The following is a piece solicited by Volume magazine for their Summer 2011 special on the so-called “Internet of things,” available now at specialist architectural bookstores everywhere. If you’ve been reading weeknotes closely, you may find some of these passages vaguely familiar, for which I apologize.
Given how quickly the social and psychological implications of living in richly networked urban environments are cropping up for so many of us, thinking clearly about the quality of trust — how it is played out and what happens to it — in the smart city seems eminently timely. Yet we stumble right at the outset of any such effort, finding that apparently neat and uncomplicated ideas like “trust” and the “smart city” unpack into rather less ruly tangles than we might have anticipated.
“Trust”? Who, precisely, are the parties implied by this necessarily mutual concept? Are we talking about the feeling two friends might have for one another, the sentiment a customer in a souk might have for a particularly well-known vendor, or that a citizen might repose in the various agencies that have been charged with her protection?
And what is this so-called “smart city”? Ought we confine our considerations to those densely networked urban environments presently being created from little more than landfill, routers, capital and aspiration, in places like South Korea, China, India and the Emirates? Or can the term encompass aspects of life in the places most people already live, the “brownfield sites” that constitute by far the greater share of urban settlement, spatially as well as historically?
Definitional housekeeping is surely a dreary business, but given the risk of confusion it might not be such a bad idea to put some brackets around what we’re talking about. For our purposes here, let’s think of “trust” as the reasonable assumption that initial and subsequent encounters between two parties will be conducted on the basis of a good-faith attempt to seek mutual interest. And by “smart city,” let’s agree that we’re talking about any place where the urban fabric, as well as people and discrete objects within it, have been equipped with the ability to gather, display, store, transmit, receive, and take action upon information.
Armed with definitions of these hard-to-pin-down ideas, we might be a little closer to getting some purchase on the shifting relation between them.
In order to understand the implications for the quality of trust of an urban fabric capable of massive, pervasive and cross-referenced data collection, we might start by considering the impact of present-day data collection in a single register, the visual.
At present, there’s almost certainly no better vantage point from which to explore this question than a street in central London. It’s become a commonplace to note that the UK is the most oversurveilled society in human history, and London its veritable fovea; the city’s 7.7 million residents are resolved by no fewer than one million closed-circuit video cameras, and one widely bruited-about figure suggests that a Londoner can expect to traverse a camera’s field of operation some three hundred times a day.
The number and variety of these devices readily apparent on the streets challenges the unprepared imagination. Some are the conventional, naked box-type; others — an increasing number of these — are modestly domed. Some blindly spool footage onto a remote drive that, unless subpoenaed, will be overwritten in six months, never to be seen by a human eye, while others are clearly and obviously being operated in real time. Some of the most dramatic are those mounted in clusters on municipal masts, morningstars to be wielded against some unknowable but evidently fearsome foe. And each of the municipal cameras, at least, is accompanied by a placard furnishing a justificatory apparatus, thoughtfully specifying its provenance, unique ID number, and ostensible purpose.
But for all that digitized, extended awareness, and for all the placards’ formal promises that cameras are there “for purposes of crime prevention and public safety,” they appear to be having a negligible impact on crime: official Metropolitan Police statistics admit that fewer than one in every thousand street crimes in London is resolved with the aid of CCTV footage. Conversely, the cameras’ advent is demonstrably making Londoners less safe. As Anna Minton observes in her essential Ground Control, the visible presence of a camera in the streetscape makes people less likely to intervene personally in emergent situations, presumably on the theory that Someone Official is observing anything untoward, and will surely intervene.
Thus the likelihood that each of us will get involved in the sort of incidents that happen ten hundred times a day on the streets of any real city — a woman slipping on sidewalk ice, audibly shattering her hipbone; an immigrant deliveryman harassed by a clutch of tweens — that likelihood shrivels. And so whatever remaining social contract governs our dealings with one another, the incentive mechanic operating in the background that underwrites so many acts of consideration, is not-so-subtly undermined. (Perhaps reflecting an inevitable co-evolutionary spiral, the Metropolitan Police also confess that the number of crimes caught on one or another CCTV system has fallen by seventy percent over the past year.) Confronted with this evidence, we might be forgiven for concluding that this “protective” CCTV is a wolf in shepherd’s clothing.
The ways in which the presence of CCTV undercuts the very security it’s supposed to enhance puts me in mind of another site where networked information technology has been folded into everyday urban experience: the “Passenger Information Monitor,” or PIM, the touchscreen interface in New York City taxicabs. As it happens, the PIM also damages the thing it’s supposed to underwrite — in this case, the trust of a paying passenger for a driver.
The PIM is an ostensible convenience that provides GPS mapping and other real-time situational information, as well as a payment interface. Its most obvious and telling quality is the place where municipal statute specifies it is to be mounted: on the rear surface of the protective partition dividing driver from passenger, visible solely to the latter. The result of this siting cannot help but be an informational asymmetry, a differential in access to salient data that only exacerbates whatever issues of mutual mistrust and class, ethnic and linguistic-cultural tension may be latent (or explicit) in the encounter between the two parties.
The placement of the screen carries along with it a not-so-subtle implication that the driver is out to screw the passenger, and if left to their own devices will surely do so. Compare this to the dashboard-mounted GPS navigation systems used by cab drivers in, say, Seoul, which are more clearly there to assist the driver in his or her negotiations with the cityscape — a primary use of such screens which does nothing to prevent their also being used to coordinate agreement between driver and passenger as to appropriate courses of action.
