Boxes And Arrows : The Design Behind the Design

Hiding in Plain Sight: An Interview with Adam Greenfield

by B&A Staff on 2006/02/27 | [5 Comments]

Boxes and Arrows caught up with Adam Greenfield on the heels of finishing his first book,
Everyware: The Dawning Age of Ubiquitous Computing, due out in March 2006. Greenfield talks to us about how computing has moved away from the desktop into every part of our lives—from soda cans to the family pet. In this interview, he allows us to imagine what our new normal might look like.

Boxes and Arrows: Congratulations on your book! What is “everyware?” Is it different from what we already know as “ubiquitous computing,” “pervasive computing,” or “invisible computing?”

Adam Greenfield: “Everyware” is computing that is everywhere around us, yet is relatively hard to see, both literally and figuratively. Broadly speaking, it is what you get when you take the information processing we associate with the personal computer and distribute it throughout the environment—embedding it in walls, floors, appliances, lampposts, even clothing. I also use the word to refer to the relatively novel interface conventions everyware requires: gestural, tangible and haptic interfaces, and to some extent, voice recognition.

The fact that it is so powerful—so insinuative and at the same time so hard to discern—makes it different in kind from the informatics we’ve grown so used to over the last twenty or twenty-five years of the PC era.

“Everyware” has a lot in common with the contemporary discourses of ubiquitous computing, so why coin an entirely new term? Each of the terms already in use—”ubicomp,” “pervasive computing,” “tangible media,” “physical computing,” and so on—is contentious. They’re associated with one or another viewpoint, institution, funding source, or dominant personality. I wanted people relatively new to these ideas to be able to have a rough container for them, so they could be discussed without anyone getting bogged down in internecine definitional struggles, like “such-and-such a system has a tangible interface, but isn’t really ubicomp.”

B&A: When did you start noticing the emergence, if you will, of “everyware” as a concept?

AG: My own first exposure to the idea came in early 2002, when Anne Galloway asked me if I was attending the Ubicomp conference in Goteborg. Like many people in the UX field, I was already familiar with Don Norman’s arguments about the invisible computer, Philips’ work with wearables, a few experiments in augmented reality, and the early manifestations of what’s now being called “the digital home.”

But discovering Mark Weiser’s work on ubicomp and “calm technology” at Xerox PARC, which dated to the late 1980s, was critical because it gave me a framework in which I could understand the interrelationship of these other, seemingly disparate developments—like being given all of the edge pieces of a jigsaw puzzle.

And because I suffer from apophenia—from the tendency to discern pattern in the world whether it’s “actually” there or not—once I had that framework in place almost every IT innovation I read about seemed to find a home somewhere within it. I started to see information processing, well, everywhere: plans to embed RFID tags in cargo pallets, individual soda cans, even the family cat. New wireless-networking techniques that allowed people to form “ad hoc” networks on the fly. Increasingly, service delivery via mobile phone.

Together, these experiences drew what seemed to me to be a pretty clear picture of where we were going with information technology. And surprisingly enough, very few people were connecting all of the relevant dots. I figured somebody should try to articulate these connections and help people understand what was in the offing.

B&A: As user experience professionals, we have been striving to create transparent systems, seamlessness. Good design is, after all, invisible. Yet, in Everyware, you maintain that knowing where the seams are may be increasingly important. What is your position on seamlessness? Is it important to respect and maintain seams?

AG: You’re absolutely right to point out that one of the central dogmas of UX work is that users should be insulated from the actual complexity of the systems that support them. They don’t need to know about what happens on the back end, we’ve argued, they just want to order a book, purchase a flight, or find an address on a map. And they want to do it quickly, “intuitively,” and with a minimum of effort.

But the deeper I’ve gotten into considering everyware, the more I’ve questioned the whole notion of seamlessness. [1] In many cases, this idea deprives the user of meaningful insight into and participation in the decisions that affect their experience.

Let’s say that you’ve recently moved your elderly mother into a home specifically designed to allow her to “age in place,” live as autonomously as possible and with as much dignity as possible for as long as possible. One of the provisions of this home is instrumented flooring: flooring that contains impact sensors that register when someone has fallen and automatically call for assistance.

