Please note: This essay has now moved over to Medium
I was recently asked to comment on ‘the street of the future’; a response for a quango responsible for the built environment and a government department responsible for transport, roads and so forth. Which means it's really the street of the near-future. I didn’t have enough time to write something short, so I dashed off the following, and I’m really posting here as a note to self, rather than an attempt to deeply discuss the everyday informational street circa 2008. Still, I hope you find it useful or engaging. The photos don't relate directly but create a kind of composite illustrative city nonetheless.
It’s deliberately grounded in the here-and-now, more or less, so it will seem rather old hat to some of you. Which in a sense it is. And in another sense, it isn't. But either way, this was a better strategy for the task-in-hand, and in imagining the scene below, via a kind of narrative, it's still remarkable to even sketchily consider how much data is already around us, and is near-invisible to traditional urban planning perspectives. And I'd suggest that this data beginning to profoundly affect the way the street feels. Some quick analysis follows the narrative, raising a series of questions for governance, legislation and the public-private partnerships that also constitute the contemporary street.
The street as platform
The way the street feels may soon be defined by what cannot be seen with the naked eye.
Imagine film of a normal street right now, a relatively busy crossroads at 9AM taken from a vantage point high above the street, looking down at an angle as if from a CCTV camera. We can see several buildings, a dozen cars, and quite a few people, pavements dotted with street furniture.
Freeze the frame, and scrub the film backwards and forwards a little, observing the physical activity on the street. But what can’t we see?
We can’t see how the street is immersed in a twitching, pulsing cloud of data. This is over and above the well-established electromagnetic radiation, crackles of static, radio waves conveying radio and television broadcasts in digital and analogue forms, police voice traffic. This is a new kind of data, collective and individual, aggregated and discrete, open and closed, constantly logging impossibly detailed patterns of behaviour. The behaviour of the street.
Such data emerges from the feet of three friends, grimly jogging past, whose Nike+ shoes track the frequency and duration of every step, comparing against pre-set targets for each individual runner. This is cross-referenced with playlist data emerging from their three iPods. Similar performance data is being captured in the engine control systems of a stationary BMW waiting at a traffic light, beaming information back to the BMW service centre associated with the car’s owner.
The traffic light system itself is capturing and collating data about traffic and pedestrian flow, based on real-time patterns surrounding the light, and conveying the state of congestion in the neighbourhood to the traffic planning authority for that region, which alters the lights’ behaviour accordingly. (That same traffic data is subsequently scraped by an information visualisation system that maps average travel times on to house price data, overlaid onto a collaboratively produced and open map of the city.)
In an adjacent newsagent’s, the stock control system updates as a newspaper is purchased, with data about consumption emerging from the EFTPOS system used to purchase the paper, triggering transactions in the customer’s bank account records.
Data emerges from the seven simultaneous phone conversations (with one call via Skype and six cellular phones) amongst the group of people waiting at the pedestrian crossing nearest the newsagent.
The recent browser histories of the two PCs with internet access in a coffee shop across the road update sporadically with use, indicating both individual patterns of websites accessed and an aggregate pattern of data transfer throughout the day. At the counter of the coffee shop, a loyalty card is being swiped, updating records in their customer database. The flat above the shop is silently broadcasting data about the occupant’s usage of his Sky+ box, DAB radio with internet connection, and Xbox Live console. His laptop noisily plays music, noiselessly accreting data to build a profile of the user’s taste in music at the web-based service Last FM. This track has inaccurate or no metadata, which means it is not registered by Last FM, in turn harming its latent sales prospects.
A police car whistles by, the policewoman in the passenger seat tapping into a feed of patterns of suspicious activity around the back of the newsagent on a proprietary police system accessed via her secured BlackBerry. A kid takes a picture of the police car blurring past with his digital camera, which automatically uses a satellite to stamp the image with location data via the GPS-enabled peripheral plugged into the camera’s hot-shoe connection.
Across the road, a telecoms engineer secures a wireless device to the telephone exchange unit on the pavement, which will intermittently broadcast its state back to base, indicating when repairs might be necessary.
Walking past, an anxious-looking punter abruptly halts as the local Ladbrokes triggers a Bluetooth-based MMS to his phone, having detected him nearby, and offers discounts on a flutter on the 3.30 at Newmarket (the Ladbrokes is constantly receiving updates on runners, riders and bets, linked to a national network aggregating information from local nodes at racecourses and bookies). The potential punter had earlier received a tip on said race from his chosen newspaper’s daily sports bulletin, delivered via his mobile’s newsfeed reader software.
As he wonders whether he could discreetly sidle into the bookies to place the bet he’d promised himself he wouldn’t, the street-lamp above his head fades down as its sensors indicate the level of ambient daylight on the street is now quite sufficient, switching into a mode where the solar panel above collects energy for the evening and delivers any potential excess back into the grid, briefly triggering a message indicating this change of state back to the public-private partnership that runs the lighting services in this borough, in turn commencing a transaction to price up the surplus electricity delivered to the grid.
