producers for mass consumption, now anyone with a smartphone and internet connection can be a producer and distributor. So far so good, as there is already a quite high level of awareness of the various roles people can take in relation to media, even if the social implications of this state of affairs are not as obvious or straightforward all the time. However, this situation is very different when it comes to producing and consuming data, or rather, it is perhaps more similar to the dynamics of media a few decades ago. That is to say that while all of us end up producing data through our ordinary activities, the consolidation, processing, and use of all of the data generated happens more at a government and industry level rather than at an individual one. This is not to say that such large-scale data is not made available to individuals, since it often is in various ways. Even something as simple as hashtags that are trending on Twitter is one such example (even as the exact mechanisms behind this are opaque). Yet there remains a significant disparity between the capabilities of large-scale corporate or governmental

Things for Us

59

entities who control the means of producing data versus those about whom the data is produced and those who wish to study it (such as academics). Now we have looked at some lenses on things in relation to the roles they play in human experience and affairs. While each of them highlights significant aspects, it is also possible to see that none of them can fully account for what things are. Since we are trying to get to the bottom of what (digital, networked, computational) things are, the next logical question is something like: Well alright then, what are these things really, if we try to get away from these more focused lenses and instead look at what is actually there, in some kind of more original sense? This question takes us into the realm of ontology.

Notes 1

An earlier version of part of this argument appeared in Wiltse (2017).

2

It should perhaps be emphasized that we will focus here only on data/ information as it exists practically in these systems, and not on the more foundational and also important aspects of these concepts in themselves. For more on the philosophy of information, see, for example, the work of philosopher Luciano Floridi (Floridi 2011).

6

Things in Themselves Turning now to more basic ontological issues, we are not concerned here with ontology and the nature of being in general, but rather the nature of being of things that are fluid assemblages. However, as we loosen our grip on existing lenses in order to try to see what is there in a more fundamental sense, there are a few considerations of more basic ontologies that will be useful. Because we are, after all, still inquiring into what is, even if only a particular subset. Two of the most basic things that need to be accounted for when it comes to the being of things are identity and change. In other words, how can something be delineated as a this thing rather than as part of an undifferentiated whole? What things are crucial for making that differentiation, and not just “accidental”? And on the other hand, how is it possible to account for change, the fact that things are not pure, stable, essential forms but rather changing their forms and relations over time? Also, when addressing fluid assemblages we explicitly need to account for these: on the one hand we need to account for their solidity, the fact that they are actual entities that exist and do things in the world; on the other hand, we need to account for the fact that they are in flux, and that any solidity is only temporary. Things can also be investigated at different scales, bringing into focus dynamics from the very local and concrete up to the level of systems. Just as we suggested in the earlier section that a basic understanding and awareness of differences between general orientations to considering the role of things in human activities can be useful, especially when working with multiple perspectives, we here suggest that sensitivity to scale is important for similar reasons. Issues of structure at different scales can be seen in the interplay of two key orientations that are foundational for a design approach: concern with

62

Changing Things: The Future of Objects in a Digital World

particulars and systemics (Nelson and Stolterman 2012), and the ways in which they relate to each other. Much of design deals with the details of particulars, both in the concrete formal and aesthetic expressions of things and in their integration in specific contexts and practices of use. Some types of design also deal explicitly with systems, and there are systems-oriented perspectives that will be discussed in each section. However, it is important to recognize from the outset the inherent interconnectedness and complexity of the real world. Thus, our purpose in identifying so many different perspectives and issues here is not in order to analytically break down our object of study into many different component parts; rather, it is to provide fine-grained conceptual lenses that can be used to develop sensitivities and analytic capacities that can be combined in order to gain a better and more holistic understanding. This holistic and systemic orientation is indeed fundamental for design, since every design “is either an element of a system or a system itself and is part of ensuing causal entanglements” (Nelson and Stolterman 2012, 57). In all designed things—and especially in fluid assemblages—there is a constant interplay between the particular and the systemic, and a temporary focus on one should not cause us to lose sight of the other. In fact, we argue that fluid assemblages require a new level of agility when it comes to moving between the two. A consideration of properties and interactions addresses what it means to be that kind of a thing at a basic level, and in which types of interactions they are generally involved. But the significance of things is in many cases not due to their existence in one instance, but rather in many—the fact that they are ubiquitous enough to be implicated in larger-scale societal patterns and dynamics. Indeed, things—whether tools, media, or computers—collectively constitute a significant part of contemporary environments. Consideration of different kinds of things in terms of systemics and scale thus addresses a more or less quantitative question—what is it like to have not just one but many, sometimes vastly many, of this kind of thing in the world? What are the dynamics and consequences of things being interconnected, and of scaling up and out? The qualitative question that is the counterpart of systemics and scale is: What are the social implications and forms of life that are supported or precluded, encouraged or foreclosed, in a world where these things hold a significant sway?

Things in Themselves

63

In the following sections we will consider three basic orientations to considering what things are: as stuff, assemblages, and objects or machines. Each of these perspectives can be seen in multiple and vast bodies of work, which can of course not be completely addressed here. The aim of this section is also the opposite of being encyclopedic: rather than accounting for the full depth and breadth of existing work, we want to get to the bottom of the basic elements and contours. The references we give are indicative rather than comprehensive, but they can also serve as entrance points for further investigation.

Stuff Viewing things as “stuff ” is perhaps the most basic way of relating to some other entity in the world with a physical presence in our lives. The very word itself is in fact telling, as “stuff ” is such a general reference that it does not really say much at all about the character of what is referred to. As such it can say something about the way that people relate to things though. Think, for example, of phrases such as “I have so much stuff ” or “I need to get my stuff ”; the reference is not individual but aggregate, a particular selection of the total accumulation of the various life accessories, basic necessities, prize possessions, curios, and detritus that are our companions at various (and variously lengthy) stages of life.1 The scale of this situation can be encountered with some force when moving and needing to sort through one’s things that must be either packed and moved or else somehow disposed of. This can be a time for making quite particular kinds of calculations around the value that particular things bring to one’s life in relation to the cost of keeping them, involving perhaps considerations of utility, history, emotional attachment, expected future value, and more. Stuff is a collection of things that have physical presence and are there for us in the ordinary dealings of our lives, variously supportive and frustrating, meaningful and mundane, broken down and repaired, treasured and discarded. They have their own temporal trajectories and narratives that are anchored in a teleological terminus of their eventually becoming waste (Viney 2014). Our stuff can also say a lot about who we are, individually and collectively, how we order our lives, what we value, what privileges we have or do not have,

64

Changing Things: The Future of Objects in a Digital World

and so on (Menzel and Mann 1994). The field of material culture takes stuff as its object of study, showing that stuff not only reflects but also helps create who we are and how we live our lives (Shove 2007; Miller 2013). The field of design anthropology considers similar issues but in a way that is more active, interventionist, and future-oriented (Gunn 2013). The practice turn in social theory also places stuff in a central role, considering the ways in which things become meaningful over time in everyday life (Cetina, Schatzki, and Savigny 2005). Diving even deeper, Martin Heidegger famously saw relations with things as fundamentally constitutive of our dwelling in the world (Heidegger 1993). At this everyday level of relations with stuff, it is not strictly rational functionality that determines the development and uses of objects. Jean Baudrillard demonstrated this at length in his book The System of Objects, arguing that “it is the whole system of needs, socialised or unconscious, cultural or practical—in short, a whole inessential system, directly experienced—which surges back on to the essential technical order and threatens the objective status of the object itself ” (Baudrillard 2005, 7). Rather than being essential to it, form becomes a connotation for function, as in the tail fins of midtwentieth-century American cars that signified speed but were functionally counterproductive (Baudrillard 2005). Baudrillard’s remarks summarizing this situation seem now remarkably prescient and relevant to fluid assemblages (Baudrillard 2005, 67): Every object claims to be functional, just as every regime claims to be democratic. The term evokes all the virtues of modernity, yet it is perfectly ambiguous. With its reference to “function” it suggests that the object fulfils itself in the precision of its relationship to the real world and to human needs. But as our analysis has shown, “functional” in no way qualifies what is adapted to a goal, merely what is adapted to an order or system: functionality is the ability to become integrated into an overall scheme. An object’s functionality is the very thing that enables it to transcend its main “function” in the direction of a secondary one, to play a part, to become a combining element, an adjustable item, within a universal system of signs.

Although Baudrillard (2005) framed consumption in terms of sign values operating within discourses, it is worth returning briefly to the more literal aspects of consumption and its roots in commodity capitalism. As Karl Marx

Things in Themselves

65

famously argued, one of the defining characteristics of commodities is that they conceal their conditions of production. A commodity does not reveal the labor, and laborers, involved in its production, but rather appears as a more or less free floating good available for exchange. This situation still maintains as a default today, which is why it is possible to feel shock when learning through news reports of dreadful conditions for workers who make our smartphones or clothing. It can also be seen in efforts to be more transparent about conditions of production such that consumers can make more ethically sound purchasing decisions, as in certification and labeling schemes for fair trade, sustainability, not using animal testing, and so on. In the more high-tech realm, the efforts of Fairphone to produce “the world’s first ethical, modular smartphone”2 are illuminating. They state that their phones “hold a complex story of the hundreds of people who helped make it,” and that they “want to open up that story, so we can make a positive impact in how phones are made, used and recycled.”3 Their efforts that they report do indeed reveal the complexities and challenges of ensuring that something is ethically produced and with longevity in mind, in everything from sourcing raw materials to production to repair and recycling. A large part of the motivation for initiatives such as these is the general recognition of accelerating climate collapse and the need to reorient present systems of production and consumption toward ones that are compatible with ongoing sustainment for our species and other life forms on earth. As Timothy Morton forcefully describes the current situation: The end of the world has already occurred. We can be uncannily precise about the date on which the world ended. Convenience is not readily associated with historiography, nor indeed with geological time. But in this case, it is uncannily clear. It was April 1784, when James Watt patented the steam engine, an act that commenced the depositing of carbon in Earth’s crust– namely, the inception of humanity as a geophysical force on a planetary scale. Since for something to happen it often needs to happen twice, the world also ended in 1945, in Trinity, New Mexico, where the Manhattan Project tested the Gadget, the first of the atom bombs, and later that year when two nuclear bombs were dropped on Hiroshima and Nagasaki. (Morton 2013, 7)

While these kinds of pressing environmental concerns and corresponding human responsibilities in the Anthropocene have been significant factors, they

66

Changing Things: The Future of Objects in a Digital World

have not been the only ones in a “turn to things” and matter that have spanned (at least) the social sciences and humanities. This (re)turn has in various ways felt around for the more solid moorings of theory, ideas, and signifiers, even if things seem to always somehow arrive after “as the alternative to ideas, the limit to theory, victims of the word” (Brown 2001, 16). There are many very rich strands of thought that can be seen as falling under the heading of a return to things and new materialisms (Dolphijn and van der Tuin 2012), which we cannot cover here; but it is worth highlighting at least a couple key points. One is a recognition of what can be the active, forceful agency of things— what political theorist Jane Bennett calls Vibrant Matter in her book of the same name, in which she uses examples such as debris, a power grid, food, and stem cells (Bennett 2010). Rather than seeing matter as simply inert and lying around until humans do something with it, this perspective calls attention to the internal dynamism and latent capacities that things possess, and their abilities to affect (and be affected by) other things. Another key characteristic of new materialist perspectives is a focus on process and change over stability and identity. A feminist new materialism sees this possibility for dynamism, complexity, and movement as a welcome and indeed essential liberation from universalist and masculine ideals (Braidotti 2003). Moreover, this ongoing becoming is characterized by a fundamental entanglement among entities and agencies that Karen Barad articulates most forcefully in her philosophy-physics, a perspective she terms agential realism. She uses the neologism “intra-action” to refer to the “mutual constitution of entangled agencies” (Barad 2007, 33). This is a fundamentally different conception of agency than has been the norm, seeing agency not as residing in things themselves as internal properties but rather constituted through their intra-actions. Here the basic ontological unit is not things but rather phenomena, in which intra-acting agencies participate in differential mattering (Barad 2007). According to Barad, things do not pre-exist, but are rather “agentially enacted and become determinately bounded and propertied within phenomena” (Barad 2007, 150). Now what she is referring to here are not any particular class of things, but all things at a fundamental level. However, there is a remarkable parallel between this description and fluid assemblages, where these dynamics can be seen more clearly than in the case of other kinds of things. Fluid assemblages do not pre-exist, but are rather

Things in Themselves

67

enacted and become determinately formed and propertied as things available for use within particular contexts.

Assemblages Unlike notions such as “stuff ” or “things,” most of us do not use the notion “assemblage” to relate to entities we encounter. Thus, it is not very straightforward to account for how relating to things as assemblages turns out in everyday life. The concept does, however, offer significant potential when it comes to analysis. The notion of assemblage was introduced in Chapter 3, thus here we will only mention a few additional aspects as we position the different ontological orientations in relation to each other. The key reason we make use of assemblage is that it allows us to be rather precise with regards to how a certain whole comes into being—even as we engage with a rather naïve understanding of this idea.4 As we try to discuss in this book, the perhaps most critical aspect of these networked computational things is how they draw matter and matters together in often very complex ways. And to better see how this is done, Manuel DeLanda’s key characteristics of assemblages are quite useful: 1. Assemblages have a fully contingent historical identity, and each of them is therefore an individual entity: an individual person, an individual community, an individual organization, an individual city (DeLanda 2016, 19). 2. Assemblages are always composed of heterogeneous components (DeLanda 2016, 20). 3. Assemblages can become component parts of larger assemblages (DeLanda 2016, 20). 4. Assemblages emerge from the interactions between their parts, but once an assemblage is in place it immediately starts acting as a source of limitations and opportunities for its components (downward causality) (DeLanda 2016, 21). With respect to such characteristics of assemblages and assembling, not only do our networked computational things combine a range of technologies,

68

Changing Things: The Future of Objects in a Digital World

computation, communication, sensors, interfaces, etc. to become what they are, they also make use of resources in a variety of places, both geographically and thematically, and on top of all these they very much operate on basis of the social fabric of everyday life. But whereas our traditional things are mostly totalities— stable wholes that cannot be taken apart as they are made of parts that have given up their individuality—these computational things are to significant extent built (in runtime!) using components that retain theirs: that can be recombined, reused, and thanks to digital technology even be made active in many places at the same time (such as when a certain piece of software is executed in multiple instances). There is a relevant relation to the notion of form to be made here, especially if we think of design as a matter of giving form to things. In Aristotle’s philosophy, form is the way matter builds things. For instance, it is the “form” that we see when looking a mountain—for how could we have the material of a mountain inside our head? (His theory of perception might be outdated, but it is still fascinating how the concepts of form and matter allow him to explain what is going on, such as how perception then is a matter of in-formation). Furthermore, form is not only about what is already real, or actual, but also about what may become. For instance, within the form of the acorn there is also the potential to become an oak. The idea of form as a matter of both what has already been realized and what has the potential to become is of special importance to the things considered here—just think about how to understand the form of computation, the code as such and the specific “machine” it instantiates when it is executed. With respect to such issues, Gilles Deleuze makes an important distinction: Now, the process of realization is subject to two essential rules, one of resemblance and another of limitation. For the real is supposed to be in the image of the possible that it realizes. (It simply has existence or reality added to it, which is translated by saying that, from the point of view of the concept, there is no difference between the possible and the real.) … The virtual, on the other hand, does not have to be realized, but rather actualized; and the rules for actualization are not those of resemblance and limitation, but those of difference or divergence and of creation. (Deleuze 1988, 97)

While this is part of building an ontology based on difference in itself (metaphorically somewhat like when Carl Friedrich Gauss developed a

Things in Themselves

69

differential geometry that in contrast to earlier algebraic solutions to how to describe curved surfaces does not depend on a reference to an external space), it actually applies also to the things considered in this book: that these things are not characterized by the distinction real versus possible, but importantly by the distinction actual versus virtual—and where the virtual is very much part of what makes these things “real.” This is important as we are here talking about things that have been designed: these are not just objects given in one sense or another, they are made. They have been made real by someone, and what Deleuze’s distinction tells us is that not only was something “actual” made real, so was something at this point “virtual.” We might even in the light of changing relations between design and use, and how these things literally come together only as we actually use them, push ourselves to define design in the context of fluid assemblages to be largely about the domain of the “virtual” along with the lines of differentiation, divergence, and creation that will turn virtual into actual. That is, designing as such deals not only with the virtual and how to make something actual—so does in this case the design output: the “things” made are not only “real” in terms of what is now after design made actual—but importantly also in terms of what is now made real in the virtual sense. Even if this would be only partially correct, it becomes very clear how insufficient it is to understand this kind of design in the same way as we worked with industrially manufactured objects: “totalities” that are “finished” as they are made, and with a prototype serving as mold for mass multiplication, the key challenge being to turn what is possible into reality. Or, to put it differently: the basic responsibility of designers lies not with what is made actual through design, but with what is made real. And in the case of these assemblages, what is made real is very much also a matter of the virtual: that which has not yet become actual, but could.

Objects/machines It is at this point following the other two sections that it in some ways becomes particularly difficult to make clear-cut distinctions among these different ways of conceptualizing things, since there is much in object-oriented perspectives that includes ideas about materialism and assemblages. However, there are

70

Changing Things: The Future of Objects in a Digital World

also a number of other key insights that are particular to the object-oriented philosophical movement that has emerged within the last decade or so. The overall umbrella of this movement has been called the speculative turn, developing as a counter to the linguistic turn that had made it impossible to talk about anything other than language that can never escape its own webs of significations to reach any reality outside of itself (Bryant, Srnicek, and Harman 2011). The movement was ignited by the philosopher Quentin Meillassoux’s diagnosis of correlationism as the condition that has haunted all of philosophy since Kant: a concern with correlation, “the idea according to which we only ever have access to the correlation between thinking and being, and never to either term considered apart from the other” (Meillassoux 2009, 5). Thinking about the world had turned into thinking about only the relation between human and world, and speaking of anything as existing “in-itself ” had become the height of philosophical naiveté. But in accepting these foundational terms for inquiry, Meillassoux suggests: Contemporary philosophers have lost the great outdoors, the absolute outside of pre-critical thinkers: that outside which was not relative to us, and which was given as indifferent to its own givenness to be what it is, existing in itself regardless of whether we are thinking of it or not; that outside which thought could explore with the legitimate feeling of being on foreign territory—of being entirely elsewhere. (Meillassoux 2009, 7)

While this rediscovery of reality might seem rather amusing to nonphilosophers, it is worth considering just how deep these correlationist assumptions run in shaping also other forms of inquiry. One example that is quite close to our concerns here is the focus on user experience in design. Certainly things themselves have been carefully shaped and examined in design, but this has been done with a view toward the meanings and experiences that they are assumed to enable. The underlying questions concern not so much what things are and do in themselves, but what they are and do for humans who use them. The linguistic turn has also been felt in design as a turn toward product semantics, what objects signify. The split between thing-for-a-human and thing-in-itself only deepens when it comes to computational things. But things in themselves also have ways of reasserting themselves. Now we realize, for example, that plastic consumer goods and packaging continue to

Things in Themselves

71

have a life of their own after humans have disposed of them, in landfills and oceans where they affect other life-forms and ecologies in ways that have nothing to do with user experience as it is considered in design. The need to account for multiple roles of things and nonhuman agencies is made more pressing by our ecological crisis. Morton articulates this need in terms of hyperobjects, a term he coined to refer to “things that are massively distributed in time and space relative to humans” (Morton 2013, 1)—the agencies and ecologies of the Anthropocene and a history that is no longer exclusively human. In the digital realm, we have also learned, for example, that social media are not only means for sharing what we are up to, but also, among other things, organization and recruitment tools for terrorists, the nexus of economic ecosystems, large-scale influence and dis-information machines, and tools that foreign governments can use to hack elections. The things we have created are in some ways taking on lives of their own. They may not be exactly like Frankenstein’s monster, but there is still something uncanny in the recognition that they are not entirely under our control or even open for our understanding. Returning to philosophy, one of the key characteristics of object-oriented philosophy is of course its focus on objects as the building blocks of reality. This is a conception of objects in a much more expansive and inclusive sense than might be our ordinary habit of mind. It tends to work with a flat ontology that sees all kinds of entities as equally existing, and does not discriminate on the basis of scale, whether something is considered real or imaginary, corporeal or incorporeal. Graham Harman (2011) explains the reasons for this in his diagnosis of the traditional “undermining” and “overmining” of objects. Undermining involves always breaking objects down into smaller constituent pieces that are seen as more fundamental in some way, whereas overmining sees objects only in terms of their perception or the effects that they have. Undermining cannot account for the emergence of objects as independent entities, whereas overmining does not recognize the reality of objects that exceeds the effects that they have in the world (Harman 2016). But Harman reserves his strongest criticism for the combination of the two that he calls “duomining,” a position he associates with materialism as “the hereditary enemy of any object-oriented philosophy” in that it both reduces down to ultimate elements while also treating them as bundles of qualities (Harman 2011, 13). In opposition to materialism, he proposes the concept of

72

Changing Things: The Future of Objects in a Digital World

immaterialism. While materialism emphasizes fluidity and change, action and intra-action, practice, contingency, multiplicity, and immanence, the principles of immaterialism focus on stability, substances and essences, interactions, singularity, and withdrawn reality over pure immanence (Harman 2016). The case study he uses to illustrate this object-oriented approach to social theory is of the Dutch East India Company and the various stages of its existence over time. Now, the purpose here is not to follow Harman in waging assault on materialism, or to provide an account of the retorts from the other side, for example, Braidotti and Vermeulen (2014). But we can at least acknowledge the truth in the somewhat milder formulation of Levi Bryant: Materialism has come to mean simply that something is historical, socially constructed, involves cultural practices, and is contingent. It has nothing to do with processes that take place in the heart of stars, suffering from cancer, or transforming fossil fuels into greenhouse gases. We wonder where the materialism in materialism is. (Bryant 2014, 2)

In fact, we see the presence and tension between dynamics associated with both materialism and immaterialism in the case of fluid assemblages: the fact that they are in flux, but also stabilize as particular things with particular capabilities and agencies. The concept of object is helpful here, referring to “any entity that cannot be paraphrased in terms of either its components or its effects” (Harman 2016, 3). This helps in recognizing entities that exist as such across different scales. Another term that has been used and that is also quite relevant for our project is machine. The distinction between tools and machines that Lewis Mumford made early in the twentieth century is still remarkably relevant: The essential distinction between a machine and a tool lies in the degree of independence in the operation from the skill and motive power of the operator: the tool lends itself to manipulation, the machine to automatic action. (Mumford 2010, 10)

This “automatic action” is indeed a highly significant difference, even as we might now more likely consider the implications of machine learning and artificial intelligence as animating agencies of machines rather than steam or electricity. Another more recent variation on this concept that is also attuned

Things in Themselves

73

to computational things is Ian Bogost’s unit operations, a term he uses to refer to “the logics by which objects perceive and engage their worlds” (Bogost 2012, 29). Bryant develops his ontology using the concept of machine to refer to “any entity, material or immaterial, corporeal or incorporeal, that exists”; and one of his reasons for using this term is the fact that it evokes this sense in which entities operate and function (Bryant 2014, 15). One of the things that objects/machines do is relate to each other. This is not to say that they establish relationships with each other, but rather that they can affect each other in various ways, according to their character. One of the main insights that Harman (2002, 2011) develops is that there is always an excess of reality to objects that is not exhausted by interactions with them, that is withdrawn in a fundamental sense. Objects are able to access only certain aspects of other objects that are accessible to them, based on their particular character. Harman illustrates this point using the example of fire burning cotton, in which the fire interacts with only the flammability of the cotton and not its other properties that may be accessible to other entities (such as humans who can appreciate its fuzzy softness). One of the ways that machines can relate to each other is through a structural coupling, a configuration that Bryant (2014) describes as one machine becoming a medium for another. It is in this way that machines come to have a purpose or use. As he states: “A machine functions as a medium for another machine not only when it amplifies or extends a sense-organ, but also whenever it modifies the activity or becoming of any other machine” (Bryant 2014, 33). This could be machines in the more industrial sense we might immediately associate with the term, but also includes what could otherwise be described as sociotechnical or even purely social assemblages. They are the entities that come to exist and to have effects in the world. Bryant suggests a practice of onto-cartography as a way of mapping the machines in a given world in order to better understand and—especially—to change it. As he describes it: In its initial formulation, onto-cartography is the investigation of structural couplings between machines and how they modify the becomings, activities, movements, and ways in which the coupled machines relate to the world about them. It is a mapping (cartography) of these couplings between machines (onta) and their vectors of becoming, movement and activity. (Bryant 2014, 35)

74

Changing Things: The Future of Objects in a Digital World

A key method for this mapping activity among entities that are fundamentally foreign to us is alien phenomenology. In his book on the subject, Bogost describes the kind of speculative practice that is needed: Speculative realism really does require speculation: benighted meandering in an exotic world of utterly incomprehensible objects. As philosophers, our job is to amplify the black noise of objects to make the resonant frequencies of the stuffs inside them hum in credibly satisfying ways. Our job is to write the speculative fictions of their processes, of their unit operations. Our job is to get our hands dirty with grease, juice, gunpowder, and gypsum. Our job is to go where everyone has gone before, but where few have bothered to linger. (Bogost 2012, 34)

One of the objects to which Bogost has devoted the most attention is the computer, an object in its own right that can clearly be seen as having its own withdrawn reality and agencies. He also argues for a practice of “philosophical carpentry,” building things as “philosophical lab equipment” (Bogost 2012, 100). This resonates remarkably well with practices of constructive design research, which in turn would need to work with an orientation of alien phenomenology in order to adequately investigate the character and possibilities of fluid assemblages. There is in fact some work already along these lines, as in work on “thing ethnography” in which things serve as participants in inquiry in capturing the world around them (Giaccardi et al. 2016); and in the “Morse Things” project that investigates what it could be like to live together with things that are networked and communicate with each other (Wakkary et al. 2017). This kind of material speculation is indeed becoming increasingly important as we are now living with entities that interact with each other not only on the basis of their “natural” physical properties (as in the case of fire and cotton), but that have designed properties that allow them to interact with each other. Designing these kinds of things means, then, not only designing things, but designing ecosystems and the connections among entities that are their mechanisms of evolution.

