Adam Greenfield (adamgreenfield) Sat 22 Apr 06 09:27
Sure. I think what we have to wrap our heads around, though, in order to understand this sort of scenario, is that everyware doesn't merely redefine the way we engage computers. Among other things, it redefines surveillance, as well - so that we're not merely talking about surveillance as something that involves cameras or microphones or wiretaps, but about all of the traces we leave behind and cannot help leaving behind as we move through the world. The example I'm about to share will hopefully put all of these claims in context. I'm going to posit a few discrete, separate, but interlinked systems. All of them either exist today as commercial products or services, or have been prototyped convincingly. The only part of this scenario that is in any way speculative is the fact of their real-time interconnection. Let's say that there's a way to determine your identity, to a reasonable level of certainty, by analyzing the pattern of your footsteps - the unique signature of your tread, your gait period and so on. And let's assume that this is something that's done by load cells and processors that are built into the flooring itself, which are imperceptible to ordinary inspection. And let's also say that there exists on the network an open database of contributions to political parties and candidates, as well as the analytic applications necessary to derive high-level inferences from the pattern of one's contributions over the past few years, so that a given person can be tagged with a rough precis of their likely sympathies: Republican, Democratic, Independent, Apathetic... And finally, let's say that there's a remote door-actuator mechanism, that either opens or bars a door based on the results of a network query. OK, fine. Nothing here is that strange, right? The tread-pattern sensor has been prototyped by ubicomp researchers Georgia Tech (among others); the database exists and is a matter of public record; and the analysis and visualization software certainly exists. The door-actuator is trivial. What do you think happens when we put all of these components together? When it's reasonably easy to build a "public" space that ascertains your identity, associates a political position with that identity, and either allows you access or bars your way depending on your conformity with a desired profile? And to circle back to your question: this accomodation or restriction is something that happens entirely as a result of invisible processes. You see a room and a door, and a stream of people who either get through that door or do not, but there is *no immediately obvious correlation* between who gets to pass and who doesn't. And what's worse is that there's not even necessarily any real-time recourse! The whole set-up is implacable, it's "talk to the hand" made utterly concrete. OK, so this is admittedly a provocation. But there is nothing in this scenario that is intrinsically impossible, very little that's even all that impractical...and it allows a space to be conditioned in a terrifyingly effective way. And none of its componentry is even perceptible, from the point of view of someone encountering it under ordinary circumstances.
Jamais Cascio, WorldChanger (cascio) Sat 22 Apr 06 12:49
A perhaps less legally-actionable version -- or, to put in another way, a more likely in the near future version -- of the above would be to replace the "who did you vote for?" database with the DHS "No Fly" list. Imagine not even being able to enter the airport, let alone get to the counter.
Adam Greenfield (adamgreenfield) Sat 22 Apr 06 15:27
Exactly! Or *not even be able to charge a ticket*. And, of course, there's no legal recourse there either. You'd never know why - you'd only know that you couldn't.
Brian Slesinsky (bslesins) Sat 22 Apr 06 21:53
I don't get it. Why does it make any difference whether it's a security guard who checks the do-not-fly list, versus someone watching through a camera and buzzing you though the door, versus a fully automatic system? In any case you don't know how you got on the list, and the legal recourse is the same: suing the government somehow.
Adam Greenfield (adamgreenfield) Sun 23 Apr 06 06:49
I don't think the two situations are equivalent, Brian, and here's why: The everyware scenario we're discussing, which presupposes widely distributed sensor nets, network access to multiple relational databases, and actuators arrayed arbitrarily throughout public space, enables a true "defense in depth" to be set up. You can be denied access to something *without knowing* what (or whose) profile you violated, and this denial can be effected at a sufficient distance from your objective in space and/or time that you fail to make the connection. The linkage between cause and effect has been (a) obscured and (b) rendered "technical," ostensibly neutral, and subject only to technical expertise. Who, indeed, do you complain to? Who do you vent your displeasure on? The locus of control is somewhere in a mesh of interlocking systems, but you don't know where, and there's no way you'll be able to find out in time to catch your flight...or, probably, your next several flights. I'll leave the question as to whether or not "suing the government" is genuinely adequate or sufficient recourse for another time.