The same increase in daylight causes a minor adjustment in four of the seven CCTV cameras dotted along the street, as they re-calibrate their exposure levels accordingly, the digital video accruing on the array of remote disk drives at a faraway control centre is rendered slightly differently in response.
In an apartment over the bookies, the occupant switches on her kettle, causing the display on her Wattson device that monitors real-time electricity usage in the flat to jump upwards by a hundred watts, whilst triggering a corresponding jump in the sparklines displaying usage on the Holmes software that tracks that data over time, and which compares her consumption to four of her friend’s houses in the same neighbourhood.
Three kids are playing an online game on their mobile phones, in which the physical street pattern around them is overlaid with renderings of the 19th century city. They scuttle down an alleyway behind a furniture showroom as the virtual presence of another player, actually situated in a town forty miles away and reincarnated as a Sherlock Holmes-ian detective, indicated on their map by an icon of a deerstalker and gently puffing pipe, stalks past the overlaid imagined space. The three play a trio of master criminals, intent on unleashing a poisonous miasma upon the unsuspecting and unreal caricatures generated by the game.
Approaching the furniture showroom’s delivery bay to the rear, the driver of an articulated lorry grinds down through his gears in frustration as he realises the road over the lights narrows to a point through which his cab will not fit, information not made clear by the satnav system propped on his dashboard. The RFID chips embedded into the packaging of the seven armchairs in his trailer register briefly on the showroom’s stock control system, noting the delivery time and identity of the driver. When formally checked in later by the showroom’s assistant with a barcode-scanner, the damage to one of the chairs is noted and sent back to base, automatically triggering the addition of a replacement armchair to the next lorry out to this town, while recalculating stock levels.
In the shoe-shop next door, a similar hand-held scanner, unknowingly damaged in a minor act of tomfoolery a day earlier, fails to register the barcode on a box of sneakers, resulting in a lost sale as the assistant is unable to process the transaction without said barcode. The would-be customer walks out in disgust, texting his wife in order to vent his furious frustration on someone. She sends a placating if deliberately patronising message back within a few seconds, which causes him to smile and respond with an ‘x’ two seconds after that. In doing so, his allocation of SMSs for the month tips over to the next tier in his payment plan, triggering a flag in an database somewhere in Slough.
Deciding to spend his money - that he unwittingly has less of than he did a few moments ago - on a book instead, he steps into the only local bookstore on the street, using the now more expensive data plan of his mobile phone service to retrieve aggregated reviews for the latest Andy McNab, which he half-reads whilst perusing the back cover of the book. Unfortunately the corresponding prices offered up by the review system are in US dollars, as the service is not localised and thus he can’t compare prices. This is fortunate for the shop, however, and so during the resulting purchase of the book, the store’s stock control system automatically orders a fresh batch of the now best-seller whilst the on-counter top 10 displays McNab’s seemingly inexorable rise up the charts on a battered old LCD monitor.
Round the corner, the number of copies of the McNab book in the municipal library remains exactly the same. Instead, the large external LED display hoisted over the door at huge expense conveys the volume of ISBNs of books being swiped by librarians inside the building, in real-time. Part of an installation by students at the local art college, the most popular genres of books taken out, inferred from the aggregate of ISBNs and cross-referenced with Amazon, are displayed every five minutes via a collage of randomly-selected movie clips from YouTube that match broadly that same genre and keywords (filtered for decency and sensitivity by bespoke software which is itself receiving updates, detailing what is considered obscene at this point). Currently, a 2-second sequence of a close-up of David Niven’s nose and moustache from The Bridge Over The River Kwai morphs into the bulging right arm of Sylvester Stallone in Rambo, cradling a stolen Soviet rocket launcher. The patterns of clip consumption at YouTube twitch accordingly.
Looking up at the display in fascination and bewilderment, an elderly lady stumbles over a pothole in the pavement. Helped back to her feet by a younger man, she decides to complain to the council about the pothole. The man suggests he can do that right now, from his iPod Touch and using the library’s open public wifi, by registering the presence of a pothole at this point on the local problems database, Fix My Street. The old woman stares at him quizzically as it takes him fifty seconds to close the website he had been looking it on his mobile (Google Maps directions for “hairdressers near SW4”, a phrase he’ll shortly have to type in again, having neglected to bookmark it) and access fixmystreet.com. He spends the next few minutes indicating the presence of a pothole outside the library on Fix My Street (unaware of the postcode, he has to select one from a few possible matches on street name), before he moves on, satisfied with his civic good deed for the day. The elderly lady had long since shuffled off, muttering to herself. Although Fix My Street smartly forwards on all issues to the corresponding council, a beleaguered under-trained temp in the also underfunded 'pavements team' is unaware of fixmystreet.com and unable to cope with the levels of complaint, and so the pothole claims five more victims over the next two weeks until someone rings up about it.