Notes 1

A beautiful art/photography project illustrating this is Material World: A Global Family, portrait by Peter Menzel and Charles Mann (1994). It shows portraits of families outside their houses with all their possessions laid out and listed.

Things in Themselves 2

75

https://www.fairphone.com/en/, accessed October 23, 2017.

3

https://www.fairphone.com/en/our-goals/, accessed October 23, 2017.

4

Indeed, as Manuel DeLanda points out, not only does the English translation of the French term “agencement” imply that we lose aspects of its original meaning, pinning down the concept is made further complicated by Deleuze and Guattari offering several different definitions so that “each definition connects the concept to a separate aspect of their philosophy, using the terms that are relevant for that aspect, so when taken in isolation the different definitions do not seem to yield a coherent notion” (DeLanda 2016, 1).

7

A Conceptual Toolkit While the kind of fluid assemblages addressed in this book are in many ways relatively new, they also connect to existing dynamics and trajectories and there is already much to work with in terms of conceptual tools that can help us better articulate and make sense of what is going on. The aim of this chapter is to identify some basic components of such a conceptual toolkit, and the kinds of analytic uses to which they can be put. The perhaps most obvious approach to accounting for existing scholarship regarding a particular set of concerns would be in terms of disciplinary trajectories, and the related historical developments in their objects of study. However, making disciplinary distinctions can be a fraught enterprise. Rather than being driven by a desire to add another installment to one particular disciplinary narrative, we want to work in a more transdisciplinary manner that is motivated by what we have in front of us and need to account for. For this reason we are much more interested in the kinds of analytic distinctions that can be useful in making sense of various aspects of things. What we are after here is conceptual precision, nuance, and utility in analytic tools that can help us see and account for key elements of things in a multifaceted, holistic way. This means that in our investigation of existing scholarship we are on the lookout for aspects of things that they can bring into focus. There is a fine line between aiming for transdisciplinarity and simply being undisciplined. In some ways it may seem that we stray toward the latter as we interpret and misinterpret, use and misuse the conceptual tools we now borrow from contexts to which we do not exactly belong. But what we aim to do is to work in a way resembling the structure of the objects we study: “things” coming together by assembling technologies and transcending concepts that

78

Changing Things: The Future of Objects in a Digital World

were initially developed for different places and purposes. This said, we also intentionally make quite extensive use of quotations, inviting other voices to join our assemblage in their own words in order to keep crucial aspects of their original context. We believe that just as these “things that change” need to be approached as fluid assemblages, so does our account of them. The perhaps not so surprising but still equally important image that this yields is one not that different from looking at something in a shattered mirror: some things are clearly seen in some places, but overall the image is fragmentary and it is truly difficult to trace continuous contours. The insight dawns upon us that this might not only be an effect of the limited reach and concern of these perspectives, but perhaps more fundamentally because the phenomena at hand will not be contained within one account, that there might not be a single complete description, no unified theory. Any notion of one perspective being more “foundational” than another is thus quite misleading, as there is no bottom to be reached, no firm ground to be found. Rather, what we face here in terms of theory is structurally not that different from the things we study: layers and layers of connections between concepts, descriptions, and perspectives that, while each making sense on their own, still require us to be mobile in order to understand how a “bigger” picture is emerging. Indeed, while there are dimensions also in a rhizomatic structure like this, there are no given directions. It is not particularly important to determine if one term is more foundational than another, but it is crucial to understand the relation between a perspective and what can be seen from that point of view. In this we in some ways still heed the mantra echoing in philosophy since Kant’s own Copernican Revolution: to understand our world, we need to turn to the things themselves—however problematic and even contradictory that at times may seem. Again, we therefore ask: what is this thing we’re making and using?

Experiential As a starting point, we begin at the place where we first encounter things in ordinary life: as they enter into experience in personally meaningful ways.

A Conceptual Toolkit

79

Equipment Arguably, the most famous and influential philosophical analysis of tools was put forward by Martin Heidegger. Before considering it directly though, we need to look at its context. The project in his major work Being and Time (Heidegger [1927] 2010) was to formulate the question of the meaning of being. The primary element of this investigation concerned the kind of being who asks this question about the nature of being as such— which is to say, human beings. This kind of being that is fundamentally concerned about its own being he terms Dasein. Following in the tradition of phenomenology and his teacher, Edmund Husserl, the structure of being of Dasein for Heidegger is always being-in-the-world. It is this structure that he sets out to investigate. The phenomenological method he uses takes phenomena as its unit of analysis, which he understands as something “showing itself in itself.” This “self-showing” of phenomena is always the basis for appearances, which, in fulfilling their function, refer back to the phenomena. This is a fundamental relationality in which we encounter things as they exist for us. Moreover, the “being toward the world” of Dasein “is essentially taking care” (Heidegger 2010, 57). This is care in an ontological sense, not in the sense of affection or distress or similar; we take care about the things we encounter because of the basic fact that they are caught up together with us as the world in which we have our being. It is only “when we put ourselves in the place of taking care in the world” (Heidegger 2010, 67) that other beings become accessible. This is (an extremely condensed version of) the path that leads Heidegger to his consideration of tools, and our relation to them. He states that the beings we encounter when taking care in the world are not just things, but rather useful things, or equipment1—things that we can use in order to do something. This structure of “in order to” that is essential in our relation to useful things also contains other references: useful things are related to other useful things. It is this totality of useful things that is discovered “always already … before the individual useful thing” (Heidegger 2010, 68). Moreover, when we deal with a useful thing, we are not dealing with or even considering it as such and in itself, but rather in the sense and to the extent that we are able to adequately appropriate it for something.

Changing Things: The Future of Objects in a Digital World

80

Here we arrive at the breakthrough insight in Heidegger’s analysis of our relations to things, in which he has been discussing the example of hammering with a hammer: In such useful dealings, taking care subordinates itself to the in-order-to constitutive for the particular utensil in our dealings; the less we just stare at the thing called hammer, the more we take hold of it and use it, the more original our relation to it becomes and the more undisguisedly it is encountered as what it is, as a useful thing. The act of hammering itself discovers the specific “handiness” [“Handlichkeit”] of the hammer. We shall call the useful thing’s kind of being in which it reveals itself by itself handiness [Zuhandenheit]. It is only because useful things have this “being-in-themselves” [“An-sich-sein”], and do not merely occur, that they are handy in the broadest sense and are at our disposal. (Heidegger [1927] 2010, 69)

And further: What is peculiar to what is initially at hand is that it withdraws, so to speak, in its character of handiness in order to be really handy. What everyday dealings are initially busy with is not tools themselves, but the work. What is to be produced in each case is what is primarily taken care of and is thus also what is at hand. The work bears the totality of references in which useful things are encountered. (Heidegger [1927] 2010, 69)

The paradox here is that the more closely we engage with something and it is thus revealed to us as what it is, the less we are actually aware of it. It “withdraws” as we focus on what it is that we are using the thing to do. It comes into objective presence only when its “handiness” turns into “unhandiness,” when our use of it breaks down for some reason as it becomes damaged or is not working well for our purpose. Rather than being withdrawn in smooth and effective use, it becomes conspicuous and obtrusive as an obstinate, broken tool. Under normal circumstances, then, when things are serving their function as equipment for us, we are caught up in our activities and the things are, in a fundamental sense, not present for us. Heidegger summarizes this interpretation by saying that “being-in-the-world signifies the unthematic, circumspect absorption in the references constitutive for the handiness of the totality of useful things” (Heidegger 2010, 75). Although Heidegger’s concern was ontology at a fundamental level, this insight about things withdrawing from awareness during effective use can

A Conceptual Toolkit

81

be applied in quite practical ways. It has, for example, been used in relation to understanding interactions with computational things in the context of human–computer interaction, which originated as a project of designing interfaces that would make it possible for humans to use complex computational systems (e.g., Winograd and Flores 1986; Dourish 2001). Withdrawal in handy use can even be seen as a goal for design: when something works well, it should disappear from the user’s awareness and allow for focus on the project at hand. When computers and graphical user interfaces were relatively new this might have seemed like an ambitious goal indeed. However, now computational devices are commonplace and use a number of familiar standard interaction conventions (e.g., dragging, swiping, opening and closing windows, and so on); so rather than being clunky and frustrating to use, they are now often (even if not always!) easy to use at the level of basic interaction, even working seemingly “automagically” or being (by careful design) addictively pleasurable to use. This sentiment can also be seen explicitly in the advertising for the iPhone X, which is described as having a screen “so immersive the device itself disappears into the experience.”2 This has significant implications as “use” of a thing is now increasingly distinctly different for different kinds of users versus for the owners of the broader system; what appears to us as a “useful thing” is in many cases just one of many manifestations of a fluid assemblage. Consider Facebook for instance: to the typical user it is a social platform for staying in touch, for sharing updates, photos, and more; to the typical advertiser, it is a highly structured marketplace where the data gathered allows advertising to be targeted to very specific groups based not only on general demographics or distribution patterns, but on personal data, interests, mobility, properties of their social network, and so on. Indeed, entering the Facebook website as an average user, one is greeted by statements like: “Connect with friends and the world around you on Facebook” and “It’s free and always will be.” Entering as a potential customer on Facebook business, on the other hand, one is informed that “Advertising on Facebook makes it easy to find the right people, capture their attention and get results” along with examples of success stories of “how businesses similar to yours are growing with Facebook marketing.” That different stakeholders, and in particular producers and consumers, have different relations to things is certainly nothing new, nor is it particularly surprising that such groups have different reasons for attending to a given thing. But what has happened in

82

Changing Things: The Future of Objects in a Digital World

cases such as Facebook and Google is they are now quite different things to different people: on one hand a way of staying in touch with friends, on the other a marketing platform; on one hand a tool for finding information, on the other a data harvesting machine to optimize targeted advertising. It is crucial that we do not dismiss this complexity in simple terms such as whether this is good or bad, right or wrong, but rather actually acknowledge that what seems to be a fairly straightforward relation to technology as something that is there for us as useful things is really much more complex. These fluid assemblages are in fact anything but straightforward when it comes to how, and to whom, they come to appear as “useful things”—and what those uses are.

A Conceptual Toolkit

83

Engagement Another aspect of things that was a focus in Heidegger’s later work is the way in which they entail a gathering of different elements. Distinguishing things from mere objects, he asserts that things stand on their own and are not dependent on our representations of them as objects are. He goes on to describe how things rather gather together in themselves the “fourfold” of earth and sky, divinities and mortals. Somewhat less poetically, he also connects to a more originary meaning of the word thing: The Old High German word thing means gathering and indeed a gathering for the negotiation of an affair under discussion, a disputed case. Consequently the Old High German words thing and dinc become the name for an affair; they name what concernfully approaches the human in some way, what accordingly is under discussion. (Heidegger [1994] 2012, 12)

This understanding of things as material gatherings around shared matters of concern has been picked up quite productively in design theory (Binder et al. 2011) and science and technology studies (Latour 2005), and is something to which we will return later on. Keeping for now though quite close to actual interactions with things, we can note that there is a certain basic ordering and connective function of things. Even if we do not follow Heidegger all the way to seeing earth and sky, divinities and mortals, in the things around us, it is possible to recognize that they are in some real sense about more than just the material object that is objectively present for us. Trying out this way of seeing with a simple exercise, we can think of the toothbrush that typically sits by the sink in the bathroom at home. It is not just an isolated object but a thing for us as we reach for it in the morning and evening (and a toothbrush in particular is even more literally a “thing for us” since its relations with humans tend to be quite strictly monogamous throughout its useful life). It is typically something that withdraws during use as we go through our daily rituals of brushing our teeth without giving it much thought (unless, perhaps, we have just been to the dentist and received remedial instructions in brushing). It is also something that does not exist on its own but also entails relations to other things: the tube of toothpaste, the sink and its running water, the cup it is stored in, etc.

84

Changing Things: The Future of Objects in a Digital World

More broadly, it also connects to other aspects of life; this could be memories of reminders from parents to brush early in life, everyday feelings of being prepared for the day ahead or a night of sleep, desire to rid one’s mouth of garlicky aromas that could negatively impact certain kinds of interpersonal relations, and so on. Expanding our scope of consideration beyond ourselves, we might also think of what will happen to a toothbrush after we dispose of it, when the plastic it is made of will exist in some form, in some place on Earth, for hundreds of years. A humble toothbrush can thus gather together in itself our everyday rituals, memories, social norms, interpersonal relations, planetary impacts and responsibilities, and more. Things, then, seem to invoke and lend themselves to certain ideas about how to carry out the affairs with which they are involved. The fact that our lives are thoroughly textured by technological things thus calls for consideration regarding the kinds of ideas and patterns that they support or discourage. This has been the project of neo-Heideggerian philosopher of technology Albert Borgmann, who develops what he refers to as the “device paradigm” to describe a pattern he identifies in the use of technology in contemporary life (Borgmann 1984). Whereas “focal things unify and gather, devices divide and scatter” (Strong and Higgs 2000, 32). He illustrates this with his famous example of the wood burning stove in contrast to modern (particularly North American) central heating systems. The wood stove gathers together a set of focal things and practices involving chopping and fetching the wood, building and tending the fire, and so on that can correspond to particular roles in a family. It serves as a focal point in the home that people gather around, perhaps cooking, warming oneself after coming in from the cold, or simply sitting around it. In short, using a wood stove entails engagement with one’s world and other people in it. In contrast, a central heating system disburdens users from the significant labor involved in using a wood stove for heating, instead providing heat as an effortless commodity in response to a setting on the control unit on the wall.3 The means of producing the heat are separated from the ends, and very minimal engagement and effort are required from users in order for it to function. Indeed, our contemporary computational devices exhibit more complex patterns of engagement, of presence and withdrawal, than we have seen before. In relation to the toothbrush discussed earlier, let us consider the development

A Conceptual Toolkit

85

of another personal tool we use on an everyday basis: a wristwatch. Although always a site for expression of technological progress and its increasing miniaturization, it has throughout also been a means of personal expression. Being part technology, part fashion, it is an interesting expression on what and how we think about and make use of technology. For instance, we may look toward the change from wearing it as part of the clothing, or on a chain in a pocket, to wearing it on the arm—and the associated differences in the acts of use. Or, we could think of the introduction of digital wristwatches, some even with computational abilities in the form of very tiny calculators, as digital technology started to diffuse into everyday life in the 1970s. More recent developments include the increasing use of sensors and communication capabilities, such as GPS to track position, accelerometers to track motion, or air pressure sensors to track altitude and barometric pressure. The use of sensors for heart rate, breathing, and other bodily measures that was first introduced for athletes and amateurs to track their training and progression has over time become part of setting and measuring an exercise or training regime also for everyday use. For instance, most GPS sports watches now connect to various other services to exploit the data collected. A key example is services such as Strava (strava.com), perhaps most known for turning every ride or run into a potential competition using a feature called “segments.” A segment is a part of a trail or road, such as a climb or a technical descent, and the ingenuity of Strava is to publish leaderboards for all segments. Thus, every run or ride can be turned into a local competition, the individual either aiming for a personal record on the segment or a better position on the leaderboard including winning trophies called “KOM” or “QOM” (King/ Queen Of the Mountain, a term borrowed from Tour de France). To illustrate, consider how a new verb—“Strava’d”—is used to describe implications for road use: In 2011, David Millar, the British pro now riding in the Tour de France, smashed the KOM on a circuit of Richmond Park in south-west London, among the most Strava’d roads in Britain. Using the bike he’d ridden to victory in the time trial at the Tour of Italy earlier that year, he completed the 6.7-mile loop in 13 minutes and 35 seconds, an average speed of 30mph. When it was pointed out the speed limit in the park is 20mph, his record was removed, the BBC took down a video of the ride, and Millar apologised for

86

Changing Things: The Future of Objects in a Digital World

his “naivety.” But the current record involves an average speed of 28mph and is still on the site, begging to be beaten. (Usborne 2013)

To understand what this becomes, we cannot just look at the different parts— the GPS watch, the sensors, the data collection, the services, the connections between them, what content users upload, etc.—we need to look at how they come together in the “things” we use. Until recently, such devices have been considered specialized sports equipment rather than general watches, but the tendency is clear: this kind of “bringing together” is quickly becoming part of things intended also for more general use. Consider the Apple Watch for instance: while it lacks some of the sensors mentioned earlier, it makes use of its continuous connection to the iPhone to provide additional functionalities. Although in some ways an extension of the interaction with the iPhone, it also adds important new aspects of wear such as the continuous tracking of certain bodily data such as heart rate and movement. It might not track your performance the way the sports watches do during a race, but there is certainly an element of performance and progress also here: how much do you move around, what is your average heart rate, how many stairs did you climb today, what was your sleep like, etc. But unlike the previous sports watches, the Apple Watch combines this new and extended functionality with an exterior design that instead strongly enforces our perception of the thing as a “watch”: ranging from its sleek and minimalistic form to the traditional faces it is often presented with, it is at the same time a statement of innovation and a strong statement of continuity. With such technological advances in mind, it is not difficult to imagine also the toothbrush as a site of bringing together, of making a wider range of relations, objects, data and more, present through the engagement with this “useful thing.” Given its place in our daily routines, it could become a means for presenting what lies ahead during the day or summarizing what has passed, equipped with sensors it could become a tool for tracking health (saliva is a promising biofluid for early detection of a range of diseases), and so on and so forth. But what is here hidden, and what is revealed? And more than anything, what is through technology transformed from one useful thing into another, as when the paved road is transformed from surface for easy transportation to race track, along with our relations to other people, the different social rules we assume apply, and much more?

A Conceptual Toolkit

87

Borgmann’s framework has been rightly criticized for its extreme pessimism about the merits and effects of devices, and for the nebulousness of his concept of engagement. Verbeek (2005), for example, argues that things that Borgmann would see as devices that disburden users in ways that threaten focal things and practices, such as a CD player or television, do not necessarily do so. While it might be possible to recognize a general pattern of disengaged consumption as a diagnosis that could be appropriate in many cases, it is less clear cut to apply it in blanket terms to one or another particular kind of thing. However, the concept of engagement is still a useful one for considering the character of interactions with and through things, what and how much they require of us, and how they connect us to or disconnect us from our material and social worlds. Similarly to Heidegger, we can also see here in Borgmann’s analysis a dynamic of presence and withdrawal, a contrast between a condition where things are characterized by a certain openness in their function and where they light up and activate networks of rich relations in the world around them, and a condition in which they hide the details of their operation from view and enable effortless, disconnected, and distracted consumption of commodities. The latter situation in which operation is hidden from view and does not require user effort or awareness seems to be particularly characteristic of modern computational technologies, and indeed increasingly the other kinds of things that now often have computational components embedded in them. Think of vehicles, for example; it is from these things that we get

88

Changing Things: The Future of Objects in a Digital World

the expression of “looking under the hood” as a way to open something up in order to try to figure out how it functions or, more typically, why it is malfunctioning and what might be done to repair it. Now, however, it seems vehicle components are being gradually engulfed in various coverings that shield them from observation and intervention by the “unauthorized,” while their computational components are visually and practically impenetrable in more fundamental ways.

Technological mediation One of the basic insights developed in philosophy of technology and related fields is that technological tools are involved in and shape human activity in non-neutral ways. As political theorist Langdon Winner states: “Technologies are not merely aids to human activity, but also powerful forces acting to reshape that activity and its meaning” (Winner 1986, 6). This makes it important to examine what technologies actually do as they are taken up and used by humans, and how they mediate human actions and interactions. This technological mediation is an explicit focus in the subfield of philosophy of technology known as postphenomenology. Postphenomenology is a philosophical approach initiated by Don Ihde around the 1980s through bringing insights and orientations from pragmatism and science studies into phenomenology (Ihde 2008). Postphenomenology attempts to get out of the subjectivity of phenomenology by analyzing the structure of relations between human and world. Variations in possible structures are emphasized in order to avoid reductionism, and variational analyses are one of the primary postphenomenological methods. One of the key moves that Ihde made was in foregrounding the role of technologies in frequently mediating human–world relations, such that human–world relations became instead human–technology– world relations. This structure is particularly characteristic of scientific practice, and since Ihde was heavily influenced by science studies this was one of his key concerns. As he states: “Inter-relational phenomenology not only provides a rigorous analytic process for understanding technologies, but also shifts and complements the understanding of science praxis as being technologically embodied and entailing human perception and action” (Ihde 2008, 7).

A Conceptual Toolkit

89

In his phenomenology of technics, Ihde (1990) develops a basic set of relational structures that have provided the primary analytic framework for the field of postphenomenology. The general intentionality relation he schematized as human-technology-world. The other initial set of relations elaborated on this, and consisted of embodiment relations, hermeneutic relations, alterity relations, and background relations. Embodiment relations [(I-technology) → world] describe the relation where a technology is taken up as part of one’s own bodily apparatus in orienting toward the world, as is the case with a pair of eyeglasses. A person wearing eyeglasses experiences them as transparent, both literally in the case of eyeglasses but also phenomenologically. Hermeneutic relations [I → (technology– world)], on the other hand, involve a technology that converts some part of the world into a text that can be read and interpreted. This is the case with a thermometer that registers the level of heat in the environment and produces a numeric display. A person can then interpret the number displayed based on previous experience of how a given temperature scale relates to how hot or cold it feels. Alterity relations [I → technology–(-world)] involve technologies that are related to as, such as a robot. These relations are becoming increasingly common in voice-activated personal assistants in smartphones or tabletop digital assistants. Background relations concern technologies such as those used for climate control, which tend to always be running in the background but receive little direct attention (unless they break down). Other formal relations have also been added to this initial set that Ihde developed, including cyborg relations [(human/technology) → world] (Verbeek 2008) in which human and technology are fused into a new entity, and digital material mediation [I → ([trace | substrate] → world)] (Wiltse 2014) in which digital technologies with multiple functional component parts mediate perception and action by making activities visible in more complex ways. Another key concept from postphenomenology is that of multistability. This points to the fact that it is possible to relate to each artifact in multiple ways. Some of these ways are at the micro-perceptual level, as with illusion drawings where one typically sees one figure in the drawing initially but can then see a completely different second figure through shifting focus (Ihde 1990). Multistability is also possible at the macro-perceptual level

90

Changing Things: The Future of Objects in a Digital World

where culture shapes representational and perceptual practices. Variational analysis is done by identifying or brainstorming different possible multistabilities. Postphenomenology, in accordance with its roots in phenomenology, is oriented toward subjectivity and the technological mediation of perception that structures and enables it. It can help us recognize the many and non-neutral ways in which technologies mediate engagement with the world. However, when it comes to fluid assemblages, this mediation becomes substantially more complex than is the case with other more traditional technologies. In Chapter 9, we will work with a couple elements in the postphenomenological toolkit to illuminate these dynamics.

Technospheres Even as it is possible to consider social implications of technological tools at the rather high levels of forms of life and the administrative systems of industrial society, it is also important to at the same time remain attentive to more intimate technological relations at the level of individuals. For after all, humans are not just hapless dupes of “the system,” even though the systems in which we live and work do exert powerful forms of influence over our activities and forms of life. Rather, we are active agents and subjects of our own lives. Yet our very subjectivity in the world is shaped to a large extent by the technologies all around us and by the connections and interactions that they mediate. In fact, as media theorist Mark Deuze has argued, we can’t get out of media (Deuze 2014). In the contemporary globalized world, individual subjectivities have a distinctly global horizon. In the context of a discussion of the disjunctures of the landscapes constituted through global cultural flows, anthropologist Arjun Appadurai (1996) proposed the concept of technoscapes as one dimension for exploration (the others being ethnoscapes, mediascapes, financescapes, and ideoscapes). He defines technoscapes as “the global configuration, also ever fluid, of technology and the fact that technology, both high and low, both mechanical and informational, now moves at high speeds across various kinds of previously impervious boundaries” (Appadurai 1996, 34). Appadurai’s other “scapes” also imply technological mediation in various

A Conceptual Toolkit

91

ways. In addition to the global interactions and imaginaries enabled by technologies, the mere presence of certain technologies and their affordances shapes our experience of ourselves and our possibilities (Kiran 2012). Another term for the technologies that constitute our environments and texture our worlds is technosphere, proposed by philosopher Michel Puech in his discussion of “the ethics of ordinary technology” (Puech 2016). Many of our decisions about how to live our lives, relate to others, configure our material conditions (to the extent possible), and go about our ordinary activities involve decisions about how to relate to technologies that shape and mediate our actions and interactions. This more intimate layer that serves as the interface between subject and world Puech terms the proximal technosphere, and it is in this existentially significant site he finds also a space of ethical significance. A key example is the mobile phone, something that now mediates many everyday activities and interactions. He calls for conscious development of ordinary, practical wisdom and associated virtues and skills that are geared to life in the technosphere.

Technical Now that we have briefly considered the human side of technological mediaion, we need to turn to what happens at the level of the technologies. And here we also need to make a very important shift from thinking in terms of human agency to thinking in terms of also technological agency. To consider what these things actually do, it is necessary to look not only at what humans do with them but what they do themselves—both in response to human action and more independently. However, at the same time it is important to remember that all technologies are, by definition, part of the artificial world that is created, configured, and maintained by humans. They are, in very important ways, always sociotechnical. Our focus on the technical here is then meant to be just that: a focus, but one that is not meant to in any way remove technologies from their social contexts of production and use. However, we also need to be able to examine their properties and character, their durable and consequential presence in the world. And it turns out that in doing this we always end up running back into the social anyway.

