The Data City

On a clear late summer morning in 1802, the poet William Wordsworth paused upon a bridge before entering London. Before him lay the entire city, visible from end to end.

This City now doth, like a garment, wear
The beauty of the morning; silent, bare,
Ships, towers, domes, theatres and temples lie
Open unto the fields, and to the sky;
All bright and glittering in the smokeless air.
[. . . ]
Ne’er saw I, never felt, a calm so deep!
The river glideth at his own sweet will:
Dear God! The very houses seem asleep;
And all that mighty heart is lying still!

Wordsworth’s sonnet “Composed upon Westminster Bridge, September 3, 1802” marks a key moment in the development of the city. This period was the last time it was possible for a pedestrian to see a major city all at once. Although he probably didn’t know it, Wordsworth was gazing upon the largest city in the world. The year before Wordsworth stepped onto Westminster bridge, Britain had completed its first national census. A million people slumbered in London on that September morning. Data could see what Wordsworth could not.

As cities have grown larger, they have become more interesting to look at and, paradoxically, more difficult to see. In the industrial city, it was important to track mass and energy. In the post-industrial city, information itself becomes important. The contemporary city is the city of data streams, each possessing a rich and confounding flow of numbers emanating from a system that is itself part of a larger transnational system. There are no more Steel Cities or Motor Cities. The post-industrial city has no single identity except as a data city. It exists at the intersection of technologies. Observing the data city means identifying its intensities, not its stable identities.

Contemporary computing systems can track everything in a city except its rats. Millions of people now wander the streets of the world’s cities snapping photographs, sending messages, conducting Internet searches on hand-held devices that can have more processing power than the computers that guided Apollo 11 to the moon and back. These activities can be stamped with precise geo-positioning data, processed, stored, then served up through Internet protocols.

But do these technologies give us a better view of the city than Wordsworth enjoyed standing on Westminster Bridge? Is the city more knowable, or less?

Toward the Radiant City

Throughout the nineteenth century, the British government watched fretfully as industrialization spawned vast slums in its cities. The teaming Victorian slum was the main impetus for the development of city planning as we now know it.

Initially, city planners tackled the problem of substandard housing by building boulevards and parks. The archetype of nineteenth-century urban design was Georges-Eugène Haussmann’s plan for Paris, first developed in 1853. Haussmann demolished the medieval core of Paris and dispersed the working classes to the peripheries of the city, conditions that exist to this day. His wide boulevards, ceremonial squares, and unified apartment blocks became the model for city beautification projects around the world. Derived from the neo-classical and Beaux Arts traditions, these projects reflected Voltaire’s belief that industry and the pursuit of urbane pleasures were the hallmarks of a progressive civilization.

Nineteenth-century census data, however, painted a very different picture. A boy born in Liverpool in 1851 had a life expectancy of 26 years, while his country cousin could live to 57. Life expectancy for Britons born in towns larger than 100,000 actually fell between 1820 and 1830, even as national industrial output went up. Industrial England suffered death rates not seen since the plagues of the Medieval Ages. As if the numbers weren’t alarming enough, the Great Stink rose up from the fetid London sewer system in the summer of 1858, fouling the air and reminding everyone of the perils of density.

Census records and improved bookkeeping provided an objective and comprehensive view of the nineteenth-century city. Yet, in the popular imagination, the realm beyond the boulevards was a poisonous labyrinth populated by masses associated with criminality, sedition, and political protest. Charles Dickens described London as “a place of squalid mystery and terror, of the grimly grotesque, of labyrinthine obscurity and lurid fascination.” George Gissing looked at London and saw “the spinnings of a huge, poisonous spider.”¹

The resolution between dismaying objectivity and uneasy subjectivity came in 1935 with Le Corbusier’s La Ville Radieuse (The Radiant City). Le Corbusier wanted to disrupt existing ways of looking at the city by de-familiarizing it, by stripping the city bare of all traces of history, memory, and desire. He stunned New York reporters in a 1935 news conference by proclaiming, “The only trouble with New York is that its skyscrapers are too small. And there are too many of them.”² To Le Corbusier’s remorselessly Cartesian eye, New York is nothing but a grandiose and cataclysmic mess. “On the day when contemporary society, at present so sick, has become properly aware that only architecture and city planning can provide the exact prescription for its ills,” predicted Le Corbusier, “then the time will have come for the great machine to be put into motion.” This is the paranoid city of absolute transparency, in which the master plan guides the very footsteps of its citizens. Le Corbusier believed that planned changes in the environment would be sufficient to produce measurable and predicable changes in people’s perceptions, mental life, habits and conduct. In the Radiant City, “nothing is contradictory any more. Everything is in its place, properly arranged in order and hierarchy.”³