The LED display board can also sniff what is being accessed via the library’s public wifi network, and displays fragments of the corresponding text and imagery. It switches briefly over to this mode, in order to denote that Fix My Street was being accessed, and displays some details of the transactions detailing the pothole issue. Before flicking back to the YouTube x ISBN installation, the display then conveys some information from the local council about a forthcoming street upgrade, blissfully unaware of the possible connection to be made between that and the pothole. Unfortunately, at that point, the pale sunlight hits the screen at such an angle that it cannot be read by two hurrying passers-by anyway. The display then dissolves into a slow pan across Keira Knightly’s delicately arched eyebrow from Pirates of the Caribbean.
In the swinging briefcase of one of the passers-by, an Amazon Kindle e-book reader briefly connects to the public library - having previously visited the library, the owner had registered the public wifi in her settings. It commences a rapid-fire series of handshakes with Amazon’s systems, swapping personal details back and forth with user profile information, and thus beginning to download a new book by Ian McEwan to the device. Despite the wealth of metadata in this rich stream of data, the Kindle’s closed system means that the library’s databases, and LED display installation, cannot possibly be made aware of this literary transaction being conducted using its infrastructure. Either way, with seven seconds the Kindle user is out of range and the download automatically fizzles out, settling back to wait for the scent of open wireless.
Behind the library, a small 19th century cottage that’s been on the market for a year now is being re-valued by estate agents. This new figure, a few thousand pounds less than the previous, is entered automatically via the estate agent’s PDA and ripples through their internal databases and then external facing systems. It doesn’t trigger any change in three other proprietary databases listing average house prices in the neighbourhood until three weeks later. This house price change subtly affects the average for the area, which is later recombined into the aforementioned map that compares with commuter times for the borough.
An employee of the local water company knocks on a door up the street, calling in order to take a reading from the house’s meter. She uses a bespoke application on her mobile phone, which should indicate the location of the meter on the property. In this case, it doesn’t, so she has to ask.
Five TomTom satnav systems in five of the twelve cars on the street suddenly crash for reasons unknown, causing an instantaneous reboot and login sequence over the course of twenty seconds. One driver notices.
The four other drivers are slightly distracted by the glow of a giant TV screen, installed and operated by the council but paid for through corporate sponsorship, which glowers over the end of a pedestrianised-shopping mall at the end of the street over the lights. It’s broadcasting the Australian Open tennis, which is being played live in Melbourne. A homeless person is sleeping underneath the screen, soaking up some of its transmitted warmth. An on-street information kiosk stands beside the screen, offering a scrollable map of the local area and directory of local businesses. It’s little-used, as the directory of businesses was always incomplete and intermittently updated, its data now rusty and eroded by time. Plus maps are available on most people’s mobile phones. Still, the printer installed in its base occasionally emits a money-off coupon for some of the local businesses.
Under the pavement on one side of the street, a buried sensor records the fact that some fibre-optic cables are now transmitting data with 10% less effectiveness than when they were installed. A rat ascends from an accidentally uncovered grille under the library’s down-pipe nearby, its whiskers containing trace elements of plastic cladding.
A blogger posts an entry on her weblog regarding some new graffiti on the library’s rear, uploading the image via her mobile phone, thanks to her blog platform’s relationship with Flickr, a popular photo sharing site. She adds a cursory description of the stencilled representation of the Mayor’s face superimposed onto a £50 note instead of the Queen’s. Shortly afterwards, she receives an SMS from the service Twitter, indicating that two of her friends are heading for a café up the street, and she decides to intercept them to discuss her find, sending back the URL of her post and the time of her imminent arrival. Her phone’s Google Maps application triangulates her position to within a few hundred metres using the mobile cell that encompasses the street, conveying a quicker route to the café. Unfortunately, none of their systems convey that the café is newly closed for redecoration.
Working from home in his small house backing onto the old cottage, a lawyer files his case notes via the password-protected intranet his company operates, his wifi network encrypted to prevent leakage of such confidential data. He then closes his network connection, switching instead to his neighbour’s wifi network - which has been left purposefully open in the interests of creating a cohesive civic layer of wireless coverage on their street - in order to watch the watch the highlights of his football team’s two-nil victory the night before. In this way, his own remarkably cheap wireless network data plan never goes beyond its monthly cap. This parasitic wireless activity is only curtailed months later, when the previously benevolent neighbour uses some free sniffer software she downloaded to detect the presence of the wifi router that’s responsible for the majority of the data usage in the street.
A local off-license has an old monitor in the window that cycles through a series of crude screen-grabs of faces of shoplifters of local stores, derived from the various CCTV systems owned by a local association of shopkeepers. Unfortunately, the face of the purchaser of the Andy McNab book is mistakenly added to the system three weeks later.
(Coincidentally, in a meeting being conducted several miles away, a project team working on council tax systems briefly considers whether a system of localised screens displaying which houses in the street had not paid their council tax yet, updated wirelessly, would be ethically sound.)