92

Changing Things: The Future of Objects in a Digital World

Protocols and interfaces Just as we considered how technologies can mediate human interactions, we need to dig down one more level in order to look at how technologies communicate, interact, and affect each other—at their protocols and interfaces. A discussion of protocols and interfaces could be quite technical, and indeed technical details can be important to examine in order to understand how things really work. However, our interest here is more at the conceptual level of trying to understand broadly how things connect, both at the level of facilitating human use (as in when a message is sent across a network) and at the more abstract level of possibilities and openness to connections. While the former shapes what people are able to do with technologies during more or less normal and intended acts of use, the latter shapes what is actually possible in terms of how things are able to be accessed and affected, and eventually how ecosystems evolve. Our concern here is with the forms that computational things can take, and particularly those aspects of form that determine their possibilities for interaction. The two terms used here as markers for these matters are quite similar, but also reference slightly different dynamics. Protocol, in a general sense, refers to formal rules governing communication or procedures in particular settings. For our purposes here we are most interested in the computing protocols that govern the exchange of information across networks, as in the TCP/IP protocol governing information exchange over the internet, or the DNS protocol for converting names into web addresses. However, it is worth bearing in mind also this more general sense of protocols not just as existing technical standards, but rules governing interaction and exchange. These in turn also determine the characteristics of interfaces for specific computational technologies, which includes their user-facing interfaces but also those that let them interact with each other. There are also both standard interfaces, such as the basic interface for a particular operating system (Mac OS, Android, etc.), and more local interfaces that are designed for particular applications. We might think of protocols as rules and standards that are in effect more or less everywhere, and interfaces as the external structures and interactive capabilities that actually exist in particular instances of computational things. As media and cultural theorist Alexander Galloway argues:

A Conceptual Toolkit

93

Protocol is a language that regulates flow, directs netspace, codes relationships, and connects life-forms. Protocol does not produce or causally effect objects, but rather is a structuring agent that appears as the result of a set of object dispositions. Protocol is the reason that the Internet works and performs work. In the same way that computer fonts regulate the representation of text, protocol may be defined as a set of instructions for the compilation and interaction of objects. Protocol is always a second-order process; it governs the architecture of the architecture of objects. Protocol is how control exists after distribution achieves hegemony as a formal diagram. It is etiquette for autonomous agents. It is the chivalry of the object. (Galloway 2004, 74–75)

Galloway’s larger argument, referenced in the preceding quote, is that protocol is the predominant form of control that exists in society after decentralization, a form of organization for which bureaucracy was then the corresponding form of control. The protocols of networked computing fall into two types: those that are radically distributed (as in TCP/IP for data exchange) and those that are rigidly hierarchical (as in DNS addressing information that uses a hierarchical inverted-tree structure). Galloway argues that this contradiction at the heart of networked protocol is what enables its generativity, which requires universalization and homogeneity, standardization and openness, control and freedom (Galloway 2004). In formal terms, protocol provides basic common structure for the connective possibilities of things. Those possibilities can be realized in concrete interfaces, while shaping the forms that they can take. This dynamic has been illustrated nicely by interaction designer Timo Arnall, who explored the protocol of RFID (Radio Frequency Identification) as an “immaterial” that provides certain types of possibilities for interface design (Arnall 2014). RFID technology consists of a reader and a small tag that can be embedded in all kinds of objects (such as a metro or library card) and store a small amount of data that it transmits wirelessly when the reader activates the tag. However, the fact that the interface is invisible means that it can be difficult for designers to work with it as a design material and for users to understand how it works (and does not work). For example, people have been concerned about the possibility for criminals to easily collect personal information without detection, say by walking by someone with an RFID reader that is able to scan the RFID-enabled cards in someone’s pocket. In order to address this, Arnall

94

Changing Things: The Future of Objects in a Digital World

worked in the mode of discursive design to visualize technical explorations of RFID as a design material. For example, for one of his projects, “Ghost in the Field,” he used the technique of light painting to create a visualization of the three-dimensional physical space in which a tag and reader are able to interact. This provides a clearer picture of what is possible (e.g., scanning the top of the tag from a few centimeters away) and what is not (e.g., scanning a tag from a meter away). While the zone of interaction between RFID tag and reader exists in a clearly delimited physical space, other types of interactions get more complex and difficult to visualize. Natural language and gesture-based interaction are prime examples. This interaction complexity at the interface promises to only increase, a situation that Lars-Erik Janlert and Erik Stolterman (2014; 2017) trace to the combination of increasing technological complexity and device miniaturization. They suggest that this situation will lead to an increase in surface-free interaction modalities such as gesture, sound, heat, smell, wind, movement, and so on, and possibly new approaches to interaction in terms of ecosystems and fields. In such an interactive environment, “users” are immersed in an ecology in which they are not conceived as interacting with particular, targeted objects one at a time but rather are moving in situational and interactional “force fields,” causing minor or major perturbations in their environment by their moves and actions while being guided, buffeted, seduced, or affected in any which way by constant movements and changes in their environment taken as a whole. (Janlert and Stolterman 2014, 531)

These are examples of an underlying trajectory set in motion by technology leaving the scale of human perception (Maeda 2000; Redström 2005). As discussed in Chapter 2, there used to be a rather direct relationship between surface and technological complexity, but when “surface” was replaced by “interface,” this relation became completely arbitrary. Whereas the surfaces of earlier technologies were partly determined by their actual workings, interfaces are designed with respect to their intended functionality, thus enforcing rather than revealing their agency. This “ecological” mode of interacting with things is quite new, in line with our experiences in the physical environment perhaps but not as much

A Conceptual Toolkit

95

when it comes to technological things. However, it also brings into focus the two senses of interface that we want to highlight here: the interface as the interaction mechanism between person and thing, and the interface as the interaction mechanism between things. It seems that these interfaces will become increasingly difficult to clearly identify and disentangle, but this does not mean that they are not there. On the contrary, it is the multiplicity of interfaces, in the formal sense of connection points between entities, that enables expanding ecosystems and interaction complexity. One good example of increasing device complexity leading to increasingly complicated interfaces is the remote control. Remote controls, particularly those for television and other media devices, are notoriously frustrating to use due to their very large number of buttons. Or, the minimalist Apple TV remote provides an example of another direction that can be taken in managing complexity, where all kinds of media come through one box and the complexity is managed through nested on-screen interactive menus rather than individual buttons on a remote. However, the remote control is a good example in another sense as well, as pointed out by Caetlin BensonAllott (2015) in her study of the remote control: remote controls offer control rather than power. Their marketing often contains connotations of the latter, but what they actually offer is control based on a pre-defined match between certain actions and effects. This is important to keep in mind as interfaces perform disappearing acts and become more “intelligent” and “intuitive”—the possibilities are programmed, and not unlimited. They are enabled but also constrained by pre-existing, if complex and responsive, structures. The increasing connectivity and complexity of devices is characteristic of a state of affairs commonly referred to as the Internet of Things (IoT), in which all kinds of things are connected to the internet and able to interact with each other and plug into a variety of data-intensive cloud computing and social media platforms and services. IoT has of course been described as bringing “revolutionary” changes to everyday life and industry, while also raising new kinds of concerns regarding security, privacy, distraction, and so on (Greengard 2015). These risks actually highlight another structural aspect of protocols and interfaces: the parameters of a thing’s openness to connection. While these are designed for certain types of intended connections, they can also enable others that are subversive or even hostile. It is in this context that hacking and

96

Changing Things: The Future of Objects in a Digital World

cyberwarfare are our new realities, with vulnerabilities also networked and distributed. However, the technical novelty and surrounding hype can distract from underlying dynamics that have historical precedent. Architect and writer Keller Easterling (2012) points to the experiments of the architects Cedric Price and Christopher Alexander with active form (in contrast to object form) as early rehearsals of an IoT. This is a way of seeing the form of space itself as carrying information through activity. It is a disposition in which the action is embodied as potential in relations and relative positions, unfolding as performance over time. Architect and theorist Lars Spuybroek makes a similar case in his reading of John Ruskin’s study of the Gothic, which he updates as an ontology of the digital (Spuybroek 2016). Gothic form was vital and active, harboring a reserve of changefulness in its internal logic of growth and variation. Spuybroek sees here a logic and aesthetic for the digital, relevant now more than ever. Finally, this last point touches on one other noteworthy aspect of interfaces: the aesthetic. Indeed, this is the subject of much work in design, a concern with form ranging from the fit or interface between user and thing (ergonomics in the physical sense, human–computer interaction in the more cognitive) to the semiotic and the webs of associations and meanings that a form activates. And of course the aesthetic is political, and computer software can be seen as having parallels to ideology (Galloway 2012). Software and interfaces mediate (always imperfectly) between the external and the internal, and operate in the realm of simulation. They both reflect and obfuscate what goes on. Moreover, they are functional: code is language that is executable. Discussing the tensions and translations between code as language and machine, Galloway also offers another observation that resonates intriguingly with our conception of fluid assemblages: The dialectical movement between fluidity and fixity, seen in the internal workings of software where states and state changes carry the day, is precisely the same political problem posed by ideology … Software might not be narrative in the strict sense of the word, but it still might have a beginning, middle, and end—to paraphrase Aristotle—even if those narrative moments are recast as mere variables inside the larger world of software simulation. Thus too might ideology be recast in digital format. (Galloway 2012, 72–73)

A Conceptual Toolkit

97

Interfaces tell stories about what is going on in machines and in the aspects of the world that they mediate, but they and the software underlying them also do things themselves. Protocols and interfaces are, then, rules and zones of connection, sites of affecting and being affected. They are the connective tissue of systems and ecosystems, the trading zones that enable flows and exchanges and cascading effects. In (at least) computational systems, they are activated by algorithms.

Algorithms Underlying the user-facing surfaces and interfaces of digital networked things is the running code that makes them tick. There are computational processes happening at the level of the device itself, and also at the level of interconnected components, services, and infrastructures. So, we turn now to consider algorithms from a perspective of considering the mediations and cultural work that these computational entities perform. First, what are algorithms in terms of how they show up in the world? Ed Finn begins his superb study of algorithms with a good roundup: The word algorithm frequently encompasses a range of computational processes including close surveillance of user behaviors, “big data” aggregation of the resulting information, analytics engines that combine multiple forms of statistical calculation to parse that data, and finally a set of human-facing actions, recommendations, and interfaces that generally reflect only a small part of the cultural processing going on behind the scenes. Computation comes to have a kind of presence in the world, becoming a “thing” that both obscures and highlights particular forms of what Wendy Hui Kyong Chun calls “programmability.” (Finn 2017, 16)

Here Finn points directly at one of the big puzzles of fluid assemblages: how can computation become a “thing” with a presence in the world? It is here at the level of algorithms that we find a key part of the answer. It is also worth noting how quotidian our interactions with algorithms have become. While many operate outside of our awareness and understanding, we are also now accustomed to algorithmic recommendations (of the form “since you enjoyed this, you might also like … ”) and wondering about how the Facebook algorithms determine what should be shown in our newsfeeds (Bucher 2012).

98

Changing Things: The Future of Objects in a Digital World

In considering algorithms, we need to first go even further back than the algorithms at work in a single thing to consider the historical logic and assumptions underlying them that guide what they do. Finn (2017) points to magic and sorcery as forerunners of the kind of logic in which words actually do things in the world, and to cyberpunk novels as explorations of how this logic plays out in the (imaginary) cultural space of computation. In addition to this mythological origin story is also the pragmatic engineering conception of algorithms in terms of problems and solutions, which is now the main logic governing computational development. This logic is also underpinned by what David Golumbia (2009) refers to in his critique of “computationalism”: a belief in the power of computing that tends to be associated with instrumental reason, the “essential politics of the algorithm” (Finn 2017, 18). The assumptions about the nature of the universe and how it can be dealt with that underlie the pragmatic conception of algorithms Finn (2017) refers to as “effective computability.” He traces it back to the Turing machine and its suggestion of universal computation, and to attempts at mathematical proofs of effective computability that seem to indicate a movement toward universal truth, but always with a remainder that does not fit left over. Algorithms have, then, always “encoded a particular kind of abstraction, the abstraction of the desire for an answer” (Finn 2017, 25). One key early site where notions of effective computability were developed was in cybernetics. Cybernetics was developed after the Second World War as a field concerned with studying messages as a means of communication and control (Wiener 1968). It claimed to be a universal science, applying to and explaining social as well as physical reality (Bowker 1993). Importantly, it was founded on a probabilistic theory of messages, reflecting the shift from Newtonian mechanics to probabilistic models of the physical universe earlier in the century. This meant that completely accurate and certain models were never a goal in cybernetics, since they were then known to not be achievable. Rather, the goal was information and models that were adequate for some particular purpose. This paved the way for computer simulations being seen as reasonable objects of study in relation to all sorts of physical, social, and biological phenomena (Finn 2017). In addition to the “good enough” logic of probabilistic simulations, one other significant legacy of cybernetics for our purposes is its focus on how entities use information to adjust to their environments in order to maintain

A Conceptual Toolkit

99

their integrity (homeostasis) and resist decay (entropy). This includes both living and non-living organisms. In the words of cybernetics founder Norbert Wiener (1968, 27): It is my thesis that the physical functioning of the living individual and the operation of some of the newer communication machines are precisely parallel in their analogous attempts to control entropy through feedback. Both of them have sensory receptors as one stage in their cycle of operation: that is, in both of them there exists a special apparatus for collecting information from the outer world at low energy levels, and for making it available in the operation of the individual or of the machine. In both cases, these external messages are not taken neat, but through the internal transforming powers of the apparatus, whether it be alive or dead. The information is then turned into a new form available for the further stages of performance. In both the animal and the machine this performance is made to be effective on the outer world. In both of them, their performed action on the outer world, and not merely their intended action, is reported back to the central regulatory apparatus.

Collecting information and feeding it back into ongoing operation is also a central function of fluid assemblages. It is also worth highlighting some characteristics of this kind of apparatus that Wiener points or alludes to in this context. First, the fact that an organism collects information means that it is equipped (we could say programmed) to do so, and moreover that the information it is interested in collecting is that which helps it with furthering its goals as an organism. Second, this information is not “neat,” but is rather transformed by the apparatus. Third, this information is then fed back into ongoing operations, and in a way that has an effect. We can then formulate each of these as questions regarding things to look for in specific instances of fluid assemblages (which can be seen as organisms in the language of cybernetics): (1) what information is it equipped to collect; (2) in what ways does it transform the information it collects; (3) how and where is the information fed back into ongoing operations; and (4) what effect does this have in other parts of the assemblage and its larger environment? We will also return to consider algorithms in more detail in the following sections as we scale up and out. But for now it is worth extending one more plank to help us make the bridge to consideration at these higher levels. Returning

100

Changing Things: The Future of Objects in a Digital World

to Finn (2017), he suggests that the algorithm serves as the alchemical realm where the material and symbolic “operate in productive indeterminacy” (p. 34). He uses the ubiquitous progress bar as an example, since it both maps “progress” in a way that we do not believe completely reflects underlying reality, and yet does reflect the functional reality that the software installation is not yet complete and nothing else is going to happen until it is. Turning then to the implications, he states: As our generally unthinking acceptance of the progress bar demonstrates, we are primed to accept these magical calculations on multiple levels. We believe in the power of code as a set of magical symbols linking the invisible and visible, echoing our long cultural tradition of logos, or language as an underlying system of order and reason, and its power as a kind of sourcery. We believe in the elegant abstractions of cybernetics and, ultimately, the computational universe—that algorithms embody and reproduce the mathematical substrate of reality in culturally readable ways. This is what it means to say that an algorithm is a culture machine: it operates both within and beyond the reflexive barrier of effective computability, producing culture at a macro-social level at the same time as it produces cultural objects, processes, and experiences. (Finn 2017, 34)

We will circle back to these aspects of algorithmic culture later on. But first, we need to dive down one more level. As protocols and interfaces are activated by algorithms, algorithms in turn rely on data and information as the material that they process.

Data and information At a basic level, computational technologies deal with data and information. This is their raw material, the stuff that they process in various ways that enable the complex behaviors they perform. It is also associated with communication, as in the common use of the term “information and communication technologies” (ICTs). This is also one point at which a distinction between a technology seen as a tool and one seen as a medium for information and communication becomes particularly blurry, as ICTs have been at the heart of changes in organizational practices, information flows, and management of personal affairs in what has been termed the “network” or “information”

A Conceptual Toolkit

101

society (Castells 2000; Kallinikos 2007). In short, they help get stuff done, and they help people do things together. But the particular angle that we are concerned with here is more to do with the role and character of information in itself within computational systems. As mentioned earlier, information emerged as a key concept of cybernetic thought that also underlies computational technologies. Data is a similar concept that typically signifies something like the smallest unit of meaning, whereas information implies something that has been in some way rendered meaningful—that can inform. Data requires abstraction. In order for the analog world to be turned into data, details must be left out according to some kind of ordering and representational scheme. The map cannot be coextensive with the territory, yet through leaving out details the map becomes much more useful in other ways. However obvious this observation may be, it is very important to keep this dynamic in mind and to inquire critically into the origins of data. This is especially the case when data shows tendencies of growing into big data, expanding outside its initially intended scope of use, or traveling between different contexts of interpretation. For insight into some of these issues, we can turn to the now classic account by science studies scholar Bruno Latour (1999) of how “circulating references” are created in scientific practice. It is noteworthy that he does not really use the term data—except to say that one “should never speak of data” (Latour 1999, 42)! Rather than considering data as what is given, he suggests speaking instead in terms of “achievements.” Because data that is produced and that can lead to better understanding is an achievement that involves a good deal of work. Latour (1999) shows how this works in his study of scientists investigating a border between savanna and forest in the Amazon to try to figure out whether it is the savanna or the forest that is advancing while the other is on the retreat. In order to collect the kind of data that will be helpful in answering this question (plants, soil samples, and so on), they must first construct a grid to impose on the area to be studied, literally unfurling lengths of string at the site in order to do so. This allows for establishing correspondence between samples and sites from which they originated. Another lovely ordering device is the pedocomparator, a box for soil samples containing a grid of smaller cardboard boxes that can fit in a drawer for storage but also turn into a suitcase for

102

Changing Things: The Future of Objects in a Digital World

transporting samples between field site and laboratory. Through this ordering, the scientists are able to both label and order the samples and also make sense of them visually, being now able to perceive in this condensed format a shift from forest to savanna in the changing color. Here there is more at stake for Latour than cataloging the practical activities and helpful technological accessories of scientific practice. It is rather a much more fundamental question regarding the connection between world and language, matter and form. Whereas the “canonical view,” following Kant, sees an unbridgeable gap between them in phenomena, Latour sees a chain of mediations operating as circulating reference. The dirt from the Amazon can be labeled with a coordinate, placed in a box in a grid, and become part of other successively higher-level representations about what is going on at the border of forest and savanna, some of which are published in a final report. While details of the local context are lost and reduced in these transformations, others are amplified as it becomes possible to compare many such references that are also much better able to travel than the box full of dirt. And importantly, it is possible to travel both ways, both up and down across this chain of mediations: from the dirt to the dense charts in the report, but also from the chart back down to the dirt samples and the location in the Amazon from which they came. These mediations are thus each phenomena, a combination of transformation and connection; and these phenomena circulate all along the chain of references. While this chain of references may exist in followable forms in scientific practice (at least ideally), things become more difficult outside of science. But it is worth remembering that any data that somehow represents some aspect of the world comes with this chain of mediations, which could, at least in theory, be followed back to the point at which it bumps up against the phenomenon in which it originated. One of the primary mechanisms involved in these mediations between phenomena and data is categories. Categories often serve as rather blunt instruments for culling away the particularities of local phenomena in order to make them fit larger ordering schemes in which they can be classified, collected, and compared. Categories can also become standards that ensure data compatibility and are used for regulation. They are also social, often messy, contested, and political in both enactment and consequences. Having a category for something enables collection of data that can make something

A Conceptual Toolkit

103

visible, but this also allows it to be more easily regulated and administered. These matters have been investigated in depth by Geoffrey Bowker and Susan Leigh Star in their important book Sorting Things Out: Classification and Its Consequences (1999). In it, they examine work practices and consequences around classifications such as the international classification of diseases (ICD), racial classification by the apartheid government of South Africa, and classification of nursing work. There are a number of reasons why classifications are important, but one that is particularly significant for our project here is the ways in which they can become embedded in information infrastructures that are in turn embedded in computational infrastructures. In this way, political and social struggles can be rendered as technical and eventually come to be naturalized (Bowker and Star 1999). Standards and categories valorize some points of view while silencing others. This is indeed necessary and inescapable, and not a bad thing in itself; but it is always an ethical choice. The same dynamic could be seen in Latour’s conception of circulating references discussed earlier, in which there is both amplification and reduction in moving along the chain of mediations between local particular and more abstract and universal reference. Even the concept of data serves as a frame delimiting what will be included for consideration and what will be left out, something that has significant consequences for approaching inquiry (Boyd and Crawford 2012; Markham 2013). Data and information are thus produced in ways that are not “natural” or given, but rather reflect particular points of view and agendas in relation to phenomena that change over time. Raw data is indeed an oxymoron (Gitelman 2013). It is always “cooked” in some way. It is (now stretching the metaphor to a perhaps inadvisable extent) prepared by particular chefs working in and for particular contexts, for consumption by particular entities with their own preferences and budgets and temporal horizons. We can also note that data collection now commonly occurs outside of scientific and work practice, encompassing many aspects of our lives. Some people even collect large amounts of information about themselves and their activities through self-tracking tools, which has become known as the “quantified self ” movement. And then of course there is all the data that is collected about us without our explicit (or fully understanding) consent when

104

Changing Things: The Future of Objects in a Digital World

we use things that are connected to the internet. It is now mightily challenging or even impossible to escape dataveillance, what digital media scholar Rita Raley describes as “the disciplinary and control practice of monitoring, aggregating, and sorting data” (Raley 2013). This is especially the case with fluid assemblages, which rely on data collection not only to customize their forms for particular users, but also to inform their larger networks of other interested parties that can later use it to more precisely target advertising, conduct surveillance, gather information about a particular population for scientific or regulatory purposes, and so on. Each function, each button or command, is thus potentially a data point reflecting a particular classification. For example, the architecture of Facebook now has a plethora of mechanisms by which users indicate how they relate to each other and to content; particularly interesting in this context is the addition of other “reactions” to the original “like” button, which allow for more nuanced responses to content but also much more precise information for Facebook. Links between things are also data—a significant dynamic when it comes to online environments and networked things. Now, if these complexities are inherent to data as such, what happens as we turn to “big data” where aspects of aggregation and thus to some extent practices of filling the gaps with projections start to have a significant impact? To begin to appreciate that quantity in this case also gets new qualities, consider the following remark from an article in The Economist in 2017: Uber, for its part, is best known for its cheap taxi rides. But if the firm is worth an estimated $68bn, it is in part because it owns the biggest pool of data about supply (drivers) and demand (passengers) for personal transportation. Similarly, for most people Tesla is a maker of fancy electric cars. But its latest models collect mountains of data, which allow the firm to optimise its selfdriving algorithms and then update the software accordingly. By the end of last year, the firm had gathered 1.3bn miles-worth of driving data—orders of magnitude more than Waymo, Alphabet’s self-driving-car division. (2017a)

In another article in the same issue, it is further argued that: Access to data also protects companies from rivals in another way. The case for being sanguine about competition in the tech industry rests on the potential for incumbents to be blindsided by a startup in a garage or an unexpected

A Conceptual Toolkit

105

technological shift. But both are less likely in the data age. The giants’ surveillance systems span the entire economy: Google can see what people search for, Facebook what they share, Amazon what they buy. They own app stores and operating systems, and rent out computing power to startups. They have a “God’s eye view” of activities in their own markets and beyond. They can see when a new product or service gains traction, allowing them to copy it or simply buy the upstart before it becomes too great a threat. Many think Facebook’s $22bn purchase in 2014 of WhatsApp, a messaging app with fewer than 60 employees, falls into this category of “shoot-out acquisitions” that eliminate potential rivals. By providing barriers to entry and early-warning systems, data can stifle competition. (The Economist 2017b)

The gaps that can be filled by “big data,” and increasingly also by the new forms of intelligence surrounding what is called “meta-data,” thus not only connect the dots when it comes to potential customers, their needs and desires, but increasingly also to the inherent logic of the market in question. While examples such as the ones mentioned earlier will soon be dated and others added, there seems to be underlying movement here that certainly points in a direction but that is also hard to predict—indeed, that it would be a company emerging from the plethora of software start-ups that eventually would shake the financial ground of newspapers worldwide by offering a different way of advertising was not something many anticipated. Perhaps it is therefore not so surprising that obtaining a “God’s eye view” using “big data” also becomes big business. To understand how such advantages are made and maintained using big data, let us consider an example from the opposite end of the spectrum, that of people forming also intimate relations using technological tools. In an article in The Guardian, Judith Duportail describes the experience and results of requesting and reading her data as registered by Tinder: I asked Tinder for my data. It sent me 800 pages of my deepest, darkest secrets. … At 9.24pm (and one second) on the night of Wednesday 18 December 2013, from the second arrondissement of Paris, I wrote “Hello!” to my first ever Tinder match. Since that day I’ve fired up the app 920 times and matched with 870 different people. I recall a few of them very well: the ones who either became lovers, friends or terrible first dates. I’ve forgotten all the others. But Tinder has not. … Some 800 pages came back containing

106

Changing Things: The Future of Objects in a Digital World

information such as my Facebook “likes,” my photos from Instagram (even after I deleted the associated account), my education, the age-rank of men I was interested in, how many times I connected, when and where every online conversation with every single one of my matches happened … the list goes on. (Duportail 2017)

Such combinations of data, however, do not only come together for the purpose of the “perfect match,” but extend beyond this particular application: The trouble is these 800 pages of my most intimate data are actually just the tip of the iceberg. “Your personal data affects who you see first on Tinder, yes,” says Dehaye [privacy activist Paul-Olivier Dehaye from personaldata. io]. “But also what job offers you have access to on LinkedIn, how much you will pay for insuring your car, which ad you will see in the tube and if you can subscribe to a loan.” (Duportail 2017)

Indeed, the notion of a “God’s eye view” based on big data might be suggestive, but there is something quite real about what new possibilities are being opened up through abstraction and aggregation here: how new structures spanning multiple contexts not only can be registered but can actually be put in use in runtime. Big data is still also just data that constantly feeds various algorithms and feedback mechanisms—scenarios such as the one involving Tinder discussed earlier is not the result of months of human analysis and intelligence, of ethical and other considerations leading to decisions; it is the outcome of machines running algorithms fed with data. It is also from this perspective we need to consider what is now emerging as “artificial intelligence”—intelligence understood not only in the sense of a capacity to learn or to reason, but increasingly instead in the sense of intelligence agencies gathering information to predict and control. An interesting example of how big data is used to form new assemblages is Netflix’s investments in producing original content. The origin of one of its early success stories, “House of Cards,” is said to stem from data harvested pointing to an effective intersection of viewing interests between political thrillers, director David Fincher, actor Kevin Spacey and the older British TV series. In an interview with The New York Times in early 2013, the company’s chief communications officer Jonathan Friedland said, “Because

A Conceptual Toolkit

107

we have a direct relationship with consumers, we know what people like to watch and that helps us understand how big the interest is going to be for a given show. It gave us some confidence that we could find an audience for a show like ‘House of Cards’” (2013). The article further comments that “Netflix has always used data to decide which shows to license, and now that expertise is extended to the first-run. And there was not one trailer for ‘House of Cards,’ there were many. Fans of Mr. Spacey saw trailers featuring him, women watching ‘Thelma and Louise’ saw trailers featuring the show’s female characters and serious film buffs saw trailers that reflected Mr. Fincher’s touch.” With respect to just how detailed such decisions based on data might be, Phil Simon in an article in Wired, the year after, uses the example of how the “House of Cards” cover was designed down to the proportions of color hues: At Netflix, comparing the hues of similar pictures isn’t a one-time experiment conducted by an employee with far too much time on his hands. It’s a regular occurrence. Netflix recognizes that there is tremendous potential value in these discoveries. To that end, the company has created the tools to unlock that value. At the Hadoop Summit, Magnusson and Smith talked about how data on titles, colors, and covers helps Netflix in many ways. For one, analyzing colors allows the company to measure the distance between customers. It can also determine, in Smith’s words, the “average color of titles for each customer in a 216-degree vector over the last N days.” (Simon 2014)

Indeed, this is a significant extension of the conceptual connection between the “programming” of computers and the “programming” of television as data is used to draw things together to form new assemblages and even define the spaces in which acts of creativity are meant to, or have to, take place. In many ways, this is an extension of what market, consumer, and user research has been about since its inception, and how designers make heavy use of “usercentered design methodology” to narrow down what might and might not work for a given target group. And still, from the wider perspective on fluid assemblages discussed here, this is moving beyond relations between forms of “programming” toward creating a close conceptual relation also between ensemble and assemblage.