In the “seamless” version, if she should happen to fall, the floor registers the event, a signal is generated, the local first responders are called, maybe the front door is automatically unlocked, and someone is able to come and see if she’s OK. It’s transparent to the user, but it’s not at all the same thing as good user experience. There are any number of places your elderly mother might want to configure this transaction: maybe she has a strong sense of pride, and doesn’t necessarily want other people to know how often she’s falling. Maybe she has a strong sense of privacy, and doesn’t want strangers—even well-intentioned ones—in her intimate and private space unless it’s literally a life-threatening circumstance. Seamlessness in such a system would deprive her of the autonomy that we all expect to be able to enjoy as a matter of course. While it was designed with the noblest aims to support her dignity, it would actually wind up infantilizing her.

We’ll see more situations like this, as we allow the tendrils of everyware into our lives. That’s why I argue that seamlessness should always be an optional mode, not an inescapable one, and why I’ve tried to challenge the discourse of seamlessness wherever it appears unquestioned, as in the current W3C position on the “ubiquitous Web.”

B&A: In your book, you suggest that everyware may threaten human agency itself. Where once we had to make choices, had to sit down at a computer and start a program, going forward, the choice will be made for us. Are we seeing the death of agency as we know it?

AG: In the abstract, I think we’ll always retain the ability to make the ultimate refusal, which is just to shut all of these ubiquitous systems down and do without. So in that sense, the answer is clearly “no.”

That said, there are all sorts of easily foreseeable reasons why this freedom to choose may be the exception rather than the rule. Marshall McLuhan said that “every extension is also an amputation,” and this was never truer than as it applies to everyware. Once we’re used to the array of powerful ubiquitous systems operating all but imperceptibly on our behalf, how many of us are going to have the will (let alone the wherewithal) to do without? Already, when I’m confronted with a question of factual knowledge, my reflex is to make immediate recourse to Google and Wikipedia. How much stronger will that reflex be when similar resources are literally everywhere around me and as free as the air?

Everyware will lever itself into our lives on the twin fulcrums of “convenience” and “security.” Once we’ve welcomed it in, we will have a hard time returning to an unmediated existence. I don’t want to say that everyware is necessarily the “death of agency,” but in its fullest development, it represents what it means to experience the world and make choices about what is experienced.

AG: I guess that sort of depends on what your everyday looks like, doesn’t it? Bluntly, there are a whole lot of people in the world whose everyday life sucks. I can easily imagine people in such situations choosing systems designed to “simplify” their lives over an unmediated experience, where whatever they’d be trading away might seem like a valueless abstraction. [2]

So, yes, some people will want their everyday experience rebuilt on a technical model. They will want to configure their bathtub or their closet or their commute to work the way we’d now configure a new laptop, either because they’ll perceive some benefit to doing so or because they won’t have a full understanding of the tradeoffs involved.

It’s not for me to judge that desire, but I feel that it’s important to point out what should be obvious: that there will be many people who do not want to remake the operations of everyday life along technical lines, and they shouldn’t be forced to do so.

B&A: Then let’s talk about opting out. When we are in control, we have the option to hide, to consider which face to put forward, even to manipulate. Will we still have this ability? And if not, are people ready to lose this control?

AG: I’m not so sure we will retain much of this ability, which in sociology is generally referred to as “presentation of the self.” With so much information about our past and current activities available to be searched, cross-referenced, and made available in real time, when we meet someone for the first time, we are likely going to lose control over the image we present to them.

Imagine what this will look like in practice. Whether you are interviewing a prospective new hire, meeting a potential romantic interest for the first time, or simply sitting next to someone on a plane, you no longer have to take a person at face value. It’s easy to see that this can occasionally be very useful, if you happen to be on the empowered end of the transaction. The trouble is that this ambient intelligence—facilitated by a ubiquitous deployment of informatic systems—cuts both ways.

And with the ability to control how others see us, I believe that we lose also a certain protective and beneficial hypocrisy that allows us to function as a society. We all, without exception, have habits, behaviors, experiences that we don’t necessarily want to share with the wider world. When you evert these experiences, and archive them, and tag them with metadata, and make them persistently accessible, it gets very difficult indeed for anyone to maintain the unimpeachable public façade our current mores require of us.