virtual community or butter? (bumbaugh) Sun 23 Apr 06 09:45
you can be denied access to things without even *knowing* you're being denied access! You don't necessarily know whether the door locked in response to *you* being there or whether it was just locked; if the reservations system just tells you "try again later" rather than "declined," you don't know you were denied a ticket rather than the system was on the fritz; and so on. Or, the denial could be less than complete -- maybe the door opens for those in the "wrong" category, but it sticks each time. (PSYOPS approach?) Maybe if every day is irritiating in these little ways, "wrong" people could learn to do things the "right way." That's awful and insidious. Great book, Adam. Thanks for bringing this to our attention. Now, the standard line of critique we hear so often: But that would be *people misusing the technology*, nothing inherent in the technology itself. You've thought about this carefully. What say you?
Adam Greenfield (adamgreenfield) Sun 23 Apr 06 12:48
Well, I think I need to be careful about just how I word this claim, lest I inadvertently prop up just the kind of technodeterminism I'm so opposed to. But I do think that each new technology enters the world with a certain spin on it - certain tendencies, certain gradients along which connection with other technologies and techniques is made easier, and others along which it is less likely to find purchase. So, for example, in the book I point out that the next version of Internet Protocol, IPv6, seems to "want" to make everything in the world a network node, due to the fabulously large address space it affords. Or that, because of its simplicity and cheapness, RFID tagging seems to "want" to mirror every artifact of the actual world in the virtual. And given that, I don't think it's inappropriate to say that everyware as a class of technologies has inherent propensities, and that if we want to maximize the benefit we derive from it it really does behoove us to find out what those propensities are. In other words, my concern is that the "proper" use of everyware can have effects that are just as deleterious to quality of life and well-being as putative misuse. Now, I don't think that saying that undercuts pointing out, at the same time, that the kind of ubiquitous systems we're discussing have those propensities as a consequence of specific design decisions, specific purchase decisions, made by knowable institutions at knowable times. None of this is to deny the place of human agency, or to imply that the tendencies we can perceive result somehow from some spooky action. Inherent is not the same thing as immanent, right? What this all leads me to argue is that yes, this class of technologies does have some (rather easily) foreseeable potential to produce less desirable outcomes, *and* that if we do a good job now raising consciousness about the implications of adopting it, we stand a much better chance of making responsible design (specification, purchase, use...) decisions.
Brian Slesinsky (bslesins) Sun 23 Apr 06 13:06
If it's a public place like an airport then I think it's pretty obvious that you should have access, and if you don't then something is clearly going on. Also, the airlines are always going to have customer representatives to complain to when something goes wrong. I think that's generally true of any business serving the public. As routine transactions are automated, it's increasingly why businesses even have front-line people. As for obscure doors that nobody knows about, I think that's always been true of some private clubs. The criteria by which some people get into some nightclubs, or some people can get reservations at some restaurants more easily, or people get admitted to universities, has never been entirely transparent. And yet we all kind of know what's going on. The thing that would be genuinely new would be the ability to automate this, making it cheaper and more common, and use criteria beyond what one person controlling admissions could easily deal with. Maybe there would be a blurring of the registration process where people somehow become members of some places without really knowing how that happened. But hiding the existence of such places is going to depend on the ability of members to keep it a secret, which is always pretty dicey and contrary to what a profit-making business wants to do. And if the club is important enough then you get into situations where people sue to join (as happened with some men-only clubs). Similarly for the subtle thing where some people get "try again later." It's indistinguishable from having a shoddy UI. Companies that want to have a good reputation with their customers will want to avoid that. There is such a thing as word-of-mouth and it's becoming increasingly important in the Internet age; the Cluetrain folks have a few things to say about it.