108

Changing Things: The Future of Objects in a Digital World

Technological ecologies Some technological systems are designed from the top down, perhaps most commonly organizational systems where use and associated routines can be enforced bureaucratically. But in more general terms, things always exist in some kind of relation to other things, and they can be pulled together into ad-hoc “systems” in use. These evolving relations and practices can then in turn feed into more explicit design to support these interconnections in better ways. This recognition has been labeled with terms like “product ecology” (Forlizzi 2008) and “device landscapes” (Stolterman et al. 2013) in the field of human–computer interaction. However, there are already existing examples of intentionally designed and curated technological ecologies. Perhaps the most well known is Apple. Its ecosystem is certainly designed (and controlled) in a quite top-down fashion, yet this has also been done by implementing connections among its devices. This has been accomplished in large part through its user accounts, currently on the iCloud platform. This system not only enables the syncing and transfer of data and preferences across devices in current use, but also between old and new devices. It is a system that allows evolution both in terms of Apple’s own collection of devices and also in terms of an individual user’s collection as it is expanded and upgraded over time. And indeed this constant upgrading and lack of attachment to any particular device is very much part of Apple’s business model, and the rapid outdating of devices accompanied by muchhyped and “best ever” updates part of every Apple device owner’s experience.4 This dynamic can also be seen in similar platforms, such as Google and the suite of interconnected applications and devices it supports. Another mechanism for supporting development in technological ecosystems is the application programming interface, or API. These open protocols basically allow apps to become services for other apps, and to plug in to certain kinds of functionality that they offer. One notable early example is Twitter, whose API allowed an ecosystem of associated apps and services to grow up around it. Other types of mechanisms that are even further beneath the surface support things like cross-platform authentication and content serving and customization. This ecological structure of artifacts evolving in relation to each other and their use can be seen at a more foundational level in philosopher of technology

A Conceptual Toolkit

109

Gilbert Simondon’s structural analysis of technology (Simondon [1958] 2017). He looks at technical objects in terms of their genesis and emergence of consistency and convergence at three levels: the element, the individual, and the ensemble. He sees technologies (such as a car engine) starting as rather abstract ideas with each component conceived more or less independently and strictly in terms of its function. However, when technologies are actually brought into existence these structures have reciprocal influences on each other during operation (through generated heat or motion and so on, in the case of thermodynamic machines). This means that in order for it to function properly these components need to work together in harmony, and pressure for good and efficient operation can lead to a convergence of functions in components that might have originally been conceived of as doing one thing in the more “abstract” initial version of the technology. The end result of this evolution through a series of technical objects is “a system that is entirely coherent within itself and entirely unified” (Simondon [1958] 2017, 29). Simondon thus saw this internal structural rationality of technical evolution as something leading toward the most essential and concrete expression of a particular technical idea (as opposed to the competing “inessential” and superficial demands of customers for particular kinds of superficial styling). However, addressing similar but somewhat more contemporary dynamics, virtual reality and digital media pioneer Jaron Lanier (2010) laments what he refers to as technological “lock-in” through software, where existing structures guide further development in ways that effectively cut off other possible paths. Lock-in is, he says, “like a wave gradually washing over the rulebook of life, culling the ambiguities of flexible thoughts as more and more thought structures are solidified into effectively permanent reality” (Lanier 2010, 9). An example he uses is the MIDI format, a rigid structure for representing musical notes that previously “transcended absolute definition” (Lanier 2010, 9). The particular dynamics of technological ecosystems vary, and there are certainly significant differences between the industrial technical objects Simondon considered and the digital ones that Lanier did. However, for present purposes, it will suffice to note in summary that technical artifacts interact with each other both directly and indirectly in complex ecosystems; and that these interactions influence both configurations at particular moments in time

110

Changing Things: The Future of Objects in a Digital World

in particular contexts and the ways in which both artifacts and ecosystems develop over time.

Platforms and infrastructures Other concepts that are often used in relation to systems and ecologies are platforms and infrastructures. These terms refer to the relatively stable structures that provide common resources for the operations of other things. For example, classic examples of industrial infrastructures would include water and sewer systems, transportation networks, electrical grids, and telecommunications systems. The internet and World Wide Web have now effectively become infrastructures in the sense that they are common resources that are used for a vast array of processes and activities across sectors and industries, work and everyday life. However, the common sense understanding of infrastructure as inert substrate is not quite accurate. Research in science and technology studies has shown that infrastructure is fundamentally relational, becoming infrastructure in relation to organized practices (Star and Ruhleder 1996; Star 1999). In a seminal article, Star and Ruhleder (1996) proposed a set of characteristics of infrastructure: that it is embedded in other structures and sociotechnical arrangements, transparent in use, having a reach or scope beyond one particular site or event, learned as part of membership in a group or organization, linked with conventions in particular communities of practice, embodying standards that allow it to connect to other infrastructures and tools, built on an installed base with a certain inertia that resists change, and becoming visible on breakdown.5 The concept of platforms is somewhat similar, but has its roots in media studies and focuses on a somewhat different set of issues. While infrastructure studies “has highlighted key features of infrastructure such as ubiquity, reliability, invisibility, gateways, and breakdown,” key features in platform studies “include programmability, affordances and constraints, connection of heterogeneous actors, and accessibility of data and logic through application programming interfaces (APIs)” (Plantin et al. 2017, 2). The term “platform” itself has also served a key role in the discourse around networked computational systems and services, as technology companies use the term flexibly to position their offerings in relation to various contexts and concerns

A Conceptual Toolkit

111

(Gillespie 2010). A social media platform can be a “platform” for self-expression and even democracy, but then also “just a platform” when questions about regulation applying to media content providers come to the fore. As Plantin et al. (2017) have argued, both concepts are needed in order to understand entities such as Google and Facebook. While they have many characteristics of platforms, being highly programmable and supporting a range of apps, platforms and platform ecologies have also become in practice like large-scale infrastructures. For example, in the case of Facebook: Viewed simultaneously as an infrastructure and platform, Facebook presents a disturbing image. As an infrastructure, Facebook is progressively expanding and embedding itself in our daily existence, taking over more and more functions formerly provided by other, less restrictive means. The API, as a gateway, transforms Facebook from a centrally controlled system into something more like a network of independently developed, yet seamlessly interconnected systems and services. As a proprietary, largely opaque platform, Facebook filters our daily communicative acts through a profit-extracting sieve, deploying its intimate view of users’ activities and relationships for the benefit of advertisers and others, who in turn provide further data (via the API) for the Facebook social graph. As a result, its power to shape our communication behavior for its own ends increases. (Plantin et al. 2017, 12)

The means by which something like Facebook acts as a “profit-extracting sieve” are particularly relevant from a design perspective, since they are embodied in the interactive mechanisms that define the thing as it is available for use. Alaimo and Kallinikos refer to this as encoding, the organizing of “user platform participation along specific activity corridors (such as sharing, following, or tagging) that heavily stylize and shape user interaction” (Alaimo and Kallinikos 2017, 176). Other familiar examples of encoding are emoji characters, standardized representations of emotion that both facilitate affective communication and commoditize affective labor for the market (Stark and Crawford 2015). After being encoded, data about user activities on a platform are then aggregated into larger entities, which are subjected to further computation that leads to higher-level descriptions of platform activity (and of users and their relationships, as operationally defined by their platform activity) (Alaimo and Kallinikos 2017).

112

Changing Things: The Future of Objects in a Digital World

Another view of things is from the perspective of “big data,” which seems to have become the contemporary instantiation of the information-driven “network society” (Castells 2000). Specifically, this data that is the new fuel for all kinds of societal processes needs to come from somewhere. This creates a drive for an increasing number and penetration of (mostly passive) datacollecting sensors in everyday life—a key means by which big data becomes embedded infrastructure (Burdon and Andrejevic 2016). The passive nature of much of this data collection provides a direct challenge to design perspectives focused on user experience, since in this case things enter (and can ultimately affect) people’s lives while—often by design—not entering into or even being hidden from conscious experience. Given these dynamics, design perspectives turning toward connecting products within systems in order to provide services (e.g., Dubberly 2017) are a step in the right direction, but still fall remarkably short of adequately accounting for the (politico-economic) role and function of connected things in contemporary (post)industrial contexts.

Societal We have already in the previous sections encountered societal dimensions of things as they are part of our lifeworlds and as they exist as technologies that do particular things in the world. Now we turn to face these dimensions squarely, looking at the character and implications of things as both products of and actors in society.

Politics and actor-networks Critiquing the conventional view of things that considers them only in terms of their making and use, political theorist and philosopher of technology Langdon Winner has argued that, rather than being only neutral aids to human activity, they are also “powerful forces acting to reshape that activity and its meaning” (Winner 1986, 6). Moreover, he asserts that as “they become woven into the texture of everyday existence, the devices, techniques, and systems we adopt shed their tool-like qualities to become part of our very humanity” (Winner 1986, 12). Seen in this light, technological design becomes a historical

A Conceptual Toolkit

113

process of world-making, a perspective much in line with current scholarship in design studies (Yelavich and Adams 2014; Fry, Dilnot, and Stewart 2015). It also points to a different set of questions than those that are typically asked about technology. As Winner puts it: From this point of view, the important question about technology becomes, As we “make things work,” what kind of world are we making. This suggests that we pay attention not only to the making of physical instruments and processes, although that certainly remains important, but also to the production of psychological, social, and political conditions as a part of any significant technical change. Are we going to design and build circumstances that enlarge possibilities for growth in human freedom, sociability, intelligence, creativity, and self-government? Or are we headed in an altogether different direction? (Winner 1986, 17)

Winner (1986) also addresses head-on the fraught question of whether artifacts “have politics.” Although it is not terribly difficult to see that they can be used for political ends, the idea that there can be something inherently political about certain technologies is much more controversial. As one example of how this might be the case, he offers the now-classic case of the low-hanging bridges in New York.6 These were part of the significant developments in the city’s infrastructure that were led by master city planner Robert Moses. One effect of these low bridges was to prevent buses from using the parkways. Moses was, according to his biographer, a racist and a classist, and through the design of his bridges he was able to effectively limit the mobility of poor people and blacks who most relied on public transit, in particular discouraging their use of his prize development, Jones Beach. This was especially significant in that the bridges enforced the prejudices of their designer even after he was no longer around. The powerful role of technologies in effectively deciding particular states of affairs suggests that they be brought under the same kinds of conditions for public decision making as is the case with other forms of legislation. As philosopher of technology Andrew Feenberg states (building on Winner): The technical codes that shape our lives reflect particular social interests to which we have delegated the power to decide where and how we live, what kinds of food we eat, how we communicate, are entertained, healed

114

Changing Things: The Future of Objects in a Digital World

and so on. The legislative authority of technology increases constantly as it becomes more and more pervasive. But if technology is so powerful, why don’t we apply the same democratic standards to it we apply to other political institutions? By those standards the design process as it now exists is clearly illegitimate. (Feenberg 1999, 131)

Now the kinds of design that Winner and Feenberg discuss tend to be more along the lines of public infrastructure than what might be labeled as industrial design of individual products for mass production. Yet at the same time the differences might be more of scale than of kind, as even individual products can embody and enforce particular ideas about how to do things, and also connections to other elements within systems of varying levels of intentional design and formality. It is also the case that categories, discussed earlier as mechanisms used for producing data, can have politics, encoding particular models of the world in persistent technical forms that have consequences (Suchman 1994). As Susan Leigh Star states, explicitly referencing Winner’s low-hanging bridge example: “There are millions of tiny bridges built into large-scale information infrastructures, and millions of (literal and metaphoric) public buses that cannot pass through them” (Star 1999, 389). And as addressed in the previous section, the information infrastructures of contemporary data-driven society involve encoding activity and steering interactions through particular platform channels. This means that when designing a user interface for interactions that connect to and through these kinds of platforms, it really is politics all the way down. In summary, looking at technological things in terms of their politics means considering the ways in which they are implicated in distributions of power and authority, and in supporting or foreclosing certain forms of life. It does not mean politics in the sense of governments, elections, political parties, or similar—although of course these particular mechanisms of politics are also vitally important, and indeed all involve technologies. But there is also a more foundational sense of the political as the collective project of figuring out how to live together well. This involves decisions about how to provide for common needs, to ensure opportunities to engage in meaningful work, to support health and wellbeing, and generally to

A Conceptual Toolkit

115

create the kind of conditions that are conducive to the collective thriving of life on the planet. These are issues in which designed technological things are very deeply implicated, and a proper consideration of their social implications means inquiring into their political dimensions also in these larger senses. One way of approaching this is in terms of actor-network-theory (ANT), a sociological approach that, like (and in response to) social constructivism, was developed in science and technology studies. In contrast to much sociology, and as described by leading proponent Bruno Latour (2007), it refuses to use the abstract notion of “the social” or broad concepts such as “power” as explanations for certain states of affairs. Rather, actor-network theory views particular forms of more or less durable social order as remarkable accomplishments to be explained. It does this by looking for actors that have particular effects in the world and serve as mediators between other actors. In identifying actors it is careful to dissolve any pre-existing conceptual distinctions between human and nonhuman, science and culture, subject and object, and so on. It is relentlessly descriptive rather than explanatory, at least in terms of explanation done by “adding” or “framing” analyses through some additional theoretical apparatus that does not emerge through following the actors themselves and accounting for their own sense-making processes (Latour 2007). Rather than “society,” ANT sees imbroglios and collectives of humans and nonhumans that exchange their properties in distinctly nonmodern ways (Latour 1999). One of the distinctive aspects of ANT that makes it particularly relevant for our purposes here is that it considers nonhumans to be actors as well as humans. For example, a speed bump is an actor that has a particular kind of agency when it comes to getting drivers to slow down (Latour 1999). City officials who want drivers to obey a certain speed limit can use other mechanisms to achieve this result, such as posting speed limit signs that are backed up by law and the possibility of fines and social censure if caught not obeying them. But the speed bump, or “sleeping policeman” as the French call it, enforces this limit in a material way that enlists drivers’ compliance not only through a morally grounded appeal to obey the law, but by addressing their selfish desire to maintain the integrity of their cars’ suspensions. In this case

116

Changing Things: The Future of Objects in a Digital World

the city officials who want to achieve this speed regulation delegate the job of enforcement to the speed bump. Through work in science studies that showed the construction (and for ANT, not only social construction) of scientific facts, it came to be caught up in the “science wars” of the 1980s and accusations that it was attempting to tear down the credibility of science and reduce it to nothing more than “social construction.” Reflecting on and responding to this situation, Latour articulates a role for a different kind of critique and critic. Drawing on Heidegger’s notion of things as gatherings, and also in the old political sense of the Thing as parliament, he suggests that “things have become Things again, objects have reentered the arena, the Thing, in which they have to be gathered first in order to exist later as what stands apart” (Latour 2004, 236). Rather than dogmatic matters of fact, he argues for a renewed attention to matters of concern (Latour 2004) and thing-centered political assemblies for making matters of concern public (Latour 2005). This notion of things as gatherers and representations of matters of concern has also been picked up in design as a way to describe and envision its political role in enabling democratic participation and the actual doing of politics (Binder et al. 2015; DiSalvo 2012). ANT can provide a useful methodological orientation and sensitivity for our enterprise here. It points to the gatherings that enable and are manifested in things that are political matters of concern—and if this the case for all kinds of things, it is especially so for fluid assemblages. For example, we might try it out with Uber, following the various actors that are involved in holding this particular assemblage together. An initial roundup might include the company and its employees, drivers, passengers, apps for drivers and passengers, regulatory frameworks, advertisements for Uber, journalistic coverage, lawmakers and regulators, citizens, Google maps that includes Uber as a transportation option, and so on.

Techniques A recognition of these political dimensions of technologies is hardly new. Many who can be seen as early theorists of technology were highly critical of it, seeing it as a force that impacted society and even humanity in deleterious ways. Moreover, they typically saw it as an overpowering monolith, Technology (with

A Conceptual Toolkit

117

a capital T) rather than specific technologies. Perhaps the classic expression in this vein was that of the late Heidegger. He challenged what he called the “anthropological view” that has an instrumental conception of technology as something only used as a means in human activity. Rather, he asserted: The manufacture and utilisation of equipment, tools, and machines, the manufactured and used things themselves, and the needs and ends that they serve, all belong to what technology is. The whole complex of these contrivances is technology. (Heidegger 1977, 4–5)

Heidegger provocatively proclaimed, moreover, that the essence of technology is not technological, but rather a particular mode of revealing of the world—enframing—such that it comes to be seen only as a standing reserve of resources to be used. He uses the example of a hydroelectric plant built on the Rhine that turns the river into merely a source of energy, which he sees as vastly different from the old wooden bridge that previously connected the two banks (Heidegger 1977). Heidegger’s nostalgic tendencies are on full display in this example, which can be (and has been) easily critiqued in its particulars. Yet his forceful portrayal of technology as something more than technological remains compelling, and particularly useful in setting us out on a path toward not-only-instrumental considerations of technologies. At a more down-to-earth level, it is also understandable to some extent how those living in the wake of the Industrial Revolution and the rapid industrialization of the twentieth century might regard technological development with some degree of suspicion and trepidation, given the massive reordering of society that it entailed and that was due in large part to new technologies and associated forms of production. However, even if it is plain, particularly after the “empirical turn” in philosophy of technology (Crease 2001; Brey 2010), that the details of particular technologies and contexts are highly significant, there are still insights we can glean from these more general, and generally pessimistic, critiques. One such concept we can use as a marker for some of these issues, and also a pointer to more concrete matters to investigate, is technique. The sociologist Jacques Ellul, writing in the middle of the twentieth century, defined technique as “the totality of methods rationally arrived at and having absolute efficiency (for a given stage of development) in every field of human

118

Changing Things: The Future of Objects in a Digital World

activity” (Ellul 1964, p. xxv). Discussing specifically the machine as the most striking manifestation of technology of the time, Ellul pointed to technique as that which “integrates the machine into society” (Ellul 1964, 5) and in so doing imposes the efficiency and rationality of the machine in all domains of human life. In other words, a technology is not just the physical object itself, but also all of the associated procedures and social arrangements that are involved in its functioning. Writing a few decades earlier, Lewis Mumford described the “cultural preparation” that precedes actual technological developments and makes society amenable to the techniques of a particular machine (Mumford 2010). He points to the clock as the key machine of the industrial age, and monasteries as sites where the necessary cultural preparation for it occurred. For the monasteries “helped to give human enterprise the regular collective beat and rhythm of the machine; for the clock is not merely a means of keeping track of the hours, but of synchronizing the actions of men” (Mumford 2010, 13–14). He suggests that the Church’s “contempt for the body” (Mumford 2010, 35) and attempt to mortify and subdue it was another key factor in preparing society to submit to subjugation by machines. Perhaps the most thoroughly developed and incisive critique of modern industrial society was written by Herbert Marcuse in his magnum opus One Dimensional Man (1964). Working in the tradition of the Frankfurt school of social theory, he integrated philosophy, social theory, and politics in order to develop a critical account of tendencies in highly developed societies. He diagnosed modern society and thought as “one-dimensional” due to the allencompassing technical rationality that both served and produced man’s needs while at the same time incorporating or rendering impossible any meaningful critique. Rather than being rational, Marcuse argued that the technical apparatus of production and the associated ordering of the social world was actually at bottom highly irrational; and that rather than leading to individual liberty, it effectively enslaved people as instruments in the machine. Although technical rationality produces a relatively high standard of living (for some), it comes at the price of a loss of liberty, autonomy, and possibility to realize the highest human possibilities. Moreover, this system precludes the possibility of any meaningful alternative, such that even apparent oppositions only serve to reinforce the existing arrangement.

A Conceptual Toolkit

119

For Marcuse, the only hope (and it is a small one) comes through making the “Great Refusal” of the current order of things. Then, perhaps, other alternatives might be imagined. The essentialist, substantive theories of technology of which perhaps especially Ellul and the late Heidegger are typical are unabating in their pessimism, the heart of which is the conception of technology (in the broad sense including techniques and so on) where the technical means are inextricably linked to ends and particular (undesirable) forms of life. The only hope, in this view, is for some kind of awakening and revealing of another ground from which to imagine other possibilities, perhaps coming from art. However, this portrayal of the almost total lack of human agency is not adequate for envisioning a program of technological reformation. This is the diagnosis and project of philosopher of technology, and student of Marcuse, Andrew Feenberg, who places agency front and center in his Critical Theory of Technology (Feenberg 1999, 2002). Feenberg follows earlier critical theorists in pointing to the prevalence of technocratic forms of administration that are legitimated through reference to scientific expertise, but also rejects essentialism in his conceptualization of the ambivalence of technologies—the possibilities for them to be developed and used in a variety of different ways and with different social consequences. His concern is the “subordinate position” that humans typically have in relation to the “technical systems that enroll us,” and he argues that we need to “begin to intervene in the design process in the defence of the conditions of a meaningful life and a livable environment” (Feenberg 1999, p. xiv). This process of democratic participation in the design of technical systems he refers to as democratic rationalization, which involves people who are “engaged in technically mediated activities and able to actualize ambivalent potentialities suppressed by the prevailing technological rationality” (Feenberg 1999, 105). By this means technical systems could be grasped and changed from within by those who understand and are affected by them, and thus able to envision different possibilities. While the techniques of the early and mid-twentieth century might be most readily associated with large industrial machines, these are not the most prominent currently. Rather, we might think of large data-driven organizational and administrative systems and even social media platforms

120

Changing Things: The Future of Objects in a Digital World

that provide the frames within which much work and human interactions of all kinds take place. As one particular example, we might think of the rise of what has become known as the “gig economy,” where digital platforms enable efficient matching between tasks and humans available to do them. These platforms make human labor available outside of the confines of stable and regulated employment relations, with predictable consequences for worker rights and economic security. One dynamic that has become characteristic of these arrangements are information asymmetries between those who own these kinds of platforms and those who do the work. This was highlighted in the case of Amazon’s Mechanical Turk by the Turkopticon project, which allowed workers who perform human computation tasks in the system to share information with each other regarding their experiences with particular employers (Irani and Silberman 2013). Another example is Uber, which leverages intentionally designed information asymmetries to extract a profit from the difference between the routes and “upfront” fare prices it shows passengers and the most efficient routes shown to the drivers that are used to calculate their pay (Kravets 2017). More generally, our data-driven economy is now fueled by digital labor that is only beginning to be conceptualized as such (The Economist 2017a). The dominant techniques of our age are arguably those involving the collection, processing, and leveraging of massive quantities of data, the creation of which is mediated in large part by fluid assemblages manifested as things available for use.

Political economy Another trajectory that we can follow in approaching especially the political and economic aspects of our current sociotechnical landscape comes from media and cultural studies. Of course there is very much more involved in terms of the role of media and culture in society, but it is the political economy of media that seems particularly salient in relation to fluid assemblages. As suggested by the term, political economy refers to both political and economic dynamics at play in the production and distribution of media, as well as consumption and use. And “at the heart of studies of political economy

A Conceptual Toolkit

121

lies the question of how social resources are controlled and by whom” (Kellner and Durham 2012, 15). What is especially significant in the context of fluid assemblages is the fact that the core product of media industries is not the media content itself, but rather the audience for that content that is then sold to advertisers as a commodity (Smythe 2012). In the US context at least, the ratings industry that developed to report on the number of viewers of network television shows mediated the production and market for the commodity audience. This market also came to define “the audience” (meaning the one attractive to advertisers) as white men aged 18 to 34, reflecting entrenched (but inaccurate) stereotypes about economic and societal power (Meehan 2012). This structure thus determined the types of content that were prioritized. The contemporary media landscape in which the audience commodity is produced is much more diverse than television, and its development has been characterized by processes of convergence among “old” and “new” media forms and increasing participation by people making up the group formerly known as the audience (Jenkins 2006). However, the underlying model has remained remarkably consistent; what has changed are the increasingly fine-tuned mechanisms and diversity of channels through which attention is aggressively bought and sold by what legal scholar Tim Wu calls the “attention merchants” (Wu 2016). The contemporary media landscape involves advertising platforms that are built on top of search engines (especially Google) and social networks (especially Facebook) and apps. In this sense we are also back in a situation in which a few big players dominate the market and the way it is structured, as well as playing a major role in society in other ways. For example, media scholar Siva Vaidhyanathan (2012) has warned about the consequences of Google’s dominance over the internet and involvement in so many of our personal activities and societal functions. One other thread that is worth picking up on here is the way in which the role and agency of the “audience” or “users” are seen. Scholars associated with the “Frankfurt school” of social research, working during and in the aftermath of the Second World War, saw media and other cultural institutions as mechanisms of mass control that reinforce existing social hierarchies and state authority. One of the strongest statements in this vein was by key figures

122

Changing Things: The Future of Objects in a Digital World

Theodor Adorno and Max Horkheimer, who excoriated the “culture industry” as “mass deception” (Adorno and Horkheimer 1997). In reaction to this view of people as passive receivers of messaging, audience studies emerged as a way to open up for audience agency. Messages were not only “encoded” into media content by producers, but also “decoded” by the audience who could provide their own alternate readings (Hall 2012). It is interesting to note this as one framing of agency: one in which the content remains the same and freedom comes from positioning oneself in relation to it. This seems not entirely dissimilar to the agency of being able to accept or reject terms of service (and thus use) for an application, or choosing which content to post and share within the structures of a social media platform. A key political economy question is: Who controls the means of production and consumption of content? We now can and need to ask new kinds of questions about who controls the means of production in this new media landscape. The “democratization” of media participation means not only that private citizens can generate and share media content, but also that, for example, the Russian government was able to expertly leverage and manipulate social media platforms in a sophisticated attempt to influence the outcome of the 2016 American presidential election.7 The internet itself is also a basic underlying “means of production” at this point, but it too can be reconfigured—as evidenced in persistent attempts to undermine or eliminate the principle of equal access to the internet known as net neutrality. These platforms are also involved in much of the basic functioning of fluid assemblages. In this sense, it is no longer only information, advertising, and cultural texts that are implicated, but also things themselves. The developmental trajectories of media industries and industrial design are thus forming a new and significant point of convergence.