Waiting at the lights, someone pays their council tax by mobile phone, triggering an internet-based bank transfer via SMS. Across the road, a car belong to a city-wide car-sharing network patiently waits to be activated by a swipe of a member’s RFID card. It transmits its location and status back to the car-sharing network’s database every few minutes.
Also in a prime position by the lights, a café is briefly office to two businesspeople having an informal meeting. Although the café’s wireless network is closed, their usage charges are paid for by the company they work for, and they barely notice the cost. The company credit card details are retrieved automatically over a secure transaction. Though it has poorer muffins, the café opens 90 minutes earlier than the library.
A series of small high-resolution displays, hanging under each traffic light and angled towards stationary drivers, alternately communicates the number of accidents that have occurred around these lights in the last year, and then the current speed limit, which can be calibrated to an optimum level for the current traffic conditions in the borough. The traffic lights also house the city’s congestion charging system’s cameras, logging the number-plates of cars passing through its network of inner-city streets.
A wireless sensor network, carefully and discreetly embedded in the trunks of trees lining one side of the street, silently monitors the overall health of the limes and planes, collating data and waiting patiently for the council’s tree surgeon to inspect the arboreal vital signs.
At the end of the line of trees, a new bench has been installed on the street. At either side of the bench, there are two standard electrical power-points freely available for citizens to recharge their phone or laptop. A small LED winks to indicate this, alongside a standardised explanatory icon drawn up by the department also responsible for the highways’ signage systems. The power running to the bench is carried via flexible cables that can twist and stretch around the growing roots of the nearest trees. The bench also carries a WIMAX transmitter as part of a research project led by the local university. As such, this bench appears as a key node on several GIS.
A cab drives through the traffic lights as they switch to green and it quickly signals to turn left, looking to nose back on itself as the presence of a fare is indicated at a nearby hairdressers, via the in-taxi control system. A faraway voice crackles over the intercom a few seconds later attempting to verify that the driver is en-route. The driver clarifies she is en-route but that she’ll take a few minutes more than usual as her satnav system indicates high traffic levels across the three normal routes taken.
At another building on the street, a new four-storey commercial office block inhabited by five different companies, the building information modelling systems, left running after construction, convey real-time performance data on the building’s heating, plumbing, lighting and electrical systems back to the facilities management database operated by the company responsible for running and servicing the building. It also triggers entries in the database of both the architect and engineering firms who designed and built the office block, and are running post-occupancy evaluations on the building in order to learn from its performance once inhabited.
In turn, and using this feed, the city council’s monitoring systems note the aggregated energy usage for the commercial buildings on the street, constantly shuffling its league table of energy-efficient neighbourhoods. The current score for the street, neighbourhood and city is displayed outside the nearby library, on a trio of vertical axis wind turbines with LEDs embedded in their blades.
A prototype of a similar monitoring system, but embedded in the bus-stop opposite the library, records the performance of the lights, travel information displays, large plasma-screen advertising display, and the chilled-beam cooling system newly installed for comfort. The travel information displays themselves receive updates in real-time via a slice of radio spectrum allocated to such data, indicating the proximity of the next five buses. This same system also conveys the latest information on the whereabouts of the no. 73 in particular, in the form of an SMS to a prospective passenger who has selected this as her ‘favourite bus’ via the transport company’s website. Around the corner, she breaks into a trot accordingly.
The plasma display is currently running an advert for the local radio station’s breakfast show (displaying a live stream of SMS messages sent to the show, filtered for obscenity and likelihood of libel). As the slightly out-of-breath imminent passenger arrives within range of its Bluetooth-based transceiver, it cross-fades to a display from the city’s premiere modern art gallery, with whom she has registered her mobile phone as a preferred mode of communication and whose systems are quickly cross-referenced for her attendance record for the last few years, and thus it informs her of a new exhibition about to start.
This she doesn’t notice at all, but one person in the loosely-defined queue around the bus stop does, and scribbles the details on his hand. Four seconds later, the display recognises another mobile phone with an open Bluetooth connection and an active account within the agglomeration of companies that have registered their databases with this advertising service, and shifts its display accordingly. The call-and-response between the queue and the screen continues until the bus finally pulls in and the screen’s transient audience dissipates. It settles back to a carousel of generic advertising messages and local information tailored to that street and its surrounds.
As the bus departs, the new passengers on-board swipe their RFID-based integrated transport system ID cards, updating mass transit databases with every possible aspect that can be gleaned from this simple activity (time of day, location, frequency of use, favourite entry points etc.) The now-empty seat in the bus-stop registers that it is indeed now empty using simple sensors, and wirelessly logs this fact with a database monitoring the usage and state of street furniture in the neighbourhood. Powered by solar panels on top of the bus-stop, it creates a pulsing ambient glow.
Across the road, another billboard displays the number of reported burglaries and bag-snatches in the neighbourhood in the last three months, live data direct from the police force systems. This causes several passers-by to feel a touch more anxious than they did a moment ago. Had they walked past a moment before, the billboard would have been displaying information on a forthcoming community sports day at the local park. One of the passers-by would have recognised their son in the video of last year’s winners, running in slow motion under the crisp typography. A moment before and the passers-by would have been subjected to a tortuous promo for a Portuguese avant-garde play currently running at the local theatre, within which a QR code displayed in the top-right hand corner could’ve been read with a mobile phone’s IR reader, delivering the website for the theatre to the phone’s browser.