Adam Greenfield (adamgreenfield) Sun 23 Apr 06 13:48
So we should just trust in the power of the market, safe in the knowledge that all these enlightened, post-Cluetrain companies have our best interests in mind?
Jef Poskanzer (jef) Sun 23 Apr 06 14:05
I think the basic complaint here is machines lying to us, as opposed to the long hallowed tradition of other people lying to us.
Adam Greenfield (adamgreenfield) Sun 23 Apr 06 15:07
Jef, there's no "complaint" here at all. To quote someone I'm quite fond of: "What it is...is up to us."
Brian Slesinsky (bslesins) Sun 23 Apr 06 15:24
I'm not saying there's no problem, but rather that we need some more fleshed-out examples than the "provocation" in #26. There's certainly a lot of potential for spooky art installations, and I can see how the underground scene could really get into this stuff, but mainstream usage is probably going to be a bit more subtle.
Adam Greenfield (adamgreenfield) Sun 23 Apr 06 15:38
Isn't that precisely the point, though? That everyware will be subtle, and therefore exceedingly hard to challenge (let alone counter)?
Christian Crumlish (xian) Sun 23 Apr 06 17:27
Adam, a few weeks ago I was in Vancouver for the IA Summit where there was a lot of joyful buzz about your book. More recently I found myself reading your book on plane trips to and from Memphis (where I'm doing some consulting for a package delivery company that can remain nameless) and despite the slim contours and cheerful ambience of the cover of the book, I found that it did not necessarily make great airplane reading. And I mean that as a compliment. I found that every sentence required savoring. That I had to frequently stop to digest the last paragraph before going on to the next. Where I might have expected a breezy survey, I was instead encountering a deep dive. So my compliments, first of all, to your achievements as a writer. As someone who has attempted to write about technology and its implications for sociology in a humane way, I believe I have a lot to learn from stealing^H^H^H^H^H^H^H studying your approach. So, a question: Despite all the examples you give of projects in the works, leaving labs and entering the commercial space, I still found myself quite skeptical about how quickly these things are going to be, well, ubiquitous. I recognize that much off this is going on by stealth, and that computer chips have embedded themselves already in many common objects that did not formerly harbor them. Still, I get that "they promised me a jetpack" feeling sometimes reading your book. Can you talk a little about the timelines for some of these aspects of real, existing everyware showing up in our day-to-day lives?
Adam Greenfield (adamgreenfield) Sun 23 Apr 06 18:57
Sure, Christian. And again, I'm delighted you're digging the book. I should probably at least namecheck the examples I give in the book, of a few systems that are ubiquitous by any reasonable definition and which are already active. Probably the single most robust such system in the world at the moment is Hong Kong's Octopus card. It's a touchless, RFID-based stored value system that was originally intended as an intermodal transit pass, and has proved to be of such utility that it's been adopted as de facto currency and access control, throughout the territory. Something like 95% of Hong Kong's population between the ages of 16 and 65 has used Octopus, it is used in far more transactions each day than there are people in the city, and you can use it for everything from paying your fare on the Star Ferry that plies the harbor to buying a Coke from a vending machine to letting yourself into your dorm or your office. Octopus has been active since 1997, so no flying cars here. MasterCard almost certainly had their eye on Octopus when they began planning their PayPass system, a similar RFID-based touchless transaction infrastructure that's marketed by Chase (at least here in the Northeast) as Blink. (I should mention here that MasterCard derives real, concrete benefit from the imperceptibility of the PayPass infrastructure, from the way PayPass transactions "dissolve in behavior": their own figures indicate that PayPass users spend up to 25% more per transaction than conventional credit-card users.) PayPass went live at the end of 2005. Another real-world example of everyware that I cite in the book - this one very robust, very elaborate - is the city of New Songdo, a development being brought into existence right now about forty kilometers southwest of Seoul. New Songdo's designers have some very ambitious plans involving everyware, including