Algorithmic culture Algorithms—the animating logic of computational systems—are becoming ever more significant in shaping societal processes. Although they have until now been considered primarily within computer science contexts and applications, their cultural significance is now hard to overlook. So just how do algorithms participate in the production of culture?

A Conceptual Toolkit

123

One way in which they participate, as argued by media scholar Tarleton Gillespie (2014), is in the production and certification of knowledge, representing a particular “knowledge logic” with particular assumptions about what counts as knowledge and what its most relevant components are. Gillespie calls these public relevance algorithms, and he outlines six dimensions that “have political valence” (Gillespie 2014, 168): 1. Patterns of inclusion: the choices behind what makes it into an index in the first place, what is excluded, and how data is made algorithm ready. 2. Cycles of anticipation: the implications of algorithm providers’ attempts to thoroughly know and predict their users, and how the conclusions they draw can matter. 3. The evaluation of relevance: the criteria by which algorithms determine what is relevant, how those criteria are obscured from us, and how they enact political choices about appropriate and legitimate knowledge. 4. The promise of algorithmic objectivity: the way the technical character of the algorithm is positioned as an assurance of impartiality, and how that claim is maintained in the face of controversy. 5. Entanglement with practice: how users reshape their practices to suit the algorithms they depend on, and how they can turn algorithms into terrains for political context, sometimes even to interrogate the politics of the algorithm itself. 6. The production of calculated publics: how the algorithmic presentation of publics back to themselves shape a public’s sense of itself, and who is best positioned to benefit from that knowledge. Each of these dimensions offer inroads into investigating the cultural consequences of algorithms. Most of them also correspond to design decisions about the ways in which fluid assemblages can be manifested as things available for use, and how these things reveal or conceal the algorithmic functioning of the larger assemblage. However, it is worth remembering that concealment of algorithms in “black boxes” has so far been the norm (Pasquale 2015). Algorithms are also shaping the production of even more traditional cultural forms. One noteworthy example, already discussed earlier, is Netflix’s production of “House of Cards.” It was made based on a prediction of relevance for fans of certain types of content, but also, and significantly, because Netflix

124

Changing Things: The Future of Objects in a Digital World

knew that it would be able to distribute the content precisely to, and packaged precisely for, these audiences. This was enabled by the shift Netflix made from sending out DVDs in the mail to streaming content, which meant that they were able to track user behavior in real time. As Finn puts it: Now Netflix can track precisely how their customers watch particular shows, how long they hesitate between options, and perhaps even how much pausing, fast-forwarding, or rewinding goes on. The instant gratification of streaming creates a different kind of rating relationship—not the evaluation of a film I watched last week or ten years ago, but right now. Netflix is no longer constructing a model of abstract relationships between movies based on ratings, but a model of live user behavior in their various apps. (Finn 2017, 91)

This is what Finn (2017) calls the aesthetics of abstraction: simplifying complex and messy details in order to provide reliable personalized services. At least Netflix lets users in on the ways in which their experience is being customized, using tags like “because you liked ….” At the heart of these new forms of interaction is what Finn calls algorithmic arbitrage, exploiting the gap between computation and culture as Netflix did in order to precisely identify the market gaps it could fill with House of Cards and its other series. But the important thing here is not just the success of this particular show, but the underlying trajectories it signals. As computational systems become more efficient, and the patina of personal data we leave behind us grows thicker, the presence of this arbitrage in our cultural lives is rapidly expanding, and beginning to reinvent what the eternal consumer present, the moment of “now,” actually means. After all, there are billions of dollars changing hands over the question of who gets to construct the present for you. When you access a website, perhaps to find out what is happening in the world “right now,” hundreds of servers are involved in auctions lasting fractions of a second to determine which advertisements will appear on the page, and maybe even organize its content according to models predicting your interest in different topics. (Finn 2017, 110–111)

In the language we have been developing here around fluid assemblages, we could say that Netflix and other things that are made available for use are literally made in a particular moment for a particular user, as defined by a

A Conceptual Toolkit

125

particular constellation of data points. Finn (2017) argues that we need to cultivate algorithmic imagination as a practice of trying to grasp these existing currents and to imagine what kinds of generative human and machine collaboration could be possible. Something very similar could be said for design. What types of creative design practices might be appropriate for working with algorithmic processes as co-creator and co-producer? How might we understand algorithmic form and aesthetics in ways that maximize their best possibilities, while avoiding being swept into service of their worst tendencies and most powerful stakeholders?

Notes 1

The original German word Zeug has been translated into English as both “equipment” and “useful things.” The Stambaugh translation revised by Schmidt (Heidegger 2010) that is used here refers to useful things, but we here work with equipment as the overarching term for referring to the issues Heidegger raises. This is in keeping with how it has been used in relevant discourses, and it also seems to be the most precise and evocative word for our purposes here.

2

https://www.apple.com/iphone-x/, accessed November 17, 2017.

3

It is standard in North America to have a central heating system in which the machine that does the heating is located in a utility room or basement. It blows warm air through air ducts and out of vents into the living space. The desired temperature is set at a small dial or control panel on the wall.

4

For the uninitiated, a helpful and quite accurate primer on the experience of owning Apple products can be found at http://theoatmeal.com/comics/apple.

5

Here can be seen the Heideggerian framing of a thing withdrawing and becoming invisible during use and becoming present and visible on breakdown, discussed earlier in the section on Equipment.

6

The accuracy and validity of this account have been disputed, but since it provides a compelling and feasible case of technology serving a political function it still makes a good example.

7

A collection of relevant coverage can be found, among other places, at https:// nyti.ms/2kbHS5m. Some specific examples can be seen in Frenkel (2017) and White (2017).

8

Assembling an Analytic Playlist Earlier in this book, we took a historical journey through technological developments using the simple act of pressing play to listen to music in order to see what is becoming of things. In what follows we will return to the example of using a “thing” to listen to music. This time, however, we will just look into one particular example of a fluid assemblage: Spotify. Indeed, wordings such as “particular example” are in a way misleading, since what “Spotify” actually refers to is perhaps more the central organizing node in the fluid assemblage that it instantiates. But at the same time, a conversation like this makes perfect sense to us: – Do you use a record or cassette player to listen to music? – No, I use Spotify. And in this way, it actually stands forth as a kind of “thing” to us. Now, let us take a look at how. Already when launching the application it is possible to identify elements that are assembled: the fact that it is an application running on a particular device and operating system, that an internet connection is required in order for it to function, and that it is necessary to log in to an account before it is possible to use it. There is also an option to log in using Facebook account credentials, thus already establishing another connection in the assemblage. The next step after logging in the first time, and periodically thereafter, is to accept (updated) terms of service and privacy policy, which is necessary for and implied by continuing to use the service. This means that one has entered into a standing legal agreement before even listening to music.

128

Changing Things: The Future of Objects in a Digital World

On the first main screen one then sees a number of featured playlists, some labeled as “sponsored” by a particular entity. In addition to the sponsored playlists are ones selected for a particular day and time and activity, for example, commuting home on a Monday evening or relaxing on a Saturday morning. This is followed by sections for top lists, new releases, personalized recommendations, and various genres. The genres include what one might typically think of as music genres, such as pop, rock, metal, soul, and so on, but also others that reference a particular mood or activity, such as sleeping, traveling, dinner, or gaming. There is also a category called “Fresh Finds” that includes playlists with selections “surfaced by Spotify’s tastemakers.” The “Trending” category contains playlists that are currently popular. All of these things combined give a strong impression that Spotify is continually updated, that it is not merely a reliable tool for playing music but also a conduit through which one is presented with fresh content. This content is both actively selected and curated by Spotify and their “tastemakers,” and assembled, presumably algorithmically to at least some extent, on the basis of current popularity. Now it is of course possible to view this in terms of media, to see the Spotify shell as the channel and what is inside of it as the content that is transmitted through that channel. With a traditional media framing, one might typically consider the ways in which media are produced and consumed, and the role of the technology itself in shaping these dynamics of production and consumption. With the advent of new media and associated practices, other possibilities and practices entered the picture, such as the convergence of media forms and industries; rise of amateur and fan production and remixing; and the collaborative creation, curation, and consumption of media (Bolter and Grusin 2000; Jenkins 2006; Löwgren and Reimer 2013). We can certainly see these dynamics in Spotify in, for instance, the ability to create and share playlists, including collaborative playlists. It is also relatively easy for artists to get their own music on Spotify; although it is necessary to use a label or distributor, Spotify has deals with a number of companies that can serve these functions for artists who do not already have a label (https://www.spotifyartists. com/guides/). This also points to the network of actors that is brought together in order to create and maintain the ecosystem that keeps Spotify running. But what about Spotify itself as a thing? What is it that is actually present, that exists in the world in some form, and what does it do? This is where things

Assembling an Analytic Playlist

129

become more complicated, because even though we can see an instance of Spotify running on a particular device, that is clearly not either Spotify “in itself ” or all the components that are necessary for it to work. And just as it is not obvious or easy to see what is actually going on in order to make Spotify appear as a thing available for use, it is also not straightforward to see or understand how the things we do are tracked, made visible, and have consequences in various ways—although if we note things such as currently trending music or personalized recommendations, we can definitely understand that this is going on somehow. To begin somewhere, we may try to understand just the Spotify application itself as it runs on a computer. To try to get down to basics, we can see what happens if the application is started without a network connection, to get a sense for what happens locally and what relies on networked resources and content. What we get is a sidebar “container” with main categories for browsing Spotify content and one’s own music, but the main window contains a message stating: “This view is not available offline! Please go online to load.” Now although this type of message is now quite common and unremarkable, it is actually very interesting when it comes to considering Spotify as a thing. Because the main section is really more than a view: it contains the primary functional components that enable playing music. Even the play, backward and forward buttons that control music playback are dimmed and inactive. A banner message at the top of the window reinforces the necessity of a network connection: “No internet connection detected. Spotify will automatically try to reconnect when it detects an internet connection (error code: 4).” In the bottom right corner of the window is another indicator that it is in offline mode. Offline mode is clearly an abnormal, inoperative state for the Spotify application. Without a network connection, it just does not do much. Local application and networked resources typically work together to produce Spotify as the thing we use, but this hybridity becomes noticeable only when there is a breakdown. It is one crucial connection in the Spotify assemblage. Of course there is also quite a lot going on even if we consider only the Spotify app, which also relies on various resources and processes provided by the underlying operating system. This is perhaps even more difficult to see and conceptualize, for non-programmers in particular. But it can

130

Changing Things: The Future of Objects in a Digital World

become visible when, for instance, the application crashes and one has the opportunity to send a crash report. Viewing the details of one such report, it is possible to see hundreds of lines of diagnostic information that includes local system information and the processes that were running at the time of the crash. There is a lot of information contained in even this first bit of the crash report, ranging from straightforward (such as path to the local Spotify application) to downright bizarre (what is a “EXC_CORPSE_NOTIFY” exception?). We might also note the very long version code that seems to indicate that there have been many versions and frequent updates. So, this one particular crash report one of us generated is specific not just to the Spotify application, but to version 1.0.28.87.g8f9312a4 running on Mac OS X 10.11.6 on a computer that has been awake for 1,200 seconds since boot and 33 seconds since wake. Returning to normal operations and establishing a network connection, the content that is the main focus and ostensible purpose of Spotify comes flooding in and is foregrounded. If we were to follow the content, we would reach networks of music distribution, licensing, curating, promoting, and so on, and of course also the musicians who make the music. We could look at the role of Spotify in reshaping business models in the music industry, and the implications of this shifting landscape for the ways in which music is produced and consumed. But let us now use the tools presented in the previous chapter to take a tour through Spotify as a fluid assemblage.

Equipment for listening When using Spotify as a “tool” for listening to music, it tends to withdraw into background awareness when it is playing music as intended. When something goes wrong—perhaps a loss of network connection, or an application crash— it comes to presence as a malfunctioning thing, and the sources of breakdown must be remedied in order to return to listening. However, as seen in the crash report earlier, figuring out exactly what went wrong is often difficult or impossible. A typical strategy for dealing with both malfunctioning computational devices and applications is to restart them and hope that takes care of the problem somehow. It is both practical tactic

Assembling an Analytic Playlist

131

and quasi-ritual, a way of entreating the processors and bits and bytes to sort themselves out and begin working again.

Engagement? Devices that provide music as a commodity without the focused effort and communal aspects of playing musical instruments receive condemnation within philosopher Albert Borgmann’s device paradigm (Borgmann 1984). But what about all the “social” aspects, such as the fact that you can see live updates about what other people are listening to after connecting to one’s Facebook account, or “scrobble” one’s own activity to a Last.fm account? Indeed, the settings options within the Spotify desktop application seem to be mostly about making connections—to other devices, to one’s Spotify and other social profiles, and to various sources of local music. More technical settings, such as cache location and proxy server settings, are initially hidden within “Advanced” settings. A user is aggressively encouraged to “engage” socially, but not to engage with the workings of Spotify itself. While we might maintain a critical stance toward blanket diagnoses of disengaged music consumption, Borgmann’s description of the tendency for things to hide their workings and provide effortless commodities seems to hold remarkably well. One can use Spotify heavily, but not become a “prodigious” or terribly creative user in the sense of someone playing a musical instrument, or indeed a hip-hop artist re-appropriating the music playback technology of a turntable as a music-making thing itself. And one of the key attractions of Spotify, which characterized moments of wonder at using it for the first time when it first became available, is just how easy it is to instantly play almost any music one would like to hear. “Engagement” is also a common term of art in marketing, as in how Spotify’s brand-facing page boasts that “your audience” is “extremely engaged,” spending an average of 148 minutes listening to music on Spotify across platforms. And further, because the audience is growing and becoming increasingly mobile, brands have “more ways to reach them” through “moments-based marketing.”1 As Spotify describes the potential for leveraging the intimacy of audio:

132

Changing Things: The Future of Objects in a Digital World

Streaming opens up an entirely new set of addressable moments for marketers. The music streaming ad revenue opportunity is worth $1.5 billion today, and it’s expected to reach at least $7 billion by 2030. Audio’s unique ability to flex to consumers’ need states makes it an especially powerful marketing tool. The mobile moments “at work” and “working out” alone have opened up $220M in ad revenue opportunity. Leverage audio to reach your audience when they’re most engaged, with messaging that matches their moment.2

Many mediations Spotify is perhaps most associated with the act of listening to music: it is the technology that mediates access to that content. But as seen earlier, it is also a way for advertisers to reach “consumers” (all music listeners naturally being subsumed in this ontological category). It is also a way for artists to reach an audience, and to get detailed data about how people listen to their music. This includes data about what tracks were streamed when and how many times, how many “followers” the artist has, and more. Users are automatically notified of new content from artists they follow, and are sent concert recommendation emails when the artist comes through their town.3 It is also possible for people to see what their friends are listening to through the social feed. Following good postphenomenological practice, we might look for multistabilities—different kinds of relations that users can achieve to Spotify. And this is certainly possible. But what is also striking is how many more possibilities for varying relations seem to lie with Spotify itself (both Spotify as thing for use and Spotify the company). It actively manages different kinds of relations to categories of listeners, artists, brands, and developers, presenting itself quite differently to each. And in fact, much of this presentation is about different kinds of mediation: Spotify as a way to access music, to access an audience, to access consumers, to access other music fans, and so on. Through its “charts” features and year-end report on global listening behavior, it also provides a window on what others are listening to in different parts of the world.

Assembling an Analytic Playlist

133

An audio technosphere Spotify’s description of “moments-based marketing” for brands might just as well be a description of its role in people’s proximal technosphere: Our lives are shaped by what we hear—and each one of us has a unique soundtrack. Ever since streaming gave us instant, on-demand access to all the world’s music and podcasts, it’s become easier than ever to personalize that soundtrack.4

And Spotify excels at being involved in every moment of life, especially through its mood- and activity-themed playlists: for studying, working out, partying, relaxing, celebrating holidays, being just so done with winter, and many more. Of course, these also serve as mechanisms for allowing advertisers to reach users at precisely the right moments, and in the right mental and emotional states to be receptive to particular messages. Of course, users also have significant agency in determining the role that Spotify will play in their lives. And it is interesting to note the possible (intended) role of music in this context of curated content for particular moments and moods—that one might turn to Spotify not just to play music (which tends to be an emotionally rich and evocative experience anyway), but to feel something, and to access and assess one’s feelings and current situation as part of the process of selecting what to listen to. Addressing the possibilities of future developments, and specifically voiceactivated assistants and devices, for delivering “deeper personalization,” Spotify tells marketers: Streaming already gives us insight into who people are, what they’re doing and how they’re feeling in the moment. As audio innovation grows, that consumer understanding will enable deeper personalization than ever before. The experts we spoke to talked about the potential of “dynamic audio,” or the ability to offer mood-based targeting and creative that can adapt to your real-time context. As devices become more connected, they’ll be able to serve up increasingly relevant content. On Spotify, for example, listeners are 100% logged in with a persistent ID across devices—and since they’re listening all day long, streaming provides deep intelligence about real-time context and emotional state.5

134

Changing Things: The Future of Objects in a Digital World

The text goes on to mention the possibility to “re-engage” listeners, “to keep the conversation going with visual reminders or sequential messages.” On this description of life in the technosphere, one might wonder if something has been lost in the understanding of what a conversation is and can be. And it seems increasingly true that we indeed cannot get out or escape.

Connection mechanisms As mentioned in the quote earlier, it is necessary to be logged in all the time in order to use Spotify. It simply does not exist as a music-playing thing until logging in. This being logged in is thus one of the primary “protocols,” or modes of connection. And the precise modes of connection available vary, with corresponding interfaces, for regular users, artists, developers, and so on. It is indeed the information for the latter that one might expect to be most revealing when it comes to protocols, and Spotify’s developer site6 does not disappoint. Here one can find information about the Spotify Web API, Web API Console, the Spotify Play Button, Spotify iOS SDK, Spotify Android SDK, and ways to integrate Spotify into other products. The play button7 is particularly interesting, especially in light of our investigation of “pressing play” in Chapter 3. It is a widget for playing a specific album, track, artist, or playlist that can be embedded in a website. It both utilizes and connects back to the Spotify service, requiring users to log in or create an account after 30 seconds if not already logged in.

Spotify’s algorithms One of the aspects of algorithms highlighted earlier is how they come with embedded assumptions about the effective computability of the world— assumptions that even if the model is not totally accurate, it at least gestures toward something like universal truth. With that in mind, consider Spotify’s comments on their “Year in Music” analysis, “Wrapped”:

Assembling an Analytic Playlist

135

With billions and billions of data activities taking place on Spotify every single day, we understand people. Every shuffle, share and song tells us a story about them. Now, we can share those stories back to every listener — and to the world.8

And further on the same page: We’re able to build experiences like Wrapped with a little bit of magic … and a lot of data. We call it our streaming intelligence. Since our audience of 140 million listeners streams in moments throughout the day, across devices, we’re able to get a real understanding of who they are, how they stream, and what context they’re in. This streaming intelligence lets us create hyper-personalised experiences for our listeners, like Your Time Capsule, which takes them back to the music they loved in their teenage years. Or Release Radar, a customised weekly playlist featuring new music from their favourite artists.

Spotify asserts that they “understand people” on the basis of their activities within the platform; that their activities on the platform tell “stories” about them; that they, Spotify, “build experiences”; and that they get a “real understanding” of who users are and what “context” they are in (context presumably defined on the basis of their device type, music selection, and so on). These are some rather audacious claims, all founded on a belief in the power of lots of data and computation to reveal reality. One might question the ways in which “people,” “stories,” “experiences,” “understanding,” “context,” “engagement,” and so on are operationally defined in texts such as this, and the algorithms they describe; but then again, how can one challenge magic and data? One way to begin to make sense of a fluid assemblage like this is to ask the questions outlined in Chapter 7 on the basis of Wiener’s description of cybernetic organisms: (1) what information is it equipped to collect; (2) in what ways does it transform the information it collects; (3) how and where is the information fed back into ongoing operations; and (4) what effect does this have in other parts of the assemblage and its larger environment? We will not here go through each of these in detail, although even the short quotes here provide a good start in finding answers that help illuminate the character and functioning of the Spotify assemblage.

136

Changing Things: The Future of Objects in a Digital World

Data and information (and understanding?) As is made even more obvious in earlier quotes, Spotify both produces and utilizes lots and lots of data about activities on its platform. It is on the basis of this data that it can make claims to advertisers about how well it knows and understands its users. But we can also interrogate just what these claims are founded on, working in the way of Latour (1999) following the data presented in a table in a journal article back to the soil samples in a lab back to the box in which they were sorted back to the location from which they were extracted in the Amazon. Take the metric of “engagement,” for example. This is defined as the number of minutes per day that people spend playing content on the Spotify platform across devices. All good categories cull away detail in order to produce uniform data that can be compared, and this is no exception. After all, how could each of these minutes of played content reflect equal levels of engagement? Surely some moments reflect more attentiveness than others, and some probably reflect a device playing to an empty room or to headphones that have been removed from ears (although perhaps headphones in the future will be able to detect proximity to ears, and thus more reliably deliver advertising content?). The messy details of everyday music listening practices are smoothed over in a system in which every minute that content is played is considered equal. This is not simply good or bad—again, data that has been generated and rendered comparable by categories in a standard system can be extremely useful. However, when consuming and utilizing data, it is important to keep in mind that it is never raw, and to be curious and critical about the methods that have been used to prepare it. We can also notice as users how action possibilities look like data waiting to be created. For example, when seeing a playlist with a follower count prominently displayed, one can assume that following this playlist will cause that number to increment by one. Playing a certain kind of playlist will also affect the kind of advertising someone on a free plan receives. Playing tracks, following artists, creating playlists, and everything else one can do within the Spotify application similarly generates data. The quite precise data-fueled view that Spotify has on user activities is brought into sharp relief in their comments on their “Year in Music” review for 2017, which they phrase as a list of 2018 goals:

Assembling an Analytic Playlist

137

Like as many things as the person who streamed “That’s What I Like” over 4,653 times this year. Sweat more than the 27,914 people who put “Despacito” on their running playlists. Have fewer issues than the person who streamed “Issues” over 3,152 times this year.9

To this list, perhaps one might add: understand myself better than Spotify’s dataveillance systems do.

138

Changing Things: The Future of Objects in a Digital World

The Spotify ecology As already mentioned, Spotify works across a number of devices, including desktop computers, smartphones, and other devices such as home speakers. It also connects to other services, such as Facebook and Last.fm. Options to purchase merchandise within Spotify link to third-party vendors, artist information is provided by Rovi, and of course ads come from marketers for brands. And surely more components and actors in the Spotify ecology could be added to this initial list. What is particularly interesting, though, is Spotify’s cookie policy. 10 Now, while this type of document might not usually make a short list for exciting reading, it is quite revealing in terms of identifying different components of the Spotify ecology. There are several different cookie types listed: those for essential operation, performance/analytics, functionality, targeting/advertising, third party, and Spotify ads. The third-party cookies even collect information about activity across different websites. And the description of Spotify ads identifies several other components and actors: We work with web publishers, advertising networks, and service providers to deliver Spotify ads on other web sites and services. Cookies may be used to serve you with advertisements that may be relevant to you and your interests on other web sites and services and to regulate the advertisements you receive and measure their effectiveness.

In the following section on managing cookies and other preferences, the same page also states: There is no accepted standard on how to respond to Do Not Track signals, and we do not respond to such signals. We use the AdChoices Icon on our website, and it may also appear on certain targeted advertisements that we (or a service provider acting on our behalf) send to you based on information about your online activities. You may click on the AdChoices icon or visit aboutads.info to receive more information about the collection and use of information about your online activities for online behavioral advertising or to learn how to opt out of having your data used for online behavioral advertising by Digital Advertising Alliance (DAA) participating companies. Canadian users can also visit

Assembling an Analytic Playlist

139

youradchoices.ca. European users can also visit youronlinechoices.com to learn how to opt out of having their data used for online behavioral advertising by European Interactive Digital Advertising Alliance (EDAA) member companies.

First, it is interesting to note the explicit position taken in relation to Do Not Track signals, which is one mechanism for configuring relations online: specifically, Spotify does not honor this kind of request. Second, there is a striking array of different entities for different parts of the world that allow one to opt out of behavioral advertising—and even these can represent an assemblage of “member companies.” Indeed, this kind of analysis of all the components that are somehow connected to the Spotify ecology could continue and become much more extensive; but hopefully this brief look is enough to indicate some potential directions that could be used for further investigation.