Of the two bars, two pubs and three cafés on the street, only one has recently checked that the location and description data overlaid on Google Maps is present and correct, and thus is fortunate to receive the custom of two hungry Hungarian tourists for a full English breakfast with all the trimmings.
Twenty metres below the ground, a tube train scurries under the crossroads, outrunning its halo of data that details its location and speed from the engine control systems, while CCTV conveys images of the carriage directly underneath. The carriage contains forty-four mobile phones seeking a signal, some with Bluetooth headphone sets; ten BlackBerries and four other PDAs likewise; thirteen mp3 players of varying brands, a couple also with Bluetooth headphones; seven sleeping laptops.
Directly overhead, ten thousand metres up, the distant roar of a commercial airliner’s Rolls-Royce engines, beaming their performance data back to engineers via satellite in real-time …
And press play ...
This somewhat banal sketch of an average high street is very deliberately based on the here and now; none of the technology lurking in the background behind this passage is R&D. Most of it is in use in our streets, one way or another, and the technology that isn’t could be deployed tomorrow. As such, given the time from lab to street, it represents the research thinking of over a decade ago.
This doesn’t even go near concepts like Bruce Sterling’s spimes, or The Living’s Living City, or Christian Nold’s biomapping, in which objects, people and buildings constantly, silently and invisibly communicate with each other, shaping each others’ behaviour and representation.
And arguably, this still underestimates significantly the size, shape and intensity of the data cloud immersing the street - it’s the tip of the iceberg. Moreover, in centring on a snapshot, it doesn’t convey the ebb and flow of systems and data over time. It’s not even particularly contrived - there are no sketches of, say:
- ‘an urban planning student measures the varying wifi signals up and down the street for her research into the informational city’ or,
- ‘three documentary film-makers deployed by the council film slow-motion sequences of pedestrians’ feet crossing the road, later digitised as part of a multimedia portrait of the neighbourhood’ or,
- ‘a dysfunctional teenager hacks into the bookie’s systems, after a morning of being ignored by his parents, and proceeds to swipe the credit card details of the four old men inside, later publishing them on the internet after placing a frivolous selection of bids at eBay in one of the old men’s names’ or,
- ‘a writer denotes the ghostly presence of a 12th century market using psychogeographical markup language’ or,
- ‘for fun, a bored intern at an urban planning consultancy drives simulated herds of cattle through a digital model of the crossroads designed for predicting the patterns of behaviour in crowds during a terrorist attack’ and so on.
Instead, this is all everyday technology - embedded in, propped up against, or moving through the street, carried by people and vehicles, and installed by private companies and public bodies. Each element of data causes waves of responses in other connected databases, sometimes interacting with each other physically through proximity, other times through semantic connections across complex databases, sometimes in real-time, sometimes causing ripples months later. Some data is proprietary, enclosed and privately managed, some is open, collaborative and public.
Yet how much of this activity is obviously perceptible on our streets when viewed through conventional means? The snapshot above, without the explanatory narrative of the systems being touched by these activities, would just like a freeze-frame of a few people and vehicles set against a backdrop of buildings. A photograph or drawing would show only a handful of people, a few vehicles and some buildings. Traditional urban planning might note patterns of flocking or grouping, when tracking the flow of people through a space, yet would they make a causal observation based on the presence of the open wifi that created a ‘flock’?
Forty years ago, the British architects Archigram suggested that “When it’s raining on Oxford Street, the buildings are no more important than the rain”. The group’s David Greene subsequently asked “So why draw the buildings and not the rain?”. Why indeed? The sketch above tries to describe data rather than rain, but they’re similarly ephemeral. The work of Archigram, and others, may provide some useful prompts for thinking about this softer infrastructure - when critiqued - though Greene himself has noted that the importance of apparently permanent buildings has persisted, even for “the electronic nomads of the global financial systems”. So the more relevant question is how do the buildings and the rain of data interrelate?
Informational systems are beginning to profoundly change the way our streets work, the way they are used, and the way they feel. This in itself presents a major challenge for the existing practice and vocabulary of planning. How much of this life of the street, this rapidly increasing torrent of human activity, is registered as a field of enquiry or activity in most planning activity? Imagine this street scene over the next few years of deployment of a more ubiquitous and pervasive computing, and the challenge to identify, understand, denote and plan for this environment is even more pressing.