flooring in housing for the elderly that detects falls and summons emergency assistance, and recycling bins that automatically credit the buyer's account with a small refund when the bottles, cans and containers they've bought properly enter the municipal recovery stream. The Tokyo ward of Shinjuku has just completed a pilot program in which some 10,000 of the ward's lampposts and utility standards communicate with appropriately-equipped mobile phones, to locate the user and describe nearby public accommodations and services. And luxury hotels are already beginning to offer their best guests networked preference files that allow an entire array of in-room environmental, lighting, entertainment and amenity options to be adjusted at the moment of check-in. The Mandarin Oriental chain began offering this, at least in its New York City properties, as far back as the beginning of last year. And most of us are familiar with other RFID-based services around us, even if we weren't necessarily aware of it, from FasTrak and E-ZPass to veterinary ID transponders. So there's a pretty broad spectrum of present-day, real-world, revenue-generating, concrete-as-hell everyware. There are others in the book, still more that I accumulated in the course of my research that didn't make it into the book for one reason or another, and a whole bunch on top of that that folks have pointed me at just in the three or four months since I locked the manuscript down. This stuff is here and now, and beginning to knit itself together in just the way I (and many other folks) had anticipated it would. That's why I think it's half-past time we started mounting an informed collective response.
virtual community or butter? (bumbaugh) Mon 24 Apr 06 06:46
(And a reminder: Readers on the Great Wide Web may query, interpose, edify, or otherwise join the conversation by e-mailing firstname.lastname@example.org to have their queries, interpositions, edifications, or other joinings posted here.)
Jon Lebkowsky (jonl) Mon 24 Apr 06 07:26
> it's half-past time we started mounting an informed > collective response Having written the book, do you see as a next step instigating such a response? Or will you leave that to others?
Adam Greenfield (adamgreenfield) Mon 24 Apr 06 07:32
Well, I expect and hope that mine will be only one of a great many voices comprising the response, but you better believe I'll be out there doing what I can to raise consciousness. I'm imagining that this will something I'll do partly through writing and speaking, partly through organizing conferences and standards bodies (should that latter possibility seem practical), and partly directly, through consulting work.
Jamais Cascio, WorldChanger (cascio) Mon 24 Apr 06 09:31
Despite having suggested the DHS/airport example, I'm inclined to believe that we won't have too many of the opaque denial situations, if only because unexplained barriers tend to provoke people into trying to get around them, which presumably is something that officials would want to avoid. However, what does seem likely is that the system Adam describes would be used to identify No Fly Listees well before they make it to an airline counter. Moreover, the same tech could be used to essentially "tag" people for observation in a wide variety of public settings. Not bar them access, per se, but prompt added attention from security guards and the like. Then there's the version of the scenario that would identify shoppers and target personalized pitches and pricing, already mentioned earlier on. Adam, another question: the "everyware" world is filled with this kind of invisible dedicated technology, but it won't be devoid of general-purpose information tools (e.g., computers as we know them now); how do the two families of information systems link together, in your view?
Adam Greenfield (adamgreenfield) Mon 24 Apr 06 09:49
I think you're right, in that there's no reason for conventional laptops and desktops and mobiles to suddenly disappear from the world just because everyware has started to appear. I do think, though, that if these devices and these familiar form factors survive, it'll be because they afford their users more or less convenient access to the constellation of ubiquitous systems and services around them. As I put it in the book, everyware will "subsume, not supplant" conventional computing. The two modes will coexist in the world for quite some time yet to come.
Christian Crumlish (xian) Mon 24 Apr 06 14:36
i know keyboards suck for many purposes but they're still a pretty good interface for writers of traditional text narratives.