Spotify as infrastructural platform Since it is now so easy to use Spotify or other similar services to instantly stream music, it is now almost difficult to remember the time when it was necessary to go to the store to purchase a physical CD or tape or record. But this is what made the capabilities that the internet brought so remarkable: that it became possible to make copies of digital music, without loss of quality, and easily share them with anyone via the internet. Of course, these behaviors were dubbed “piracy” by the music industry, which launched into full-on assault against them. Yet the internet had already shown its enormous promise for being a new kind of infrastructure for a great many things, music distribution and consumption very much included. Spotify entered the scene as a way to leverage this infrastructure and the practices that had grown up around it, but to also bring the “pirates” back within a legal framework—paying for the service either with their attention on the adsupported free version or with a monthly fee for the premium version. Artists were encouraged to not care so much about exactly how much they made through making their music available on the service, since it was for the greater good of establishing this model as a viable framework for music distribution in

140

Changing Things: The Future of Objects in a Digital World

the internet era. In other words, the Spotify platform has been set up from the beginning to be the infrastructure for distributing and listening to music. Spotify and similar services continue to be the focus of strong criticism on the basis of how little artists actually make through using them. In Spotify’s artist guide,11 it is truly remarkable that virtually no information is given about how much artists are paid—this seems to now be left entirely to music distributors. Rather, the emphasis is on the ability to find out about listeners, and how well an artist’s music is performing. The number of monthly listeners is prominently displayed for each artist in Spotify, follower counts are shown for playlists, number of listens for tracks—reflecting the ways in which Spotify platform participation is encoded, data about user activities aggregated, and that data further computationally processed to lead to higher-level descriptions of what goes on in the platform (the year in review feature being a good example).

The politics of Spotify How is Spotify involved in distributions of power and authority? As mentioned in the previous section, it emerged in response to the music industry losing control in the face of peer-to-peer online music sharing. Their goal was to move users back within a legal framework—in other words, it was explicitly political, setting up a framework for how things should be done in contradistinction to another model. We might also look for the various actors that are gathered together around the both material and political thing that is Spotify. As mentioned throughout the previous sections, there are many. While we are not able here to go through a full analysis, it is perhaps at least worth noting that what might on the surface seem to be two of the most fundamental kinds of human actors—artists and listeners— often seem to be themselves used, rather than, or in addition to, being served by it.

Spot the techniques Spotify quite clearly utilizes the data-driven techniques that increasingly characterize administrative, economic, and sociocultural processes. It produces vast quantities of data about activities on its platform that it feeds into its own

Assembling an Analytic Playlist

141

operations and also provides as part of the service. This includes precisely tailoring how Spotify shows up for specific users, ideally enabling a better experience for users who see things they are interested in while also enabling precise targeting of advertisements. We might also consider what have become techniques around listening to music, such as creating playlists that include music from different sources, and listening to music as a service rather than owning a physical copy. Within fluid assemblages such as Spotify, these techniques also become forms of digital labor that provide value for others in the assemblage.

Spotify’s political economy We have already seen how one of the main functions of Spotify is to produce an audience that can be sold to advertisers in quite sophisticated ways. Other questions we can ask in relation to Spotify’s political economy concern who controls the means of production and consumption. And here we can remember that Spotify has positioned itself as a means for controlling the distribution and consumption of music, also providing much more detailed data about consumption than was possible in previous models that relied on physical media and radio distribution. It also opened up a way for artists not signed to a major label to distribute their music—although it is noteworthy that an artist needs to use a label, distributor, or aggregator to deliver their music to Spotify, in order to “make sure everything on Spotify is properly licensed.”12 This is different from a service such as SoundCloud, which allows artists to upload tracks directly, without these intervening layers of actors and mechanisms of control.13

Algorithmic culture and Spotify Spotify is driven by algorithms that produce and process enormous quantities of data. It does this as a matter of course during normal use, but also as part of A/B testing of different versions at massive scale that is used to optimize development of the platform for certain metrics. Algorithmic processes are used as co-creator and co-producer, leading to new types of content (such as

Changing Things: The Future of Objects in a Digital World

142

branded playlists and new kinds of advertising formats14) and new kinds of possibilities to discover and play music and see what others around the world are listening to. Spotify is not just another media format and associated device that makes it possible to press play—it is a fluid assemblage that incorporates many different actors and entities in its data-fueled functioning and continuous development.

An open playlist The notion of partial perspectives has probably never been more appropriate than in the case of fluid assemblages. For instance, when reading our story earlier it is crucial to remember that we here do not speak from the privileged position of actually knowing how Spotify works. We can only articulate what we can access, and here we only see what we are meant to see. Thus, this is just one piece of the overall story: the one from the perspective of those who are using it. Another, and equally important, part of the story is completely missing: the story told from the perspective of those who are making it. But then again, this is unfortunately quite often the case when it comes to these things. And so we here present our story not as a final complete album, but as an open playlist—with the invitation for others to add to it.

Notes 1

https://spotifyforbrands.com/uk/, accessed December 19, 2017.

2

https://spotifyforbrands.com/uk/feature/power-of-audio/, accessed December 19, 2017.

3

https://artists.spotify.com/blog/nerd-out-on-your-data, accessed December 19, 2017.

4

https://spotifyforbrands.com/uk/feature/power-of-audio/chapter-1, accessed December 19, 2017.

5

https://spotifyforbrands.com/uk/feature/power-of-audio/chapter-3, accessed December 19, 2017.

6

https://developer.spotify.com/, accessed December 19, 2017.

Assembling an Analytic Playlist 7

143

https://developer.spotify.com/technologies/widgets/spotify-play-button/, accessed December 19, 2017.

8

https://spotifyforbrands.com/uk/insight/2017-wrapped-data/, accessed December 19, 2017.

9

https://spotifyforbrands.com/uk/insight/2017-wrapped-data/, accessed December 19, 2017.

10 https://www.spotify.com/uk/legal/privacy-policy/#s13, accessed December 19, 2017. 11 https://artists.spotify.com/guide, accessed December 20, 2017 12 https://artists.spotify.com/faq/music#how-do-i-get-my-music-on-spotify, accessed December 20, 2017. 13 https://soundcloud.com/upload, accessed December 20, 2017. 14 https://spotifyforbrands.com/uk/gallery/, accessed December 20, 2017.

9

Making Concepts When we address these new technologies as “things,” we do not mean that fluid assemblages after all are still just things. They are not. Whereas the traditional thing has a surface, these assemblages have nested structures of interfaces. Whereas the traditional thing is stable, these are fluid and their properties immanent. And whereas typical things sport a rather sharp distinction between making and using them, between producing and consuming, to use a fluid assemblage is to become part of something created in runtime. Indeed, the differences are so many and so substantial that it is at first difficult to see why one would think of these assemblages as things in the first place. As we turn to how they present themselves to us, however, and how we come to incorporate them into our lives, the resemblance becomes more obvious. The original meaning of the word “thing” (“þing” in old Norse, “Ding” in German, “ting” in Swedish) is “assembly,” a gathering of the governing people to settle matters of shared concern. A related heritage from the Vikings is how the Swedish word for a thing (an object) is “sak” (“Sache” in German), now present in English as “sake” (as end, purpose), thus also having this double meaning related to both object and issue (cf., also “res” in Latin). Things are always complex entanglements of the social and the material, but we seem to have entered a new phase in the development of “thinging” and “thingness” where networked computational technologies make entirely new kinds of assemblies and assemblages possible. And while they may not be typical things, they are certainly part of a long history of stuff standing forth as “things” to us. Indeed, it is precisely this “standing forth” that is the basic reason for approaching the examples discussed here as things. To be a “thing” is not a physical property, but a state of, or coming into, being. Reasoning around what makes a thing a thing, Heidegger remarks:

146

Changing Things: The Future of Objects in a Digital World

The jug is not a vessel because it was made; rather, the jug had to be made because it is this holding vessel. The making, it is true, lets the jug come into its own. But that which in the jug’s nature is its own is never brought about by its making. (Heidegger 1971, 166)1

And further: But what the vessel of this aspect is as this jug, what and how the jug is as this jug-thing, something we can never learn—let alone think properly— by looking at the outward appearance, the idea. … Instead of “object”— as that which stands before, over against, opposite us—we use the more precise expression “what stands forth.” In the full nature of what stands forth, a twofold standing prevails. First, standing forth has the sense of stemming from somewhere, whether this be a process of self-making or of being made by another. Secondly, standing forth has the sense of the made thing’s standing forth into the unconcealedness of what is already present. (Heidegger 1971, 166)

Here, we have tried to understand and articulate this twofold using a range of lenses and concepts. With respect to the first, the stemming from, we have primarily used the notion of assemblage to explore how something is being gathered, or made, to stand forth as thing. Whereas the jug in Heidegger’s example has a distinct maker, and a kind of stable presence as jug-Thing once made, the fluid assemblages are constantly in the making. In a sense, their presence as thing is immanent, but they are nevertheless able to stand forth as “things” when they, literally, “do their thing.” With respect to the second, to stand forth into unconcealedness, we have discussed how assemblages come to present themselves through interfaces. But we have also tried to show that there are significant limitations to what is actually made to “stand forth into unconcealedness.” And so, when we get to the question of whether the fluid assemblages are really things or not we face both aesthetical and ethical design dilemmas: aesthetical ones because of how current aesthetics attuned to seamless “wholes,” or totalities, conceal much of what is actually going on; ethical ones because we may doubt that these things are “genuine” in any reasonable sense of that term. To Heidegger, “Only what conjoins itself out of world becomes a thing” (Heidegger 1971, 180). Certainly, these things come into being by means of

Making Concepts

147

a gathering, a bringing together that moves above and beyond what it came from. At the same time, their forms of disclosure are at times deceptive. They may, or may not, genuinely bring about nearness as they are equally about the annihilation of distance. And with respect to gathering, to assembling, they are definitely assemblages, not totalities. Because of the exteriority relations, all their components retain their integrity, and the assemblage makes full use of this fact—not the least in the way we as users do not really perceive that we are now part of a large machine as we keep all of our individual properties. What we see here is, therefore, a related but still different form of “thinging.” While still a way of standing forth as a “thing” to us, the condition of being a thing has largely been transformed by this process of constant becoming. This new form of becoming lacks much of the continuity of earlier forms, and what may appear as near, as genuinely close to us as beings, might as well be just the absence of distance. We do not use these things, we become part of their coming into being. This thinging is, therefore, mirrored by processes of objectification. In what follows, we present a set of concepts formed to articulate such aspects. They are by no means an exhaustive list of what new notions we might need to account for the thingness of fluid assemblages. Rather, they are meant as a kind of design examples of what we think could be created as part of an emerging conversation about these things: a set of hopefully inspirational prototypes of concepts we might add to the long list of notions discussed earlier in this book.

Present-as-particular In design, as well as when it comes to experience, we are not only interested in what things in general are. Rather, what we really care about are particulars. When designing, one does not make a chair in general—one makes a chair. When sitting down in a chair, one cannot sit on a chair in general—one sits on a chair. And when you use a mobile phone, you do not use a phone in general—you use that particular phone. It is not technology in general that makes itself present to us, it is some particular expression and instance of it. In fact, we cannot even produce a chair in general, nor sit on one, should we want to. Plato suffered substantially from this insight.

148

Changing Things: The Future of Objects in a Digital World

While it is true for all things artificial, as well as most other kinds and categories, that we get to know them through encounters with particulars, the fluid assemblages discussed here take aspects of particularity much further. Whereas the mass-produced object may become unique through imperfections of the production process, through wear and tear, and so on and so forth, fluid assemblages may receive their particular expression and presence only at the moment of use. Rather than being “things,” a more accurate description of what happens is that they make themselves present as things as we use them. To understand this constant becoming, we need to find complementary ways of discussing more precisely how these things make active use of what becomes part of our experience and frame of reference, and what does not. For a fluid assemblage to work in this way, the mutual presence of both parts is critical: that we, on one hand, honestly think we are using the mobile phone that we hold in our hand, but on the other hand somehow also know that this phone without its connectivity and constant use of resources elsewhere in the world would not at all be the phone as we know it. The expression “bricked,” used to describe an utterly non-responsive state of a computational device, was coined for a reason—but importantly, we use it to describe the state of the thing, the device as such, not the system or structure it is part of or the relation between these parts. Now, of course we can withdraw from each experience and instead try to seek that which is invariant across all these encounters and from these draw insights into what they in general might be. But we can also do the opposite, and instead of analyzing invariance of instances ask the question of how such technologies can be made present to a user, how the particular instance of a technology comes about—how they make themselves present-as-particulars. Already in the basic question “what is this thing?,” we can see a trace of how something becomes present to us as a “thing.” In this question, the “this” reveals several aspects of what is going on. First, “this” positions the thing as a particular, as something that can be distinguished from a “that.” Second, “this” connects the expression to the act of expressing it: while we may not exactly point our finger toward the thing, “this” assumes the presence of some concurrent act of pointing out what particular it is we orient toward. Indeed, if we do not clearly “see” this, our first response is likely to be “which one do you mean?”. This act of pointing out what “this” refers to, in turn, tells us that

Making Concepts

149

to be a thing, it needs to be distinct in terms of place and time. Without either place or time, we cannot differentiate between this and that thing and we lose track of this particular. Consider for instance what happens when we encounter a number of massproduced and completely identical copies of something. Here, “this” may still refer to a particular, to an individual thing. Although we may question the relevance of the remark, we still understand what a person—such as a child choosing between many seemingly identical plush toys—means when saying “no, I don’t want that one, I want this one.” Had our reference “this” been to the essence or general character of the thing, this would not have worked. But, as it refers to “here and now” (or “here and then,” or “there and now,” etc.) it does not at all depend on essence or character, only a specific place and time. In other words, the way that the thing presents itself to us as a particular is largely based on the possibility of differentiation: that we can separate it from the background or from other instances. That our distinctions do not depend on essence but rely on difference opens up enormous possibilities when it comes to bringing something new into someone’s life: a person does not need to understand what the thing actually is at all in order to be able to still understand that it is a “thing” and allow it to take presence as such. Had our understanding of what makes a thing a thing been based on grasping also what it does, this would not have been possible. From this we may learn that what makes something a thing to us to a significant extent depends on how it presents itself to us: to be a thing, it must be able to be present for us in certain ways. And so, to unpack how we come to understand these massive technological systems composed of distributed components communicating with each other as “things” present to us here and now, we perhaps should not ask what they in general are or how they in general work, but rather ask how they make themselves presentas-particulars. This is why it is so important to turn to actual examples of how these technologies come to appear, or stand forth, to us, and not just toward how they in principle function, which is otherwise a typical way of approaching technology. Unlike more traditional technologies now since long part of the philosophical canon, like the hammer or the microscope, these new technologies lack continuity between their presence-as-particulars and what they are capable of doing, most notably because the here-and-now

150

Changing Things: The Future of Objects in a Digital World

“this” that we encounter not even remotely represents (or, “makes present”) what they actually entail. As technologies, they are not, in any sense of word, unconcealed through the interfaces we encounter. Indeed, they are particular about what they make present in use.

Making Concepts

151

Multiinstability One of the key features of a post phenomenological account is its attention to the multistability of artifacts and the many variations that are possible in use; that is, any given artifact can be perceived and appropriated in a vast number of ways. Recognition and analysis of these multistabilities counter technologically deterministic readings of things that might see them in the more narrow light of intended use cases and dominant narratives around particular kinds of (technological) progress. It thus helps ward off reductive approaches that can risk misunderstanding or not seeing what technologies actually do in the world in various contexts, and it does this by pointing to the rich variety of predispositions and possibilities that creative human beings bring to their engagements with the world. The concept of multistability points to the instabilities surrounding artifacts that are introduced on the side of what might be called use or appropriation— that is to say, on the human side of human–technology relations. It signals the varying relations that can be achieved to any given technology, where the technology does in fact remain more or less given, while it is the human that introduces an element of potentially significant variation. Of course we might recognize that each unique technology will be in some ways different to others of its kind, but we also generally assume a quite high level of stability with respect to the character and constitution of the technology itself as an artifact. Indeed, we can recognize this stability in our common usage of words such as “hammer” or “telescope” or “park bench,” where just the term itself, as a sort of category descriptor, evokes a quite clear picture of a certain kind of physical form and associated possible uses (both typical and potentially more creative). However, if we consider more technologically advanced examples, such as those that might fall under the general categories of “computers” or “smartphones,” we find that significant sources of instability in human– technology relations in many cases actually come from the side of the technologies themselves. Many of the technologies we now live with are capable of not only multistability with respect to the relations into which they enter, but also with respect to their very constitution as things available for use. We might even say that just as variability can be introduced by humans in deciding

152

Changing Things: The Future of Objects in a Digital World

how to relate to things, we now also have variability that is introduced from the side of technologies themselves—what we might call a multiinstability. Consider, for example, a computer. One of the first things you do when starting up a new computer for the first time is to create a user account. Initial basic configuration choices such as language, region, and time zone will be set for that account, and in the case of Apple products one is prompted to create and/or log in to an iCloud account that, among many other things, will sync preferences and data across devices. Even at this point the new computer is no longer a generic thing, but rather one customized for a particular person in a particular place. The customization continues as apps are installed and set up, which now typically also includes more processes of creating and/or signing into accounts and syncing associated preferences and data. This introduces a kind of variability that comes from the side of the “thing” itself. As discussed extensively in the examples, with such dynamically customized things different users will see different things—not only because of the different experiences, predispositions, cultural heritages, and so on that they bring as different humans, but also because they are literally seeing and interacting with different things. A relevant example is the “bttn,” a big red physical button that can be configured to connect to a wide variety of web services such as IFTTT, Facebook, Twitter, SMS, various “smart home” systems, and more (https://bt.tn/). Pushing a bttn might thus have no effect if it has not been configured, or it may initiate a quite complex chain of events including everything from communication to controlling physical devices. This is an object that is almost like a blank slate. With the exact same physical form, it can be configured to be many different things that enable as many different relations. While the action that a user takes with the bttn remains exactly the same (one of pressing), the mediating relations it enables can vary dramatically. If we were to look for multistability on the side of use, we would thus almost entirely miss the multiinstabilities inherent in the bttn. Furthermore, such instability or variation also means that such things have quite distinct “sides.” The bttn has a physical form, but this outward form clearly does not fully reveal what it actually is in any given configuration. Understanding its character requires looking also at its “back end” in order to see the various services and other things to which it connects. Its physical

Making Concepts

153

presence not only does not exhaust its capabilities, it does not even come close to revealing them through its outward physical form. We can see this dynamic in a somewhat different way when we consider something like Google, which has a business model of profiling users in order to sell their attention to advertisers. What ordinary users experience in terms of their interactions and the use value they get from Google are only a small part of what these interactions and the overall system functionality actually entail. And even as a site like Google has information pages that face different types of users (“ordinary” users, advertisers, and businesses), this is not just a case of a company providing different types of offerings to different types of clients: in this case, they are all inextricably linked components of the same system. What any given user sees as search results are the product of a complex interplay of past use of the system, contextual variables like location, dynamic ad markets, and ongoing tuning of the Google algorithms. This is a complexity that is once again hidden behind Google’s iconically simple search page. Interestingly, when the functionality and scope of use change significantly in relation to what is already familiar, we may or may not be aware of such changes. In some cases, such changes are presented and broadcasted as narratives of innovation and technological progression; in others, they are much less explicit and even hidden from use. A prime example can be seen in the case of Google, which, partly thanks to its iconically simple interface, is still primarily perceived as a search engine; however, it equally accurately could be described as a customized information experience used to create a marketplace for selling user attention for advertisement. In light of examples such as the ones discussed in this book where it is obvious that the intentions of the user cannot be said to exhaust the actual working of the thing, but where a range of other activities are taking place simultaneously, we think philosophical accounts countering the kind of technological determinism evident in some of these designs are more important than ever. As discussed earlier, much of this could be approached using other terms than “things,” instead working with notions such as systems, media, etc. But since many of these designs explicitly build on the ways we relate to, and experience, things through use, there is a need to account for these encounters also from the perspective of “thingness.” In our view, the

154

Changing Things: The Future of Objects in a Digital World

use of simple interfaces mimicking aspects of their physical and mechanical predecessors is quickly becoming not primarily a convenient design solution to a complex interaction issue, but increasingly a pervasive strategy necessary to unpack in the light of technological determinism as packaged in current consumption practices. These fluid assemblages also need to be understood by their multiinstabilities—all the ways in which these things change dynamically to configure their relations to particular humans, even as particular humans also configure their relations to the things. Not only that, these fluid assemblages also come into relations with other things. This happens during the course of normal operation and may not even be visible to users. This can lead to shock when we find out, for example, that our mobile phones are constantly reporting our locations and other data about us to third parties (Goodin 2015; Örstadius and Larsson 2015). One of the main implications is that we cannot take things for granted, assuming that they can be easily identified and will just “sit still” while we try to figure out how they come into relations with humans. This is, after all, a basic assumption in much analysis of what is going on at the human side of the equation, evident in notions such as “multistability.” But these things are increasingly having their own kinds of agencies, complexities, and variabilities. There are thus tensions between stabilities and instabilities that play out over time in ways that are not necessarily obvious or straightforward. Treating such things as stable objects only ever reinforces them as “given,” as presented to us ready to be used, whereas we in fact need to shift our attitudes toward them as being “designed,” as being made and thus also open for intervention and change. This, we believe, is a most pressing issue in countering technological determinism: our understanding of design versus use, of making versus appropriating and interpreting, needs to become less of a dichotomy and more of a continuum. To “use” a fluid assemblage is to literally become part of its making. As a consequence, we need theory that crosses this divide, addressing the continuity between making and using by breaking with the still prevalent habit and tradition of considering things as stable things ready for use and analysis. It is as if we have to account for not only being present- or ready-athand, but for becoming-in-hands.

Making Concepts

155

Multiintentionality Whereas the notion of multistability inquires into a certain kind of variability on the side of the human in human–technology relations, it still assumes certain invariability when it comes to perspective. Of course, different people may engage differently with a tool, thus exploring its multistability, but we retain the integrity of these individuals when it comes to defining what the thing is through use. In other words, while the “user” may come up with different ideas and interpretations of what a thing should be, we never really leave the perspective of the user. This is a heritage from Heidegger and the idea that the most primordial encounter with a tool is through use. Fluid assemblages, however, introduce a different complexity into the matter of perspectives as any given system might be quite different things to different users: what appears to someone appears as a map and navigation service is to someone else a tool for mapping people’s movements and how much time they spend in different places. Also, a traditional tool could be transformed into something making such multiple intentions possible. Taking a hammer for instance, one could equip it with sensors and more so that while the carpenter makes use of it as a hammer, a researcher is able to use it as an instrument to study muscular forces and strain during work. But in comparison, the diversity in basic perspective when it comes to these networked computational things is staggering. To exemplify, when entering Facebook the headline reads “Connect with friends and the world around you on Facebook”; when entering Facebook blueprint the headline reads “The tools you need to learn how Facebook can help grow your business.” These differences are not just about something becoming picked up through use in varying ways, this is about an assemblage presenting very different views on what it is, what it does, and why. This is not just a matter of multistability that different users interpret and make use of the thing in different ways. Rather, this is about an assemblage becoming present as fundamentally different “things” framed by different intentions. What we’re looking at here is something we may call “multiintentionality.” While we may recognize this, we are clearly also reluctant to acknowledge just how far the implications of this multiintentionality reaches. Indeed, it is

156

Changing Things: The Future of Objects in a Digital World

precisely because these different perspectives, or interfaces, have been made separate that this multiintentionality can evolve. Had all these possible ways of accessing or using the assemblage been available through the same interface, the situation would have been very different as this would open up as a more or less continuous field of possible interpretations. Their design, however, is based on the opposite strategy: to offer an interface as simple as possible with respect to the identified and intended forms of use in question: to listen to music, to find a restaurant, to play a game, to send a message to your loved one, to buy a piece of clothing, to track user locations, and so on and so forth. This is the primary design program of the “app” genre. Thus, you access the assemblage not through an open question about what it can do for you, but via carefully designed interfaces that offer not only a specific functionality but also a careful framing of the assemblage as a particular thing. If we bring this observation back to the phenomenological notion of bracketing, and the method of eidetic reduction, we get another view on the significant difference between surface and interface. When Edmund Husserl formed a method to seek the essence of things as they emerge through experience, he looked for something that would transcend the particularities of each individual experience and allow us to grasp what is invariant. At the risk of overly reducing these extensive philosophical investigations, there are a couple of moves that stand out as crucial in trying to understand the “essence” of something in this way. First, he made the seemingly paradoxical claim that in order to “turn to the things themselves” we need to bracket the “natural” world (epoché). This means that we turn to how the things present themselves to our consciousness, but that we at the same time suspend all judgment regarding what may or may not actually be there independent of human experience. The second move is to see what is invariant by actively performing a kind of systematic variation upon these phenomena. To get an idea of how this method, called eidetic variation, works, imagine that you are trying to form an idea of what the essence of “tree” is: rather than just standing at some point looking at a tree, you would walk around it, look at it from different angles and distances. You would then turn to another tree and do the same, and so on and so forth. And throughout this process, your attention is tuned toward what stays invariant across all these experiences of trees; you try to transcend them and form an understanding beyond what is present

Making Concepts

157

in any given experience. Eventually you understand what “tree” is. Eidetic variation, or reduction, is somewhat similar to this only that it operates not (only) on sensory experience, but on how things, or phenomena, appear in consciousness. Obviously, this does not come close to capturing the depth of phenomenological method, but even this brief glimpse of it allows us to reflect upon the significant differences between surface and interface. This method is about tracing the outlines of something assuming that its metaphorical “surface” (as that which is, again metaphorically, presented to us) is continuous—that the different variations presented through consciousness are not only connected but come to express specific kinds of continuity across instances, namely one that is directly related to its essence. The notion of “shape/form” is more metaphorical here, but that there is a basic relation between form and idea is a notion at least as old as Plato, and here it comes to expression as a way of understanding essence through what is invariant. Indeed, the origin of the word “eidos” holds meanings related to both “form” and “shape” as well as being the start of our word “idea.” Here, however, we must also note the importance of intentionality. According to phenomenology, the human consciousness is always directed toward something; it has an inherent directionality. Thus, when we try to grasp the essence of a hammer, the way we approach its variations to seek what is invariant will not only depend on how the thing presents itself in some kind of, on our side, completely passive act of mere registration, but importantly through how we actively relate to it. This is where Heidegger makes his move, emphasizing that it is through use that a thing like a hammer will present its most significant characteristics to us. This is because, unlike the tree, the hammer is primarily a tool, and thus “the less we just stare at the thing called hammer, the more we take hold of it and use it, the more original our relation to it becomes and the more undisguisedly it is encountered as what it is, as a useful thing” (Heidegger 2010, 69). Turning now to the notion of interfaces, precisely this matter of what it means to turn to something transforms. First, because we do not face a surface in a general sense, but an already from the start defined framing of how interior and exterior are functionally related to each other. And second, because this framing is not just a matter of our own intentions as we approach the thing

158

Changing Things: The Future of Objects in a Digital World

as a tool, but importantly also something based on the intentions framing the interface as such. While a surface can be functionally specific, an interface has to be. We can talk about general interfaces, but they are still somehow based on functionality, on a certain connection between interior and exterior. In other words, whereas a surface is open for continuous variation across instances, an interface is not. Compare performing a kind of eidetic variation on a tree versus trying to do the same with a fluid assemblage. Moving around trees, looking at them from different perspectives, the experience of “tree” is continuous in a certain sense. Turning to our fluid assemblage, we can of course make it present itself to us in a number of different ways; we can compare use across different devices, we can use it in different countries or log in as different users with different histories and preferences and register the differences in results and perspective this brings about, and so on. But there are distinct limitations to what variations we can experience, and we know that there is much about this thing that falls outside what we actually experience through this particular interface. This means that the act of “bracketing” starts to perform in a very different way: instead of suspending judgment about the natural world, like we did when we did not engage in, say, the genetic biology of plants when trying to grasp the phenomenological essence of “tree,” what we instead get here is a bracketing of what falls outside what this particular interface gives us access to.