Considering the non-visual senses might be a better analogy when it comes to perceiving the way data affects i.e. looking at the way the street sounds, feels or smells. References here would include Finnish architect Juhani Pallasmaa’s book The Eyes of the Skin, and the collection Sense and the City edited by Mirko Zardini (particularly geographer Stephen Flusty’s work on new descriptions of spaces, such as ‘stealthy spaces - that cannot be found’, slippery spaces - spaces that cannot be reached‘, ‘prickly spaces - that cannot be comfortably occupied’, ‘jittery spaces - that cannot be utilised unobserved’, and so on). The somewhat esoteric world of psychogeography - the layered informational and psychological history and geography that conveys a sense of place - might be another reference point. Sound could be a particularly useful analogy, as data has similar characteristics, and is at least partly understood and considered by existing policy, even if it generally focuses on noise abatement, rather than the potential of encouraging what some research is calling ‘positive soundscapes’.
So the patterns of data in the streets, the systems that enable and carry them, the quality of those connections, their various levels of openness or privacy, will all affect the way the street feels rather more than street furniture or road signs. Holes in data, public and private, may become more relevant than the pothole in the pavement - until you trip over it, at least. (This latter point isn't frivolous - for we should be aware of the limits of information services, until made physical. Either from a phenomenological point-of-view, or from the view that just says these systems tend to be transient, it's important to keep a sense of perspective.)
The sketch above deliberately traverses quite a few modes of activity - from private to public; individual to civic; commercial to recreational; residential to vocational. And in all instances systems are in flux, in development, or require implementing, testing and shaping.
In many of these instances there are decisions to be made about openness, responsibility, privacy, security, interaction, experience. Some of these will be directly under the aegis of government, some through public-private partnerships, some though architects of the built environment, some through architects of this informational environment, some through commercial enterprises, some through NGOs, some through municipal institutions, some through education, some through individuals or community groups, and so on.
There are decisions to be made about raw infrastructure - the equivalent of transport networks and power supply. Should the street be enabled by fibre optic, copper wire, WIMAX, 3G, 802.11x wifi, and so on. How close does that run to the curb? Who will install and operate? How smart should street furniture be? How should it convey itself to the world? Should it expose its seams, to aid understanding and engagement, or withdraw silently into the woodwork, reducing clutter and complexity? What would happen to the flow on the street if the street furniture attracted clusters of people within its halo of connectivity?
Without this infrastructure, the street only half-exists, becoming a residual dead-zone in the city. And yet should areas on the street deliberately be dead-zones, shielded from connectivity in order to provide respite, reflection, quietude? How is that to be managed and conveyed?
Car engines may be limiting their speed themselves, via GPS - should traffic control systems have some say in their real-time calibration too, to ensure optimum speed levels for the city? As control and monitoring systems become pervasive, how should the relationship between private and public infrastructure, behaviour and legislation respond? In terms of public transport, car-sharing schemes and real-time congestion pricing, are mesh networks the right way to go?
As the location and timing of work and leisure becomes ever more entwined and slippery, how should the infrastructure and shape of the street respond to this? How to deal with legacy issues, like the defunct kiosk mentioned earlier, given the rate of change in technology and cultural understanding of technology? Are leasing or ownership models appropriate for the hardware and for the software?
Who is responsible for the representation of the street on the various informational systems underpinning the above sketch? Or, conversely, the representation of those informational systems on the physical street? Who looks after the digital equivalents of ‘Blue Plaque’ heritage signs, in which the layered history of the street can be explored - do they need a form of planning permission?
What is the power consumption of this street? As Buckminster Fuller might have said, “how much does this street weigh?” The displaced energy consumption of the street, distributed on to servers and power sources all over the world, could be immense. Conversely, much of the connectivity may rely on a handful of key undersea cables providing high-speed bandwidth to the city. These will occasionally convey a sense of fragility in the network, perceived but not comprehended by users. As the scenario is largely here and now, it appears unaffected by the issues of ‘peak oil’ and the different kind of supply that renewable energy sources might entail. Any future scenario planning needs to consider that aspect in particular rather more carefully.
In each case, who is best qualified to deliver all this?
The complex interplay of these fibres will help define the sense of the street, and planners first of all need to identify this rich, sometimes slippery network, before going on to decide where it can be shaped and where it should be shaped.
Note: It’s also worth bearing in mind that complex software models can convey some of this data fairly exactly, yet still not tell us about the cultural memory of the street. This system is in turn a part of other, more complex systems of behaviour and representation.
There are two caricatured possible futures that can be deployed to flush out a few more issues.
1. Locked down street
The depiction of the highly-privatised data environments which constantly reach out to potential consumers is familiar from numerous hackneyed science fiction plot-lines. For example, the now infamous and faintly ludicrous scene in Minority Report where the protagonist is assailed by highly targeted ads as he walks through a shopping street.
Systems that are focused around a user’s private data, then played back to them with little or no chance of opt-out, are likely to emerge nonetheless. This we can infer from current web-based systems. Spam filling an in-box is debilitating enough without the physical urban environment adding to the problem. Social software systems rarely let you cleanly unsubscribe. Thus the call for secure privatised systems for advertising display, as in the bus stop sketch above, will be strongly made by companies claiming to prevent irrelevant or misleading messages polluting the physical environment.