Cogito, Ergo Dubito (robertflink) Mon 24 Apr 06 15:52
Adam, would you comment on the relation of everyware to the experience that humankind has been having in dealing with bureaucracy for some centuries now? Perhaps infotech amplifies the problems as well as the benefits to such a degree that it may be considered a difference in kind?
Adam Greenfield (adamgreenfield) Tue 25 Apr 06 07:18
Christian, agreed, in spades - but I think you'll find the standard QWERTY keyboard instantiated in some surprising ways in the next few years. Bureaucracy, hmm. To be honest, I haven't given that specific subject very much thought. I'd suggest, though, that everyware is a sword that cuts two ways in this regard. On the one hand, as a first pass response, I'd imagine that the vastly enlarged scope of data collection activities implied by everyware tends to proportionally reduce the necessity for large or elaborate bureaucracies - which are, after all, nothing but human data-filtration mechanisms. Let's say, for example, that all of my financial transactions in the course of a year generate a data trail. Not merely the dollar amount, but the context in which that amount of money changes hands. What need is there for a phalanx of accountants, or an elaborate IRS bureaucracy, in this scenario? The government merely has to craft an algorithm that reifies whatever tax code the legislature has passed, and apply it to those transactions. Nominally, such a thing could even be rolling, real-time, right? "April 15th" would no longer be a date to fear and despise. But something in my bones tells me that this is a naive response, even if one accepted the assumptions at its heart. If nothing else, information technology has strewn more hassle through my life than it's obviated - never underestimate the power of an automated customer-assistance line to make you want to rip your teeth out of your head. For sheer Kafkaesque perversity, they can't be beat. I'd rather deal with a human bureaucracy any day of the week. So whereas in theory we could design things such that ubiquitous, ambient information processing provides a much-needed balm for all the hassles private- and public-sector bureaucracy graces postmodern life with...I'd never bet on it actually coming to pass that way. Call me a cynic.
Jon Lebkowsky (jonl) Tue 25 Apr 06 14:13
We can come up with many scenarios for ubiquitious computing, from the obvious to the far-fetched, but perhaps we should discuss constraints that might deter the development of truly ubiquitous systems? For example, the more sources of, and purposes for, messages we conceive, the greater the complexity of the systems for transmission and processing. Aren't bottlenecks and errors inevitable? Or could we handle this with more distributed networks?
Adam Greenfield (adamgreenfield) Wed 26 Apr 06 07:41
I'm certainly no expert, but it seems to me that good old IP networks scale pretty well - and like we've mentioned, the 128-bit address space provided for in IPv6 is so absurdly capacious that I have a hard time imagining we'll ever run out of space. It's also, of course, the case that nothing takes place in a vacuum, so that while we're over here talking about the ubiquity of computing, other folks are wrestling with the sort of thing you mention - distributed processing, grid computing and so on. When you put these two ideas together - networked processors of similar capacity to present-day CPUs, costing about what a dimmer switch does now and similarly strewn through the built environment - you have the potential for some heavily processor-intensive work to be shared between them. Somewhere in the book I have language like "an invisible agora going in behind the walls." It's gonna be interesting. At this point, I want to shift gears a little bit. I'm sure everyone's heard by now that Jane Jacobs passed away yesterday, which was a real loss for all of us who care about the life of the city. And Jane always being a major influence on my thought and my work, I've been a little frustrated by the fact that there doesn't seem to be anyone addressing the challenges of ubiquitous information technology in public space the way she might have. My sense is that, in its way, everyware poses the same sorts of challenges to the quality of urban life that urban renewal did in the 1950s and 60s. So I'd like to talk about the city a bit, get a sense for how people reading this imagine that everyware is going to shape their experience of the urban environment. Mind you, I'm aware that in the preceding paragraphs I've collapsed "urban life," "the urban environment," "the city" and "public space," and I think it's worth pointing out that these are all different things. But make of the question what you will.
Members: Enter the conference to participate