Making Concepts

159

And while this may capture my understanding of what this thing is and does, it will not include aspects I was not meant to see through this interface. To get to a more complete and continuous outline of what a fluid assemblage actually is, we therefore cannot stay with our own intentional directionality but must include also other, and sometimes completely different, trajectories. We must turn to other interfaces, and even try to get to what exists between interfaces. It might even be that no one actually has a complete grasp on the assemblage. We may even question whether there is any essence to be found at all, as these things might not have such. What is more clear, however, is that we now have to relate to some kind of “multiintentionality” where the transcendence we need to aim for more significantly than before needs to embrace diversity, and of acknowledging that there is no one point from which one can form a more “complete” conceptual framing of something like a fluid assemblage. The observation made earlier, that to “use” an assemblage is to become part of it, therefore takes on very profound meanings. While a term such as “consumer data harvesting” may be a description of what an analyst is up to, it also highlights how being a consumer in a very practical way is also to perform a function, to be a component in an assemblage, that among other things generates such data.2 Thus, what seems to us to be a situation where it is our intention and objective that determine what is going on and what is going to happen next, we might at the very same time be acting as a component without really knowing precisely what function we perform.

Tuning formations Use of things that are fluid assemblages entails participating in these larger structures and processes; and designing them means configuring the terms of this participation for the entities involved—humans and things, users and usees. Thus, there are extensive ongoing feedback loops in which things are updated and customized, and data about use is collected in order to inform future customization and, perhaps even more important, to turn our everyday practices into data that feeds the processes of “cybernetic capitalism” (Raley 2013). Dataveillance—“the disciplinary and control practice of monitoring, aggregating, and sorting data”—now not only monitors but also predicts

160

Changing Things: The Future of Objects in a Digital World

and prescribes (Raley 2013, 124). Big data analysis is structured by nontransparent and self-reinforcing mathematical models that can have very real consequences for, among many other things, whether someone is able to get a job, get insurance, or get parole (O’Neil 2016). It is possible to see these intertwined processes of ongoing design, use, and assembling as processes of tuning.3 The forms that things take are tuned, as are the structures of the larger formations of which they are part. The processes of formation themselves are also subject to this tuning; it is perhaps even these processes that are the main site for design intervention, taking precedence over the honing of physical forms that has so far been the dominant mode of industrial design. Fluid assemblages form, operate, and evolve through ongoing processes of tuning within larger formations. These formations typically orient tuning processes toward extracting the most useful data, and thus to tuning human behaviors in order to enable that extraction. It is perhaps worth emphasizing again at this point that fluid assemblages are not strictly good or bad. The fact that a thing that is a fluid assemblage is a thing for use and also a thing in a larger assemblage can be the source of fantastic possibilities as well as the most misdirected uses of big data and systems of dataveillance. The personalization of things can be delightful, even as it can also border on creepy. However, it is also important to recognize the degree to which things have well and truly changed. The contract between design and use that was inherent in industrial systems of production has been replaced by terms of service, licensing agreements, privacy policies, forced updates, and opaque applications. Legal and political frameworks have not even begun to keep pace with these new realities. On the other side, as people who use fluid assemblages, it is not yet clear how we should make sense out of personalization that is not personal at all, or relate to machines that optimize their relations to us on the basis of massive amounts of collected data and machine learning. It is in this context that it seems we need practices of onto-cartography that are geared toward following the tuning formations of fluid assemblages. Here Levi Bryant’s concept of gravity is particularly helpful: gravity “refers to the way in which the structural openness, movement, and becomings of one machine are mediated by another machine” (Bryant 2014, 193). Onto-cartography involves mapping the gravitational relations between machines. In the context of fluid assemblages, we can see that there are certain entities that exert tremendous gravitational force

Making Concepts

161

over others—what Bryant terms “bright objects” (Bryant 2014, 202). Examples could be the giants of the web that are almost impossible for connected systems to avoid, such as Google, Amazon, and Facebook. Government dataveillance programs can also be a significant factor. And then of course we also need to map the relations among the various other entities involved. The purpose of such mapping is in order to understand and intervene in worlds. Bryant calls this a practice of geophilosopy, with three dimensions: ontocartography—producing topographical maps of assemblages; deconstruction— severing relations between machines that sustain particular patterns; and terraformation—adding machines to existing worlds “to create new paths of movement and becoming” (Bryant 2014, 274). The practice of deconstruction aligns with the possibility that Tonkinwise identifies for design to be also about designing away, not only creating the new (Tonkinwise 2014). The practice of terraformation resonates perhaps most strongly though with design as an intentional practice of bringing about change in the world. So far the biggest overall gravitational force that influences the becoming and operation of fluid assemblages is that of cybernetic capitalism. Indeed, capitalism has always been the driving force that fuels industrial design, so in this sense it is nothing new. However, what does seem to be decidedly new is the culmination of the development of market feedback loops being embodied in things themselves, and the ongoing and dynamic relations that they entail. It is optimization of things based on market logic, users being constantly interpellated as consumers and disciplined subjects. In this new landscape of fluid assemblages, design as a primarily formgiving practice is no longer adequate. To design, and to design with, fluid assemblages is to design forms and processes of becoming; to assemble and be assembled; to tune forms and formations.

Aesthetics of immanence When we introduced the notion of fluid assemblages earlier in this book, we argued that current design aesthetics seem to be oriented toward an ambition to craft totalities. Indeed, there was a need for such an aesthetic orientation in light of how industrial production transformed the way things were made,

162

Changing Things: The Future of Objects in a Digital World

and how the sense of a whole made and cared for by the craftsman had to be replaced with something that would bring the various parts and machinemade components together. In many ways, it is the very same orientation that has made our current condition possible, that of providing strong and unitary experiences of something that seems to us to be one thing, when the underlying technology in fact is far from anything like a coherent and consistent whole. As we have tried to show, however, the effectiveness of this approach hides other issues that we also need aesthetics to address as we consider the social contract that design is meant to establish; that is, we both literally and metaphorically can trust the things we use. Indeed, the crafting of such a contract between design and use can be seen as one of the most central implications of design being a service discipline (Nelson and Stolterman 2012). It has been remarked many times in interaction design discourse that the temporality of computational things is central. Computational things have behavior and a complexity unfolding over time that an aesthetics primarily oriented toward visual form does not account for, and that we therefore need aesthetics dealing also with acts, motion, process, etc. But following the argument presented earlier, we need to push this point even further: the difference here is not just one of time and temporality, it is a difference between transcendence and immanence. Or in other words, whereas design expression is currently built around an aesthetics of transcendence because this was the most critical thing to achieve in industrialized mass production, fluid assemblages instead call for an aesthetics of immanence. While incorporating temporality certainly prepares for some of the issues in how fluid assemblages can be designed to present themselves in and through use, we also need to take care of how we actually are able to make sense of, interpret and understand, these things as what they actually are, as assemblages. To achieve this, we cannot rely on an aesthetic fundamentally oriented toward making totalities. Indeed, just as industrial design needed new aesthetics to address the challenges posed by industrial production, we believe we are going to need new aesthetics to deal with fluid assemblages. When it comes to aesthetics, therefore, it is not enough to address the fact that these networked computational things change over time in the sense of changing appearance, added functions, and so on—what we typically call “interaction” and deal with in for instance interaction design. If such temporal

Making Concepts

163

aesthetics still operate under the assumption that the meaningful whole conveyed in experience needs to be a kind of “totality,” then we still fail to address what truly characterize these things, namely that they are assemblages. Here, it is worth recalling Jane Bennett’s introduction, and how “assemblages are living, throbbing confederations that are able to function despite the persistent presence of energies that confound them from within. … Each member and proto-member of the assemblage has a certain vital force, but there is also an effectivity proper to the grouping as such: an agency of the assemblage. And precisely because each member-actant maintains an energetic pulse slightly ‘off ’ from that of the assemblage, an assemblage is never a stolid block but an open-ended collective, a ‘non-totalizable sum.’” (Bennett 2010, 23–24) This inherent instability, and in particular the complex agency of assemblages that in no sense is subject to “user control,” is so central to how these things work that to solve the issue of how fluid assemblages come to expression through use— that is, how our aesthetics help us deal with these matters when designing—by instead trying our best to hide them is in so many ways just the same as trying to make mass-produced objects look like they were handmade. At first this may seem comforting, but as the previous way of making never had to deal with the levels of complexity that then come about, the limits of the approach will eventually cause it to break down. We argue our current situation is very similar: industrial aesthetics still work, but the issues it cannot deal with are amassing. And so, we arrive at the question: What would an aesthetic proper to these new things then be like, and what would an aesthetics of immanence entail? This question we cannot answer here, the reason being that it calls for something else than what has been our current task at hand. This question about aesthetics is not an analytical problem. This is a question for design: a question that calls for extensive experimentation, for new ways of working, probably even for new practices of design.

Concluding remarks We cannot through analysis arrive at an answer to what the design and use of fluid assemblages should be like. Nor can we through philosophical articulations arrive at all the necessary tools we need to live well with these

164

Changing Things: The Future of Objects in a Digital World

technologies. But what we can do is to start to conceptually furnish a place where design can go to work, and where use can be explored as a curious and critical activity. Much like designing needs it spaces such as the studio and the workshop, it needs its conceptual spaces, its tools to think with. But while designing in the workshop makes use of the tools available, there is still a basic difference between using tools and doing design. The situation here is quite similar. We have aimed to present conceptual tools, along with introductions and ideas about how to use them, but the basic distinction remains: also here there is a difference between using conceptual tools and doing design. In this book, we have tried to set up a workshop for a different kind of conceptual work, along with some ideas about what materials, issues, stakeholders, contexts, and concerns to work with in this space. We have also argued that this workshop is not only meant for what we used to call “design,” but importantly also for what we used to refer to as a matter of “use” (and understanding use and its consequences). With fluid assemblages, this is no longer a dichotomy but a spectrum of acts, agencies, intentions, and more. Thus, above all, it is a “workshop” equipped for rethinking the ways things currently are in light of what we hope they will become. In short, it is about creating a place for changing things.

Notes 1

What is here translated as “making” (and in other translations as “producing”) is in the German original hergestellen. Thus, one important aspect that is lost in translation is the intimate connection to the many other words related to stellen that he is using, perhaps most notably Gestell (enframing).

2

This resonates with Hamid Ekbia and Bonnie Nardi’s notion of inverse instrumentality, describing a sociotechnical system that requires humans to perform certain functions as component parts (Ekbia and Nardi 2012).

3

Richard Coyne’s (2010) investigation of the role of media technologies in the tuning of place has helpfully influenced the conception we develop here.

References

Adorno, Theodor, and Max Horkheimer. 1997. Dialectic of Enlightenment. Translated by John Cumming. London and New York: Verso. Akrich, Madeleine. 1991. “The De-Scription of Technical Objects.” In Shaping Technology/Building Society, edited by W Bijker, 205–24. Cambridge, MA: MIT Press. Alaimo, Cristina, and Jannis Kallinikos. 2017. “Computing the Everyday: Social Media as Data Platforms.” The Information Society 33 (4): 175–91. doi:10.1080/019 72243.2017.1318327. Appadurai, Arjun. 1996. Modernity at Large: Cultural Dimensions of Globalization. Minneapolis, MN, and London: University of Minnesota Press. Arnall, Timo. 2014. “Exploring ‘Immaterials’.” International Journal of Design 8 (2): 101–17. Barad, Karen. 2007. Meeting the Universe Halfway: Quantum Physics and the Entanglement of Matter and Meaning. Durham: Duke University Press. Baudrillard, Jean. 2005. The System of Objects. Translated by James Benedict. London and New York: Verso. Bennett, Jane. 2010. Vibrant Matter: A Political Ecology of Things. Durham and London: Duke University Press. Benson-Allott, Caetlin. 2015. Remote Control. New York and London: Bloomsbury. Bill, Max. 2010a. “Beauty from Function and as Function.” In Form, Function, Beauty – Gestalt, 32–41. London: AA Publications. Bill, Max. 2010b. “The Basis and Aim of Aesthetics in the Machine Age.” In Form, Function, Beauty – Gestalt, 60–70. London: AA Publications. Binder, Thomas, Giorgio De Michelis, Pelle Ehn, Giulio Jacucci, Per Linde, and Ina Wagner. 2011. Design Things. Cambridge, MA: MIT Press. Binder, Thomas, Eva Brandt, Pelle Ehn, and Joachim Halse. 2015. “Democratic Design Experiments: Between Parliament and Laboratory.” CoDesign: International Journal of CoCreation in Design and the Arts 11 (3–4): 152–65. doi:10.1080/15710882.2015.1081248.

166

References

Bogost, Ian. 2012. Alien Phenomenology, or, What It’s Like to Be a Thing. Minneapolis, MN, and London: University of Minnesota Press. Bolter, Jay David, and Richard Grusin. 2000. Remediation: Understanding New Media. Cambridge, MA: MIT Press. Borgmann, Albert. 1984. Technology and the Character of Contemporary Life. Chicago: University of Chicago Press. Bowker, Geof. 1993. “How to Be Universal: Some Cybernetic Strategies, 1943–70.” Social Studies of Science 23 (1): 107–27. Bowker, Geoffrey C., and Susan Leigh Star. 1999. Sorting Things Out: Classification and Its Consequences. Cambridge, MA: MIT Press. Boyd, Danah, and Kate Crawford. 2012. “Critical Questions for Big Data.” Information, Communication & Society 15 (5): 662–79. doi:10.1080/136911 8X.2012.678878. Braidotti, Rosi. 2003. “Becoming Woman: Or Sexual Difference Revisited.” Theory, Culture & Society 20 (3): 43–64. Braidotti, Rosi, and Timotheus Vermeulen. 2014. “Borrowed Energy.” August 12. https://frieze.com/article/borrowed-energy/. Brey, Philip. 2010. “Philosophy of Technology after the Empirical Turn.” Techné 14 (1): 36–48. Brown, Bill. 2001. “Thing Theory.” Critical Inquiry 28 (1): 1–22. Bryant, Levi R. 2014. Onto-Cartography: An Ontology of Machines and Media. Edinburgh: Edinburgh University Press. Bryant, Levi R., Nick Srnicek, and Graham Harman, eds. 2011. The Speculative Turn: Continental Materialism and Realism. Melbourne, Australia: re.press. Bucher, Taina. 2012. “Want to Be on the Top? Algorithmic Power and the Threat of Invisibility on Facebook.” New Media & Society 14 (7): 1164–80. doi:10.1177/1461444812440159. Burdon, Mark, and Mark Andrejevic. 2016. “Big Data in the Sensor Society.” In Big Data Is Not a Monolith, edited by Cassidy R. Sugimoto, Hamid R. Ekbia, and Michael Mattioli, 61–75. Cambridge, MA and London: MIT Press. Bush, Vannevar. 1945. “As We May Think.” The Atlantic, July. https://www. theatlantic.com/magazine/archive/1945/07/as-we-may-think/303881/. Carr, David. 2013. “For ‘House of Cards,’ Using Big Data to Guarantee Its Popularity,” February. http://www.nytimes.com/2013/02/25/business/media/for-houseofcards-using-big-data-to-guarantee-its-popularity.html. Castells, Manuel. 2000. The Rise of the Network Society. Oxford: Wiley-Blackwell. doi:10.1002/9781444319514 Ceruzzi, Paul E. 2003. A History of Modern Computing. Second edition. Cambridge, MA and London: MIT Press.

References

167

Cetina, Karin Knorr, Theodore R. Schatzki, and Eike von Savigny. 2005. The Practice Turn in Contemporary Theory. London and New York: Routledge. Coleman, B. 2011. Hello Avatar: Rise of the Networked Generation. Cambridge, MA: MIT Press. Coyne, Richard. 2010. The Tuning of Place: Sociable Spaces and Pervasive Digital Media. Cambridge, MA: MIT Press. Crease, Robert P. 2001. American Philosophy of Technology: The Empirical Turn. Edited by Hans Achterhuis. Bloomington, IN: Indiana University Press. DeLanda, Manuel. 2016. Assemblage Theory. Edinburgh: Edinburgh University Press. Deleuze, Gilles. 1988. Bergsonism. New York: Zone. Deleuze, Gilles, and Félix Guattari. 1987. A Thousand Plateaus: Capitalism and Schizophrenia. Translated by Brian Massumi. Minneapolis, MN, and London: University of Minnesota Press. Deleuze, Gilles, and Félix Guattari. 1994. What Is Philosophy? Translated by Hugh Tomlinson and Graham Burchill. London and New York: Verso. Deuze, Mark. 2014. Media Life. Cambridge, UK, and Malden, MA: John Wiley & Sons. Dilnot, Clive. 2014. “Companion to DRS 2014: Design’s Big Debates.” Edited by Johan Redström, Erik Stolterman, Anna Valtonen, and Heather Wiltse. Umeå: Umeå University. DiSalvo, Carl. 2012. Adversarial Design. Cambridge, MA: MIT Press. Dolphijn, Rick, and Iris Van Der Tuin, eds. 2012. New Materialism : Interviews & Cartographies. Ann Arbor, MI: Open Humanities Press. Dourish, Paul. 2001. Where the Action Is: The Foundations of Embodied Interaction. Cambridge, MA: MIT Press. Dubberly, Hugh. 2017. “Connecting Things.” In Encountering Things, edited by Leslie Atzmon and Prasad Boradkar, 153–65. Broadening Design to Include Systems, Platforms, and Product-Service Ecologies. London: Bloomsbury. Duportail, Judith. 2017. “I Asked Tinder for My Data. It Sent Me 800 Pages of My Deepest, Darkest Secrets.” The Guardian, September 26. https://www. theguardian.com/technology/2017/sep/26/tinder-personal-data-dating-app-messageshacked-sold. Easterling, Keller. 2012. “An Internet of Things.” E-Flux Journal. http://www.e-flux.com/journal/an-internet-of-things/. Ekbia, Hamid, and Bonnie A. Nardi. 2012. “Inverse Instrumentality: How Technologies Objectify Patients and Players.” In Materiality and Organizing: Social Interaction in a Technological World, edited by Paul M. Leonardi, Bonnie A. Nardi, Jannis Kallinikos, 157–76. Oxford: Oxford University Press.

168

References

Ekman, Ulrik, Jay David Bolter, Lily Diaz, Morten Sondergaard, and Maria Engberg, eds. 2015. Ubiquitous Computing, Complexity and Culture. New York and London: Routledge. Ellul, Jacques. 1964. The Technological Society. Translated by John Wilkinson. New York: Vintage Books. Ensmenger, Nathan L. 2012. The Computer Boys Take Over. Cambridge, MA and London: MIT Press. Feenberg, Andrew. 1999. Questioning Technology. London and New York: Routledge. Feenberg, Andrew. 2002. Transforming Technology. New York: Oxford University Press. Finn, Ed. 2017. What Algorithms Want. Cambridge, MA: MIT Press. Floridi, Luciano. 2011. The Philosophy of Information. Oxford: Oxford University Press. Forlizzi, Jodi. 2008. “The Product Ecology: Understanding Social Product Use and Supporting Design Culture.” International Journal of Design 2 (1): 35. Frenkel, Sheera. 2017. “For Russian ‘Trolls,’ Instagram’s Pictures Can Spread Wider Than Words.” The New York Times, December 17. https://nyti.ms/2k8soRk. Fry, Tony, Clive Dilnot, and Susan Stewart. 2015. Design and the Question of History. London and New York: Bloomsbury Academic. Galloway, Alexander R. 2004. Protocol. Cambridge, MA and London: MIT Press. Galloway, Alexander R. 2012. The Interface Effect. Cambridge, UK, and Malden, MA: Polity. Giaccardi, Elisa, Nazli Cila, Chris Speed, and Melissa Caldwell. 2016. “Thing Ethnography.” In Proceedings of the 2016 ACM Conference on Designing Interactive Systems. 377–87. New York: ACM Press. doi:10.1145/2901790.2901905. Gibson, James J. 1979. The Ecological Approach to Visual Perception. Boston, MA: Houghton Mifflin. Gillespie, Tarleton. 2010. “The Politics of ‘Platforms’.” New Media & Society 12 (3): 347–64. doi:10.1177/1461444809342738. Gillespie, Tarleton. 2014. “The Relevance of Algorithms.” In Media Technologies: Essays on Communication, Materiality, and Society, edited by Tarleton Gillespie, Pablo J. Boczkowski, and Kirsten A. Foot, 167–193. Cambridge, MA and London: MIT Press. Gitelman, Lisa. 2006. Always Already New: Media, History, and the Data of Culture. Cambridge, MA and London: MIT Press. Gitelman, Lisa, ed. 2013. “Raw Data” Is an Oxymoron. Cambridge, MA and London: MIT Press. Goffman, Erving. 1959. The Presentation of Self in Everyday Life. New York: Anchor Books. Golumbia, David. 2009. The Cultural Logic of Computation. Cambridge, MA and London: Harvard University Press.

References

169

Goodin, Dan. 2015. “User Data Plundering by Android and iOS Apps Is a Rampant as You Suspected.” Ars Technica, November 4. http://arstechnica.com/ security/2015/11/user-data-plundering-by-android-and-ios-apps-is-as-rampantas-you-suspected/. Greengard, Samuel. 2015. The Internet of Things. Cambridge, MA and London: MIT Press. Gunn, Wendy. 2013. Design Anthropology: Theory and Practice. London and New York: Bloomsbury. Hall, Stuart. 2012. “Encoding/Decoding.” In Media and Cultural Studies, edited by Meenarkshi Gigi Durham and Douglas M. Kellner, 137–44. West Sussex: WileyBlackwell. Harman, Graham. 2002. Tool-Being. Chicago and La Salle, IL: Open Court. Harman, Graham. 2011. The Quadruple Object. Winchester and Washington, DC: Zero Books. Harman, Graham. 2016. Immaterialism. Cambridge, UK, and Malden, MA: Polity. Hayles, N. Katherine. 2008. How We Became Posthuman. Chicago and London: University of Chicago Press. Heidegger, Martin. 1971. “The Thing.” In Poetry, Language, Thought, translated by Albert Hofstadter, 163–80. New York: HarperCollins. Heidegger, Martin. 1977. “The Question Concerning Technology.” In The Question Concerning Technology and Other Essays, translated by William Lovitt, 3–35. New York, London, Toronto, and Sydney: HarperPerennial. Heidegger, Martin. 1993. “Building Dwelling Thinking.” In Martin Heidegger Basic Writings, edited by David Farrell Krell, translated by Albert Hofstadter, 243–55. London and New York: Routledge. Heidegger, Martin. [1927] 2010. Being and Time. Translated by Joan Stambaugh and Dennis J. Schmidt. Albany, NY: SUNY Press. Heidegger, Martin. [1994] 2012. “The Thing.” In Bremen and Freiburg Lectures: Insight into That Which Is and Basic Principles of Thinking, translated by Andrew J. Mitchell. Bloomington and Indianapolis, IN: Indiana University Press. Ihde, Don. 1990. Technology and the Lifeworld: From Garden to Earth. Bloomington, IN: Indiana University Press. Ihde, Don. 2008. “Introduction: Postphenomenological Research.” Human Studies 31 (1): 1–9. doi:10.1007/s10746-007-9077-2. Irani, Lilly, and M. Six Silberman. 2013. “Turkopticon: Interrupting Worker Invisibility in Amazon Mechanical Turk.” In Proceedings of CHI 2013, 611–20. Janlert, Lars-Erik, and Erik Stolterman. 2014. “Faceless Interaction—A Conceptual Examination of the Notion of Interface: Past, Present and Future.” HumanComputer Interaction 30 (6): 507–39. doi:10.1080/07370024.2014.944313.