Equally, many of the scenarios above indicate highly personal data - browser usage patterns, credit card details, traffic control systems and so on. So it’s clear that the street will be a platform for highly secure data transfer and proprietary systems. Personal data security will be an increasing focus for individuals and service providers, as the often haphazard approach to personal data will not be allowed to continue, through regulation and compliance (cf. the UK government’s recent mishandling of social security records). Equally, systems with potential to reduce public safety if misused - the traffic light controls above, for example - will need to be carefully regulated and deployed as highly secure systems.
The intriguing area, however, is around systems in which personal data is used more openly. Where patterns of behaviour are played back to people, as per current social software systems like Facebook, Last FM, Flickr, Twitter etc.
Judging by uptake of this social software over the last few years, it seems the average user cares little for privacy of much personal data. Look at Facebook et al, and you’ll see an awful lot of personal data given away. Users can play fast and loose with the veracity of this data, but one can assume that a fair proportion is accurate. However systems can overstep the mark. Witness the response when Facebook did try to deploy a highly targeted advertising model based around personal data and that of users’ friends; the resulting furore meant that Facebook had to withdraw the functionality, at least temporarily and at least in that form. With commercial proprietary systems based around social data, a desperate reliance on the users’ goodwill and ongoing patronage can mean a self-moderating system to some degree. If a proprietary system is perceived as using data inappropriately, when re-using it on its users, it’s likely to be subject to resistance.
So the general movement within social software systems - which we might be able to take as canaries in the coal mine for wider patterns to some extent (though it might be very different when physical) - is towards an openness in terms of recombining data. An optimistic viewpoint perhaps. (And one solely derived from observing that social software written under a largely Californian ideology, whose mileage may vary massively in different urban cultures.
Equally, closed systems will rarely be able to take advantage of the aggregated patterns of data that could emerge from a neighbourhood. For instance, a concept like the Personal Well-Tempered Environment relies on the various sensors and meters implanted in energy-monitoring devices being able to communicate with one another over open standards, irrespective of who constructed, distributed or runs the system. Proprietary systems, while ideally suited to high-security purposefully-closed networks, are intrinsically unlikely to enable a form of creative aggregation and connection unintended by the owners and makers. If BMW’s ConnectedDrive system was open and shared by other manufacturers, imagine how useful elements of that automobile engine data might be to transport and environmental planners too?
So as the street begins to produce data on a vast scale, locking all of it down seems both impossible and counter-productive, as a suspicion of using such data inappropriately, well-founded or not, will lead to resistance.
Moreover, the locked down street, based around centralised proprietary systems, can be surmised to struggle with localisation. (See the ‘price finder in US dollars’ example above.) Intrinsic local detail will tend to require collaborative updates from local users themselves. Broader platforms that draw on such openly produced collaborative efforts (cf. Google) may well continue to be popular - note how Google Maps conveys the street - but is dependent on the kindness, and quality, of strangers.
Crucially, better GIS data (across multiple independent systems) will become fundamental to the way the street is articulated. Omissions in data already lead to issues for British streets - note how some articulated lorry drivers end up careering through small country towns, their GPS systems had leading them there to avoid traffic, only to discover that the roads are too small for their trucks. A street that does not accurately convey its state to such systems is in danger of being ignored or inadvertently misused. An open approach to descriptive data can enable a far richer local dataset to emerge, more accurately conveying the sense of what the street is. It may also be patchy, however.
Given the commercial environment most of this technology is developed in, it’s likely that multiple, fragmented and disconnected systems will embed themselves in the increasingly responsive and interactive façades and fabric of the near-future street. This may mean familiar issues, but now manifest in the built environment:
- Frustration of data locked in one system that can’t be transferred to another;
- Systems unaware of other systems, and thus do not enable useful connections to be made (physical or digital);
- Poorly-designed systems which inadvertently convey too much personal information;
- Tracking of patterns of behaviour that are not made evident to the user;
- Systems that convey information poorly;
Amongst others. This commercial development is a given, and often not related to those whose job it is to shape streets.
Yet traffic systems, signage and display systems, mass transit systems, municipal buildings, street furniture, landscaping, the quality and typology of infrastructure laid down, public databases and information systems, acquisition and communication of community information, and encouragement of commercial activities through levies, grants and public-private partnerships are all examples of wider civic concerns for those who do have to shape streets.
Planners and architects have to decide, and soon, whether they want to have a say in this scenario. By their actions in terms of infrastructure, schemes and thought leadership, how will they shape the street? It might well mean a new multidisciplinary and holistic approach to the street. As Reyner Banham said, when you’re running with technology, you’re in fast company - and you may have “discard the professional garments by which you are recognised”.
2. Open source street
Many of the examples in the sketch above have been left deliberately ambiguous as to their openness. The various information modelling systems - the building information modelling system; those conveying the state of local services; those broadcasting the presence of a bus - could be built with openness in mind. Why do this? In order to enable maximum coverage and to stimulate engagement and innovation, with occasional possibility for unintended creative use. And often, it's public data and therefore part of a civic relationship.