This is something that people who consider ubiquitous computing from a purely instrumental or technical perspective frequently miss: it’s not just a change in the way we use computers, it’s an alteration in some of the very foundations of the self as it’s been constructed in the West for the last few centuries. We’re in for a wild ride.

B&A: You know that when we’re doing something “on the record,” we tend to act and speak a bit differently, even in contrived ways at times. So how will this awareness of everyware affect how we present ourselves?

AG: Well, I think you’ve hit the nail on the head. Anyone who’s had, for one reason or another, to get used to being in front of cameras or microphones with any degree of regularity knows how hard it is to be “natural” when confronted with the prospect of being recorded, or transmitted to a large audience, or both.

When artifacts like cameras and microphones (to say nothing of sensors capable of recording one’s position and location, and verifying one’s identity via unique biometric signatures like retinal patterns or even gait period) are embedded in the objects and surfaces of everyday life, we all potentially become subject to the most intense kind of mediation. Barring some regulatory or other intervention, we’ll be forced to assume that we’re at least potentially “on,” just about all the time. And the sheer ubiquity of output modes offered by the robust deployment of everyware means that whatever once goes into the network can come out again just about anywhere.

Among other complications, this strikes me as being very likely to give rise to many of what MIT sociology professor Gary T. Marx calls “border crossings”: irruptions of personal information at an unexpected place or a time, in an unexpected context. Again, I don’t think we’re even remotely prepared for what this is going to do to social cohesion.

B&A: There seems to be a new transparency of social value. Flickr tells us who considers us real friends versus contacts. We are on at least one person’s blocked buddy list. We know how many times we’ve been hotlisted. Are we losing anything by being so aware of our perceived social currency?

AG: Sure. Or at least I surely think so. Again, we lose control; in this case, control over whether we are unintentionally causing someone else hurt or offense. Social networking software, at least here in its infancy, very often seems tone-deaf to the nuances of social concourse that every second-grader has internalized.

In other words, a list of my friends, especially when it’s ordered in some way, is not at all a neutral document. To put it in Flickr’s terms, why do you consider her a “friend,” but me only a “contact”? When people I don’t know make me a contact of theirs, isn’t there at least some shred of a reciprocal obligation? Doesn’t refusing such a request put me in the highly awkward position of snubbing someone I’ve never met, but still have no wish to insult?

And so for the most part, these moments of awkwardness are mediated by the personal computer. Imagine what happens when the mediating artifact is something as omnipresent as a mobile phone or has disappeared entirely into the surrounding architectural space. How much more raw will these conflicts appear? How much more deeply will we internalize the damage to our self-image? This is why I recommend that, wherever possible, everyware be designed so that it is “conservative of face”; that it allows us to preserve a healthy amount of amour-propre.

B&A: When do you disconnect? In other words, when is it valuable for you to escape everyware?

AG: Me personally? Ironically enough, perhaps, I’m actually a pretty disconnected guy. I rarely carry a mobile phone, for example.

But beyond that, there are plenty of areas of life that would not be improved in the slightest if they were network interconnected: having a pint of Guinness with good friends, going for a long sweaty run, spending time with my wife. I already cherish every moment of my life I spend in such pursuits, I wouldn’t necessarily want to see them macerated and fed into the global mnemotechnical system. And I would imagine that there are a great many people who feel just the same way.

B&A: What are you doing to ensure that the information in Everyware, the book, is itself ubiquitous?

As soon as practically possible, subject to negotiations with my publishers, I’d like to make the book available as a full and free download covered by a Creative Commons license.

I’ve also set up a discussion board for anyone who wants to find out more about ubiquitous computing in all of its aspects, or share what they’ve learned with others. And of course I’ve committed myself to touring in support of the book well into next year.

Beyond that, the book offers five broad guidelines for the ethical development of ubiquitous computing (originally proposed in an article right here on Boxes and Arrows), and there’s already been some discussion on how to incorporate these guidelines into something like an international standard. Clearly, there’s a long, long way to go on this latter front, but it’s one of the most exciting things to have come out of this work. Should it come to pass, I would find it extraordinarily gratifying.

Whether it’s these particular guidelines, some modification of them, or others entirely, the most important thing is the recognition that people do not have to sit passively by and accept the everyware they’re offered. The stakes are too high for that.

Everyone who will be affected by this class of technologies should have a voice in shaping its emergence.