170

References

Janlert, Lars-Erik, and Erik Stolterman. 2017. Things That Keep Us Busy: The Elements of Interaction. Cambridge, MA: MIT Press. Jenkins, Henry. 2006. Convergence Culture: Where Old and New Media Collide. New York: New York University Press. Jones, John Chris. 1992. Design Methods. Second edition. New York: Van Nostrand Reinhold. Kallinikos, Jannis. 2007. The Consequences of Information. Cheltenham, UK, and Northampton, MA: Edward Elgar Publishing. Kellner, Douglas M., and Meenarkshi Gigi Durham. 2012. “Adventures in Media and Cultural Studies: Introducing the KeyWorks.” In Media and Cultural Studies, edited by Meenarkshi Gigi Durham and Douglas M. Kellner, 1–23. West Sussex: Wiley-Blackwell. Kiran, Asle H. 2012. “Technological Presence: Actuality and Potentiality in Subject Constitution.” Human Studies 35 (1): 77–93. doi:10.1007/s10746011-9208-7. Kravets, David. 2017. “Uber Said to Use ‘Sophisticated’ Software to Defraud Drivers, Passengers.” April 6. https://arstechnica.com/tech-policy/2017/04/ uber-said-to-use-sophisticated-software-to-defraud-drivers-passengers/#p3. Lanier, Jaron. 2010. You Are Not a Gadget. New York: Vintage. Latour, Bruno. 1999. Pandora’s Hope. Cambridge, MA and London: Harvard University Press. Latour, Bruno. 2004. “Why Has Critique Run Out of Steam? From Matters of Fact to Matters of Concern.” Critical Inquiry 30 (2): 225–48. doi:10.1086/421123. Latour, Bruno. 2005. “From Realpolitik to Dingpolitik.” In Making Things Public, edited by Bruno Latour and Peter Weibel, 14–43. Cambridge, MA and London: MIT Press. Latour, Bruno. 2007. Reassembling the Social. New York: Oxford University Press. Lessig, Lawrence. 2006. Code: Version 2.0. New York: Basic Books. Löwgren, Jonas, and Bo Reimer. 2013. Collaborative Media: Production, Consumption, and Design Interventions. Cambridge, MA and London: MIT Press. Maeda, John. 2000. Maeda@Media. London: Thames & Hudson. Manovich, Lev. 2001. The Language of New Media. Cambridge, MA and London: MIT Press. Marcuse, Herbert. 1964. One-Dimensional Man. Boston, MA: Beacon Press. Markham, Annette N. 2013. “Undermining ‘Data’: A Critical Examination of a Core Term in Scientific Inquiry.” First Monday 18 (10): 323–50. doi:10.1177/1077800405275056. McLuhan, Marshall. 1964. Understanding Media: The Extensions of Man. London and New York: Routledge.

References

171

Meehan, Eileen R. 2012. “Gendering the Commodity Audience.” In Media and Cultural Studies, edited by Meenarkshi Gigi Durham and Douglas M. Kellner, 242–53. West Sussex: Wiley-Blackwell. Meikle, Graham, and Sherman Young. 2012. Media Convergence. Houndmills, Basingstoke, Hampshire, UK and New York: Palgrave Macmillan. Meillassoux, Quentin. 2009. After Finitude: An Essay on the Necessity of Contingency. London and New York: Continuum. Menzel, Peter, and Charles C. Mann. 1994. Material World. San Francisco, CA: Sierra Club Books. Miller, Daniel. 2013. Stuff. Cambridge, UK and Malden, MA: Polity. Morton, Timothy. 2013. Hyperobjects: Philosophy and Ecology after the End of the World. Minneapolis, MN and London: University of Minnesota Press. Mumford, Lewis. 2010. Technics and Civilization. Chicago: University of Chicago Press. Nelson, Harold G., and Erik Stolterman. 2012. The Design Way: Intentional Change in an Unpredictable World. Second edition. Cambridge, MA and London: MIT Press. O’Neil, Cathy. 2016. Weapons of Math Destruction : How Big Data Increases Inequality and Threatens Democracy. Great Britain: Penguin Books. Örstadius, Kristoffer, and Linus Larsson. 2015. “Mobilen Spårar Dig – Överallt.” Dn. August 31. http://www.dn.se/ekonomi/mobilen-sparar-dig-overallt. Pasquale, Frank. 2015. The Black Box Society. Cambridge, MA and London: Harvard University Press. Peters, John Durham. 2012. Speaking into the Air. Chicago and London: University of Chicago Press. Plantin, Jean-Christophe, Carl Lagoze, Paul N. Edwards, and Christian Sandvig. 2017. “Infrastructure Studies Meet Platform Studies in the Age of Google and Facebook.” New Media & Society 20 (1): 293–310. doi:10.1177/1461444816661553. Puech, Michel. 2016. The Ethics of Ordinary Technology. New York and London: Routledge. Raley, Rita. 2013. “Dataveillance and Countervailance.” In “Raw Data” Is an Oxymoron, edited by Lisa Gitelman, 121–45. Cambridge, MA and London: MIT Press. Redström, Johan. 2005. “On Technology as Material in Design.” Design Philosophy Papers 3 (2): 39–54. doi:10.2752/144871305X13966254124275. Redström, Johan. 2008. “RE:Definitions of Use.” Design Studies 29 (4): 410–23. doi:10.1016/j.destud.2008.05.001. Ries, Eric. 2011. The Lean Startup: How Today’s Entrepreneurs Use Continuous Innovation to Create Radically Successful Businesses. New York: Crown Business. Shannon, Claude E., and Warren Weaver. 1959. The Mathematical Theory of Communication. Urbana, IL: University of Illinois Press. Shove, Elizabeth. 2007. The Design of Everyday Life. New York: Berg.

172

References

Simon, Phil. 2014. “Big Data Lessons from Netflix.” Wired. https://www.wired.com/ insights/2014/03/big-data-lessons-netflix/. Simondon, Gilbert. [1958] 2017. On the Mode of Existence of Technical Objects. Translated by Cecile Malaspina and John Rogove. Minneapolis, MN: Univocal. Smythe, Dallas W. 2012. “On the Audience Commodity and Its Work.” In Media and Cultural Studies, edited by Meenarkshi Gigi Durham and Douglas M. Kellner, 185–204. West Sussex: Wiley-Blackwell. Spuybroek, Lars. 2016. The Sympathy of Things. London and New York: Bloomsbury. Star, Susan Leigh. 1999. “The Ethnography of Infrastructure.” American Behavioral Scientist 43 (3): 377–91. doi:10.1177/00027649921955326. Star, Susan Leigh, and Karen Ruhleder. 1996. “Steps toward an Ecology of Infrastructure: Design and Access for Large Information Spaces.” Information Systems Research 7 (1): 111–34. Stark, Luke, and Kate Crawford. 2015. “The Conservatism of Emoji: Work, Affect, and Communication.” Social Media+ Society 1 (2): 1–11. doi:10.1177/2056305115604853. Stolterman, Erik, Heekyoung Jung, Will Ryan, and Martin A. Siegel. 2013. “Device Landscapes—A New Challenge to Interaction Design and HCI Research.” Archives of Design Research 26 (2): 7–33. Strong, David, and Eric Higgs. 2000. “Borgmann’s Philosophy of Technology.” In Technology and the Good Life? edited by Eric Higgs, Andrew Light, David Strong, 19–37. Chicago and London: University of Chicago Press. Suchman, Lucy. 1994. “Do Categories Have Politics?” Computer Supported Cooperative Work CSCW 2: 177–90. Suchman, Lucy. 2007. Human-Machine Reconfigurations. Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo: Cambridge University Press. The Economist. 2017a. “Data Is Giving Rise to a New Economy.” The Economist, May 6. https://www.economist.com/news/briefing/21721634-how-it-shaping-updatagiving-rise-new-economy. The Economist. 2017b. “The World’s Most Valuable Resource Is No Longer Oil, but Data.” The Economist, May 6. https://www.economist.com/news/ leaders/21721656-data-economy-demands-new-approach-antitrust-rulesworldsmost-valuable-resource. Tonkinwise, Cameron. 2014. “Design Away.” In Design as Future-Making, edited by Susan Yelavich and Barbara Adams, 198–213. London and New York: Bloomsbury. Turkle, Sherry. 1997. Life on the Screen: Identity in the Age of the Internet. New York: Simon and Schuster. Turner, Fred. 2010. From Counterculture to Cyberculture. Chicago and London: University of Chicago Press.

References

173

Usborne, Simon. 2013. “Can Cycling App Strava Change the Way We Ride?” Independent, July 3. http://www.independent.co.uk/life-style/gadgets-and-tech/ features/can-cycling-app-strava-change-the-way-we-ride-8685996.html. Vaidhyanathan, Siva. 2012. The Googlization of Everything. Berkeley and Los Angeles, CA: University of California Press. doi:10.1525/j.ctt1pn8d3. van Dijck, Jose. 2013. The Culture of Connectivity. New York: Oxford University Press. doi:10.1093/acprof:oso/9780199970773.001.0001. Verbeek, Peter-Paul. 2005. What Things Do: Philosophical Reflections on Technology, Agency, and Design. University Park, PA: Pennsylvania State University Press. Verbeek, Peter-Paul. 2008. “Cyborg Intentionality: Rethinking the Phenomenology of Humantechnology Relations.” Phenomenology and the Cognitive Sciences 7 (3): 387–95. doi:10.1007/s11097-008-9099-x. Viney, William. 2014. Waste. London and New York: Bloomsbury. Wakkary, Ron, Doenja Oogjes, Sabrina Hauser, Henry Lin, Cheng Cao, Leo Ma, and Tijs Duel. 2017. “Morse Things.” In Proceedings of DIS 2017, 503–14. New York: ACM Press. doi:10.1145/3064663.3064734. White, Jeremy B. 2017. “These Are the Most Bizarre Russian ElectionMeddling Posts That Have Just Been Released.” Independent, November 2. http://www.independent.co.uk/news/world/americas/us-politics/facebook-russiaads-posts-hillary-clinton-most-bizarre-posts-election-2016-a8032436.html. Wiener, Norbert. 1968. The Human Use of Human Beings. London: Sphere Books Ltd. Wiltse, Heather. 2014. “Unpacking Digital Material Mediation.” Techné: Research in Philosophy and Technology 18 (3). Philosophy Documentation Center: 154–82. doi:10.5840/techne201411322. Wiltse, Heather. 2017. “Mediating (Infra)Structures: Technology, Media, Environment.” In Postphenomenology and Media, edited by Yoni Van Den Eede, Stacy O’Neal Irwin, and Galit Wellner, 3–25. Lanham, MD: Lexington Books. Wiltse, Heather, Erik Stolterman, and Johan Redström. 2015. “Wicked Interactions: (On the Necessity of) Reframing the ‘Computer’ in Philosophy and Design.” Techné: Research in Philosophy and Technology 19 (1): 26–49. doi:http://dx.doi. org/10.5840/techne201531926. Winner, Langdon. 1986. The Whale and the Reactor: A Search for Limits in an Age of High Technology. Chicago: University of Chicago Press. Winograd, Terry, and Fernando Flores. 1986. Understanding Computers and Cognition. Boston, San Francisco, New York, Toronto, Montreal, London, Munich, Paris, Madrid, Capetown, Sydney, Tokyo, Singapore, Mexico City: Addison-Wesley. Wu, Tim. 2016. The Attention Merchants. New York: Knopf. Yelavich, Susan, and Barbara Adams, eds. 2014. Design as Future-Making. London: Bloomsbury.

Index

abstraction 1, 98, 100, 101, 106 active roles 57–8 actor-networks 112–16 actor-network-theory (ANT) 115 distinctive aspects of 115 methodological orientation and sensitivity 16 ad hoc groupings 33–4 advertisements 18, 28, 116, 124, 138, 141 advertisers 18, 21, 81, 111, 121, 132, 133, 136, 151 aesthetics of abstraction 124 aesthetics of immanence 161–3 agencies 57–9, 66, 71, 72, 74, 106, 154, 164 agential realism 66 “agile” development 36, 37 aging 11 Alaimo, Cristina 111 algorithmic arbitrage 124 algorithmic culture 122–5, 141–2 aspects of 100 algorithmic imagination 125 algorithmic objectivity 123 algorithms 106 cultural consequences of 123 pragmatic conception of 98 alien phenomenology 74 alterity relations 89 ambivalence of technologies 119 analog/digital converters 13 analog technologies 21 analytic distinctions 77 analytic frames 43, 45, 89 analytic playlist 127–30 ANT. See actor-network-theory (ANT) “app” genre 156

Apple 15, 18, 152 iAd program 28 Apple Watch 85 application programming interface 108 Aristotle’s philosophy 68 artifacts, ecological structure of 108 artificial 1, 5, 17 artificial intelligence 72, 106 assemblages 33–5, 67–9, 145, 155 agency of 33 character 56–7 characteristics of 67 character of computers 56 complex agency of 163 design aesthetics 39–42 design methodology 35–9 emergent properties of 34 ideas about 69–70 notion of 33, 146 topographical maps of 161 assembler 56 assembly/assembling 16–17, 24–5, 145 line 41 automata, historical fascination with 13 automatic action 72–3 background relations 89 Baudrillard, Jean 64 Bennett, Jane 33, 34, 66 Benson-Allott, Caetlin 95 big data 104–6 perspectives of 112 black boxes 44, 123 Bogost, Ian 72–4 Borgmann, Albert 83–4, 86, 131 framework 86

Index Bowker, Geoffrey 98, 103 bracketing 156, 158 branding 16, 25 bright objects 161 Bryant, Levi 70, 72–3, 161 business forms 18 calculated publics, production of 123 canonical view 102 CD players 23–4 "random" playback order in 24–5 Ceruzzi, Paul 55–7 chain of references 102 change 9, 11, 12, 14, 15, 61, 95 circulating references 101, 102, 103 classic record 23 CMC. See computer-mediated communication (CMC) Coleman, B. 53 collaborative media 51 commodities capitalism 3, 64–5 characteristics of 64–5 communication 50–4 conditions of 51 face-to-face 53 mathematical model of 51 media for 48 practices of 54 processes of 50 situated practices of 54 complex computational systems 81 complex ecosystems 109 complexity 5, 12–13, 22–3, 31, 49, 57, 62, 66, 82, 94, 95, 153, 155, 162, 163 compositions and breach of contract 29–31 computational components 87–8 computational device 56, 81, 84, 130, 148 computationalism 97 computational technology 13–14, 87, 92, 101, 145 computational things 14, 16, 21, 22, 48, 55, 59, 67, 68, 70, 72–3, 92, 155, 162 computer-mediated communication (CMC) 51 computers 54–7, 151 assemblage character of 56 programmability of 56

175

conceptual tools 77 connection mechanisms 134 consumers 57, 132, 159 consumption 15–17 configuration of 37–8 contemporary design 30 contemporary media landscape 121 contemporary music-playing systems 25 contract, compositions and breach of 29–31 correlationism 70 craft methods 47 cultural preparation 118 cultural significance 122 culture industry 122 culture machine 100 customers, demands of 107, 108 customization 13, 15, 25, 27–9, 108, 152, 159 cyberculture 55, 56 cybernetic capitalism 159, 161 cybernetics 52, 98–100, 135, 159, 161 significant legacy of 98 cyberwarfare 95–6 cyborg relations 89 cycles of anticipation 123 data collection 102–3, 112 combinations of 106 and information 100–7, 136–7 massive quantities of 120 points 125 processing 54–7 data-driven economy 120 data-intensive processes 50 dataveillance 104, 159 programs 161 deconstruction 161 DeLanda, Manuel 67 Deleuze, Gilles 33, 34, 43, 68, 69, 75 democratic rationalization 119 “democratization” of media participation 122 design/designing 4–5, 15–17 approach, foundational for 61–2 central implications of 162 domains of 22, 23 intervention 160 methodology 35–9

176 studies, scholarship in 113 theory 5, 17 types of 62 worldview 22 devices connectivity and complexity of 95 landscapes 108 paradigm 83–4, 131 digital materials 30–1, 57 digital media 51, 104, 109 digital music players 21, 24 digital networked things 14, 50, 97 digital technology 13, 14, 84. See also technology disclosure, forms of 147 discursive frameworks 44 discursive practices 44 distribution patterns 81 Do Not Track signals 138 duomining 71 Duportail, Judith 105 dynamic ad markets 153 dynamic audio 133 Easterling, Keller 96 “ecological” mode of interacting 94–5 ecosystems 21, 49, 71, 94, 95, 97, 108–10, 128 EDAA. see European Interactive Digital Advertising Alliance (EDAA) effective computability 98, 100, 134 eidetic reduction 156 eidetic variation 156–8 Ellul, Jacques 117–18 embodiment relations 89 engagement 82–7, 131–2, 135–6 complex patterns of 84 entanglement with practice 123 equipment 40, 58, 74, 79–82 evaluation of relevance 123 exteriority 34, 39, 41, 147 external storage device 24, 55 external vs. internal complexity 12–13 Facebook 81, 111, 155 advertising on 81 marketing 81 face-to-face communication 53 feedback mechanisms 106

Index Feenberg, Andrew 113–14, 119 Finn, Ed 97–8, 100, 124–5 fixity 96 flat ontology 71 flexibility 13, 55–7 fluid assemblages 1, 39, 66–7, 78, 124, 141, 142, 145, 146, 154, 160, 162 becoming and operation of 161 conception of 96 context of 120, 160 description of 6, 17 designing 41, 42 example of 127 function of 99, 122 outline of 159 producers and consumers 18 sense of 135 shifting to 16, 18 specific instances of 99 thingness of 145 tuning formations of 159 fluidity 35, 72, 96 formation, processes of 159 function 11, 13–17, 19, 21–4, 28, 33, 38, 40, 41, 46, 48, 64, 73, 79, 80, 85, 86–7, 99, 100, 104, 108, 109, 118, 121, 122, 127–30, 138 functional reality 100 Galloway, Alexander 92–3, 96 general demographics 81 geophilosopy 161 Gibson, J. J. 10, 12 gig economy 120 Gitelman, Lisa 53 globalization 16 God’s eye view 105, 106 Golumbia, David 98 Google 82, 153 ads 28 GPS 15 gravity 160 Guattari, Felix 33, 34, 43 hacking 95–6 handiness 80 Harman, Graham 70–3 HCI. See human–computer interaction (HCI)

Index Heidegger, Martin 20 n.2, 64, 79, 86–7, 116–17, 145–7, 155 analysis 80 heritage from 152 ontology 80–1 hermeneutic relations 89 hippie entrepreneurs 55 historical logic 98 Hobbes, Thomas 9 human activity 45–6 domains of 45 mediational structure in 58 human agency 47, 91, 119 human analysis 106 human–computer interaction (HCI) 48, 81, 108 human experience 5, 39, 45, 47, 59, 156 human perception 35, 88, 94 human subjectivity 47 human–technology relations 151 human–world interface 49 hypermediacy 51 hyperobjects 71 ICD. See international classification of diseases (ICD) iCloud platform 108 ICTs. see information and communication technologies (ICTs) identity 9, 16, 34, 46, 52, 53, 61, 66 immanence 35, 41, 72, 161–3 immaterialism 71–2 immediacy 51 industrial capitalism 3 industrial design 5, 22–3, 47, 160, 161 branding 16 mass consumption 15–17 objective of 11 product of 15 streamlining 16 industrial infrastructures 110 industrial opportunities 2 industrial production 2, 15–16, 161–2 Industrial Revolution 1 information processes of 50 types of 55 information and communication technologies (ICTs) 100

177

information-based processes 50 infrastructures 100–12 understanding of 100 initiatives, motivation for 65 intelligence 55, 72, 106, 135 intentionality, importance of 157 interactions 62, 162 design discourse 162 interfaces 92–7 characteristics of 92 local 92 software and 96 structures of 145 interiority 34, 39, 41 international classification of diseases (ICD) 103 Internet of Things (IoT) 12, 19, 95 interpretation 11–12, 51, 52, 80, 101, 155, 156 IoT. See Internet of Things (IoT) iTunes 25 data and experience 26 Janlert, Lars-Erik 94 Jones, John Chris 47 Kallinikos, Jannis 111 knowledge logic 123 production and certification of 123 KOM 85 Latour, Bruno 101–2, 115, 116, 136 “lean” development 36, 37 legal frameworks 139–40 listening equipment for 130–1 to music 21, 127 machine 72 animating agencies of 72–3 concept of 73 learning 160 manually building 35 Marcuse, Herbert 118–19 market feedback loops, development of 161 Marx, Karl 64–5 mass consumption 16, 58 massive surveillance, programs of 55

178 mass multiplication 69 mass-produced object 148, 163 mass production, requirements of 36 materialism 71–2 ideas about 69–70 materialist perspectives 66 materials 5, 9, 11, 16 speculation 74 McLuhan, Marshall 51 means of production 58, 122 mechanical construction 12–13 mechanisms of evolution 74 media 50–4 for communication 48 definition of 53 dynamics of 58 forms, convergence of 128 landscape 58 technologies 51–2 mediations 88–90, 132 kinds of 132 Meillassoux, Quentin 70 meta-data 25–9, 105 military intelligence 55 Millar, David 85 Minimum Viable Product 37 modern computing 56–7 moments-based marketing 131, 133 Morton, Timothy 65 Moses, Robert 113 MP3 file format 25 multiinstability 151–4 multiintentionality 155–9 multiple complex dynamics 45 multistability 20 n.1, 89, 132 concept of 151 notion of 154 Mumford, Lewis 72, 118 music 25–9 distribution 130 industry, business models in 130 listening to 21, 136 playback, techniques for 21 music players 25 production and customization of 27 short history of 23–5 music-playing technologies 21. See also technology pre-digital examples of 23

Index music-playing things 28 assembling of 28 “natural” physical properties 74 Netflix, production of “House of Cards” 123 net neutrality 122 networking capability 2, 13–14 object-oriented perspectives 69–70 object-oriented philosophy 71 object-oriented programming 35–6 objects 57–8, 64, 73 concept of 72 functional design of 22 resembling the structure of 77–8 "undermining" and "overmining" of 71 ontocartography 73, 161 practices of 161 ontology 43, 59, 61, 68, 73, 80, 96 openness 87, 92, 93, 95 open playlist 142 open-source software 15 passive roles 57–8 patterns of inclusion 123 perceptual systems 10 personalization of things 160 phenomenological method 79, 157 foundations of 47–8 phenomenology of technics 89 tradition of 79 philosophical carpentry 74 philosophical lab equipment 74 philosophy 5 “empirical turn” in 117 philosophy-physics 43–4 physical form 1, 9, 152, 153, 160 changes in 11 physical properties 10, 74, 145 Piaget, Jean 10 piracy 139 Plantin, Jean-Christophe 111 platforms 15, 18, 81, 82, 108, 110–12, 119–22, 131, 135, 136, 139–40 Plato 147 playlists 128 pocket calculator 13 political economy 120–2, 141

Index political frameworks 160 politics 112–16 postphenomenology 19–20 n.1, 88–90 development of 47 praxis 49 present-as-particular 147–50 privacy policy 127 processor 13 product ecology 108 production configuration of 37–8 forms of 2 profit-extracting sieve 111 programmed thing 14–15 programming 107 properties 62 protocols 92–7 connective possibilities of things 93 definition of 92 of networked computing 93 of RFID 93 proximal technosphere 91 public relevance algorithms 123 QOM 85 “quantified self ” movement 103 quotations, extensive use of 77–8 Radio Frequency Identification (RFID) 57 Raley, Rita 104 Rdio3 online application 28–9 realization, process of 68 “re-engage” listeners 134 relations 57–9 remediation 51 remote activity 14 representational practice 51 RFID. See Radio Frequency Identification (RFID) roles 57–9 Ruhleder, Karen 110 runtime assembly 27–9 runtime customization, source of input for 28 Ruskin, John 96 scale 44, 62 scapes 90 scripts 49

179

seamless totality 35, 41 search engine 18, 53, 121, 153 segments 85 self-expression 111 “self-showing” of phenomena 79 Shannon, Claude 51 shape/form 157 ship of Theseus 9 Simondon, Gilbert 109 Simon, Phil 107, 109 simplicity 22–4, 31, 48 sleeping policeman 115 smart home systems 152 smartphones 14, 15, 34–5, 38, 65, 88, 151 social contract 7, 30, 162 social media 28, 51, 53, 71, 95, 119, 122 social theory 64, 72, 118 societal dimensions of things 112 software and interfaces 96 start-ups 104–5 updates 15 solidity 10, 48, 61 speculative realism 74 speculative turn 70 Spotify 18, 127 algorithmic culture and 141–2 algorithms 134–5 application 129–30 assemblage 135 attractions of 131 build experiences 135 cookie policy 138 data-fueled view 136 description of 138 dynamics in 128 ecology 138–9 functions of 141 as infrastructural platform 139 political economy 141 politics of 140 techniques 140–1 workings of 131 stability 11, 37, 66, 151–4 stages of life 63 Star, Susan Leigh 102, 109–10, 113 Stolterman, Erik 94 streaming 27, 124, 132–3, 135 structure, issues of 61–2

180 stuff 1, 12, 19, 49, 50, 57, 62, 63–7, 67, 74, 101, 145 subject 57–8, 90, 91, 96, 115, 160, 163 Suchman, Lucy 48–9 systemics 62 system monitoring tools 24 systems approach 56 tape players 23 technical apparatuses 51 technical artifacts 5, 17 technical complexity 22 technical evolution, internal structural rationality of 109 technical novelty 96 technical rationality 118 technological advances 49, 86 technological determinism 51, 153, 154 technological developments 2, 3, 4–5, 18 technological ecologies 108–10 technological ecosystems, particular dynamics of 109 technological progress 84, 151, 153 technological things 83–4, 114 technological tools 48, 49, 87, 90, 105 technology 148 abstract initial version of 109 ambivalence of 119 complexity of 22 development of 21 domain of 22 “empirical turn” in 117 essence of 117 instrumental conception of 117 instrumental use of 49 powerful role of 113 structural analysis of 109 substantive theories of 119 technospheres 90–1, 134 terraformation 161 theoretical frameworks 4 things/thingness/thinging 1, 18–19, 27, 61–3, 83, 145–7 actions and effects of 2, 14 aspects of 44–5 assemblages 16, 67–9 characteristics of 3, 11, 12 conventional view of 112 designed 5, 6

Index development of 145 digital networked 14 ethnography 74 experience and engagement 5 fluid assemblages (see fluid assemblages) forms and functions 11 function of 83 general orientations to 44 “general” understanding of 46 mechanical construction of 12–13 objects/machines 69–74 personalization of 160 perspective of 153–4 physical form 1–2 relation to 9 stuff 63–7 technical artifacts 5 understanding of 83 value of 3 thing theory 5 time-sharing model 56 tools action and perception 46–9, 47 domain of 22, 47 influential philosophical analysis of 79 for specific purposes 48 technological 48 totality 34, 39–40, 163 seamless 35 transcendence 159, 162 transdisciplinarity 77 “trending” category 128 tuning 159 formations 159–61 Turkopticon project 120 turn to things 18, 19, 65–6 unit operations 73–4 “useful thing” 79–82, 86 user experiences 14–15, 17, 25, 43, 70, 71, 112 Vaidhyanathan, Siva 121 variations 88 Verbeek, Peter-Paul 47, 86 “walled garden” approach 25 Weaver, Warren 51 web applications 18

Index web-based music players 26, 27 wicked problems 22 Wiener, Norbert 98, 99, 135 WinAmp player 25

wireless internet 15 World Wide Web 51, 57, 110 “Year in Music” analysis 134–6

181