Just as good street planning might leave a space open to possibility, and not over-prescribe its program, so informational systems can leave themselves open to possibility. The powerful notion of an API (Application Programming Interface) means that a database can be machine-readable (by code in other systems) and that data can be used elsewhere. APIs can have various qualities - be read-only or read-write, open or closed, and so on.
The library LED installation above might be rather more creatively and usefully imagined as a read-write system. For instance, so that others could insert local information - say, a database of the nearest school’s football team scores - into the stream of information. The bus information system could have a read-only API to enable others to build new visualisations of when buses are due, perhaps deployed even in sculptures or using sound, or enable users of more obscure or outdated mobile devices to have representations of the service programmed for their handsets (which is unlikely to happen with commercial development). The API on those Rolls-Royce engines 30000 feet up should be particularly closed, however.
Interestingly, the equivalent of road signage - say, digital displays of civic information - might be best served not through an open platform. As traffic signs and markings in the UK comply with the Traffic Signs Regulations and General Directions 2002 (TSRGD), it raises the fundamental question of whether a digital equivalent for civic/public data needs to exist. The flexibility of representation in the digital realm - known on the web as separating content from presentation - means that numerous possibilities exist for displaying data. But should public/civic data have its own hallmark of quality, and undergo the kind of stringent analysis and compliance that physical signage has? A Kinneir and Calvert design for displaying public data? Or should the flexibility of the medium enable data to float freely across numerous platforms and modes, enabling different presentation across in-car head-up displays, mobile phone screens, physical displays as per the wind turbines above, façades on municipal buildings and so on. Could we do both?
In a sense, the entire street itself can now be thought of as having an API, conveying its overall behaviour to the world, each aspect of it increasingly beginning to generate and recombine data.
From a governance perspective, all this real-time data should be invaluable to those whose job it is to maintain and develop the street - forming a kind of post-occupancy evaluation for the entire neighbourhood - but if the same data were made open, accessible and approachable (anonymised appropriately) the community that uses the street could end up feeling far more engaged in their environment. The patterns of use in their data become as self-evident as that shortcut worn through the grass in front of the library.
If they can add further detail to this environment, using the street as a platform, all the better.
For instance, simply triangulating location via mobile cell (as per Google Maps Mobile) is currently not accurate to the building level but can often infer the street with some accuracy. With this simple application, wayfinding is enhanced instantly - this is the street you’re on, at the touch of a button. The street then has an opportunity to announce itself to the user, via business listings, historical information etc., all capable of being listed via Google Maps layers (or equivalent). Here is an example of an open system i.e. data is not moderated or prevented, but it is enabled by a third party. (Ownership is a complex issue here. The street’s data is contributed by various parties, but what does Google own?) Aspects of the street not perceptible in such information systems - like the other cafés the hungry Hungarians didn’t know about - may become invisible, in a sense. Does the library exist on this street? Does the smart bench? Should their digital representation relate to their physical wayfinding elements?
So there are a huge variety of possible approaches here, as finely grained and multi-dimensional as there are approaches to planning the built fabric of the street.
In terms of built fabric, the physicality of these informational systems becomes important too. In the locked down street, systems may record behavioural data invisibly. As with Facebook, this sense of being invisibly watched is likely to raise suspicion. If the sensors are made visible, and if the results are made visible, with feedback mechanisms, that's a different story, at least partly.
[Image by Kazys Varnelis, from this post]
For instance, meters and sensors can often be deployed on the street with little information about their purpose. From now on, it may well be important to make the invisible visible here too, to communicate its function and purpose. Though this might appear counter many trends in contemporary product design (e.g. iPod) - where the goal is often to hide, disguise or remove the seams that indicate how the product works, or is constructed or articulated - actually showing the seams of an object is far more likely to engender trust, engagement and appropriation.
[Image courtesy of Timo Arnall]
The city information model briefly described in the sketch would benefit from an open API, enabling the kind of innovation described in the Personal Well-Tempered Environment idea. Again, appropriate decisions need to be taken about privacy and ethics - should councils be able to zoom down to house level to see individual environmental behaviour, or should individuals be able to see their neighbour’s efforts, or their council’s efforts? Might public, or public-private, bodies be a contender for managing the overall data around environmental behaviour at the street, neighbourhood and city levels? By exposing this data in an open fashion, they again engender trust, stimulate innovation, and have some say in the informational civic relationships emerging from the street, a civic relationship as important as any other.
And yet, an open-source approach can mean that the range and depth of information, and the quality of execution, are left in the hands of others. Will the wisdom of crowds conjure a street rich enough to, as Baudelaire had it, enjoy taking a bath of such multitude?
The reality of course is that the near-future street will simultaneously encompass both scenarios - and many more - as in the sketch, due to the multiplicity of clients, services and products that make up a neighbourhood. Designing the contemporary street means attempting to understand and shape the most complex set of relationships between humans and the built environment yet imagined.