And in what has to be reckoned an extraordinarily clumsy and hamfisted way of undermining any common feeling between the person in the front seat and those behind the partition, the PIM screens run advertising. These ads are predictably loud and irritating, they load automatically and continue running unless manually shut off, and they generate revenue for the taxi operator every time they are viewed. (The passenger is provided with an Off button, but it is designed so as to be relatively obscure and hard to engage.) The cab driver is therefore incentivized to tolerate a system behavior that’s clearly detrimental to the experience of the paying customer. So much for that “good faith effort to seek mutual interest.”
These are design decisions. There is nothing inherently wrongheaded with choosing to site a passenger interface on the back of a taxi’s partition, nor is there necessarily anything wrong with providing the passenger with information that will reassure them as to the wisdom of the driver’s choices.
But in each of the above cases, as a result of bad design, the interests of driver and passenger have been allowed to become uncoupled from one another, with terrible repercussions for their ability to trust and feel comfortable with the other — both locally to this specific ride, and across whatever rides take place in the future, for as long as this particular envelope of technological and design decisions remains intact.
Happily, the presence of information gathering and display in everyday urban life need not lead to such baleful outcomes.
“Trust,” after all, is not the same thing as “faith.” While the construction may seem mechanistic or reductive, trust is at some level the result of a rational calculation, one informed by the best available information about actors’ prior behaviors, their present potentials and intentions, and the possibility space they shape together. We’ve seen how any such calculus on the part of a citizen/user is almost gleefully undermined in both the London CCTV and the New York City PIM cases, ruled out from the very start by design choices that deny them access to the information such a reckoning requires.
By contrast, the work of my own practice, Urbanscale, is animated by an underlying belief that the timely provision of information to the citizen is both in itself an act of respect, and the kind of offering that can lead to more robust and resilient feelings of trust between people and the cities they live in.
Consider something that happens hundreds or thousands of times a day in any city where public transit uses RFID-based fare-payment cards. (Hong Kong’s Octopus, London’s Oyster, or the Clipper card recently launched in the San Francisco Bay Area are all examples.)
The current generation of technology offers riders no self-evident way of knowing how much value remains on a given card before they’re actually at the turnstile, or boarding the bus. This leads to an unpleasant situation many of us have been unlucky enough to experience, at least occasionally: you’re already moving through the turnstile, in anticipation of bus or train (and maybe you can even see it, about to depart without you) when you’re brought up short by insufficient fare or an expired card.
This causes three kinds of trouble. First, and not insignificantly, it’s simply awkward and embarrassing; nobody likes to inconvenience others, especially when those others are impatient big-city commuters, lined up close behind you. The delay of having to backtrack to a payment machine, refill the card and tap through again might cause you to miss the next bus or train, whether or not you can afford to miss it. Finally, at peak hours, the turbulent flow this circumstance sets up at the turnstile can materially impact a station’s throughput, its optimal ability to handle passenger load, with real economic consequences.
Again, as with the PIM, the problem’s nothing inherent in the technology itself. The hassle experienced at the moment of use is simply a consequence of thoughtless design decisions, and perhaps of institutions that lack a clear understanding of how information might be used as a material in design. But what if there was a way for a rider to know how much remained on their card before they hit the turnstile? What if that information was presented in a friendly way, at a sufficient size that one could read it at a glance from across the room?
This is the brief for the enhanced transit payment card we call Farevalue, which laminates an e-ink panel (similar to that on a Kindle) onto an otherwise ordinary RFID-based farecard. The provision of that one piece of information, couched in carefully crafted, natural-language statements (“There is $19.15 remaining on this card, which is enough for about eight rides”) ought to be enough to let people make better decisions about how, when and under what circumstances they want to use the card — the crux of the trust calculus we believe is operating here, and in virtually all circumstances in which ordinary people engage urban-informatic systems.
Though we obviously hope it does succeed as a commercial proposition, in the long term Farevalue’s fortune in the market is less important than how well it functions as propaganda of the deed. We want it to furnish existence proof that confidence-eroding informational asymmetries aren’t necessary qualities of the networked city, and can be mitigated or eliminated entirely from the user experience by a thoughtful process of design. Our wager with Farevalue is that a relatively minuscule informational intervention — amounting to a single line of copy, presented in the right voice, in the right place and time — has disproportionate power to transform our encounters with the pervasive networked infrastructure that now undergirds so much of urban life.
If the slow colonization of daily experience by informatic systems can sometimes seem like a fait accompli, though, we should be very clear that there are competing visions of its becoming. Ultimately, as it happens, I’m not particularly interested in the clean-slate “smart cities” so many enterprise-scale technology vendors want to saddle us with, but in designing better, more empowering and more widely accessible interfaces to the places most of us already live: a layer of information to help us make wiser decisions, as individuals and communities.
The lesson of artifacts like CCTV and the New York City PIM, though, is that design that enhances our ability to place trust in one another does not happen by accident, and is not likely to occur at all if not explicitly valued and taken into consideration early in the development process. (This is most especially true in domains like information technology, where products are more likely to be engineered than designed per se.)
If, as my favorite go-to McLuhanism suggests, “every extension is also an amputation” — an observation that strikes me as worth recalling to mind in these days of algorithmic trust engines and distributed reputation services — we need to ensure that the mantle of performative augmentation newly afforded us by the networked city doesn’t undermine the very qualities we otherwise strive to achieve in our design of place.
More than ever, as our cities become still denser, as the diversity of those we encounter in them grows along every axis, as a larger population competes for a shrinking pool of resources and the world confronts us with reduced room for error in all our calculations, we can’t afford to live in places where important aspects of the built environment undermine our ability to rely upon each other. What is the relationship between trust and the smart city? It is that any meaningful hope for the latter hinges entirely upon our understanding the processes that give rise to the former.