Mapping the Data City

In order to build his Radiant City, Le Corbusier wanted to level existing cities and plant fields of mid-rise skyscrapers in their place. This never happened, but new technologies promise to create Radiant Cities out of existing ones. IBM’s Smarter Cities initiative is designed to place even the most unruly cities under the benign control of the server farm. MIT’s Senseable Cities program is a collection of projects intended to put the power of ubiquitous computing in the hands of ordinary citizens. The Senseable Cities home page announces, “The real-time city is now real!” Under the auspices of the MIT program, plain old Singapore will become “LIVE Singapore!” In prose that neatly elides any references to the city-state’s authoritarian government, the website describes the program as “Developing an open platform for the collection, elaboration and distribution of real-time data that reflect urban activity. Giving people visual and tangible access to real-time information about their city, allowing them to take their decisions more in sync with their environment, with what is actually happening around them.” LIVE Singapore! has not yet gone live, but its promise is very exciting. Armed with real-time data about our urban environment, we will no longer get lost or feel frustrated. Alienation will be a thing of the past.

Or so one hopes. Experiments in creating real-time views of the city are well underway, and so far the results are disappointingly impersonal. The Centre for Advanced Spatial Analysis (CASA) at University College London is a more modest version of MIT’s Senseable Cities program. Fabian Neuhaus, a CASA researcher, specializes in urban maps constructed by geopositioning data “harvested” from Twitter. His maps are landscapes of tweets. The peaks represent the most intense Twitter activity, while valleys and plains indicate lesser activity. Neuhaus calls his Twitter maps “cycle studies” to stress the dynamic nature of his models. “Cycle studies are the science of everyday life, as normal as it gets,” Nuehaus explains. “Its focus is the daily routine, with its habits and rhythms as they occur in most citizens’ lives. It is the power of the normal that brings stability and the routine that ensures security.”

The Cycle Studies maps show that people tweet a lot when they’re near Times Square and in Soho in London. In other words, Neuhaus has verified that people tweet most often in exactly the places we would expect them to, that people feel compelled to express themselves most strongly around the totemic places of the city. The most recent form of mass communication is transformed into a pre-historical landscape of active volcanos, a Mannahatta of the digital age.

Wordsworth’s privileged instant of space and time, a one-shot view of London before it resumes its furious bustle, attenuates into the slow accumulation of data over time, then precisely distributed over imaginary space.

A more sophisticated synthesis of the dryly academic and the vaguely mystical can be found in the “Geography of Buzz” constructed by Elizabeth Currid of USC and Sarah Williams of Columbia University.⁴ Using geo-positioning data from Getty Images, their Geography of Buzz maps out patterns of cultural consumption. Events range across a variety of cultural activities, including film screenings, concerts, fashion shows, gallery and theater openings. Currid and Williams found, to no one’s surprise, that buzz tends to be centered around well-established marque venues such as New York’s Lincoln Center and the Kodak Theater, where the Oscars awards ceremonies are held. Allegedly cool Brooklyn remains dark, its buzz too weak to register.

Currid and Willams’ study is less a document of cultural consumption than a way of seeing the city. The researchers developed a method for gathering and modeling what Williams calls “shadow data,” or “the traces that we leave behind as we go through the city.” By capturing and modeling the traces of daily life, Currid and Williams have opened up a whole new realm of investigation, using data to make visible something ephemeral and abstract, to discover the agitated hearts of America’s twin culture capitals. New York and Los Angeles are publicity cities, vast fields of murmuring punctuated by outbursts of delirium.

By pinpointing buzz, one more mystery of the city has been cleared up. Access to mass data sets, once reserved for large corporations and intelligence agencies like the NSA, has been placed at the disposal of the humble social scientist. Currid and Williams have verified that it is possible to architect mass social experiences, just as Le Corbusier predicted.

Topologies of the Data City

As formerly hidden realms of the city become exposed to the technological eye, urban planning and architectural practice have adjusted. Several commentators have credited Google Maps and Google Earth with initiating the current vogue for green roofs. The same technologies have habituated us into to seeing cities as clusters of hyperactivity in a continuous flow of territory. We never quite step off Westminster Bridge. We merely zoom in for greater detail.

There are two potential consequences to the modeling of real-time data. One is that the city will assume the same shape as Haussmann’s Paris, with a data-rich center surrounded by information deserts. Because the hand-held devices aren’t universally available, or may be used in districts that don’t produce enough data to satisfy the algorithms, entire sections of the city will remain anonymous and mysterious, just as they were in the nineteenth century. Everything will appear in its proper place, as Le Corbusier wished, except the hierarchy will reflect the irrational industrial city he deplored.

But there’s a more optimistic consequence as well to the new data modeling techniques. Data flows traverse all boundaries, so cities will appear more integrated with their surrounding suburbs and countryside. The old city versus country antagonism will be revealed as the illusion it had been since Wordsworth’s day. There will be visual evidence that the city isn’t an exception to the suburban/rural norm. Cities won’t be realms of separate meanings, just intensities of meanings available elsewhere. Yet, at the same time, as they burn more brightly with data, cities can once again assume the role Voltaire assigned them: as the hallmarks of a progressive civilization.