The Right Way to Make Cities Smart

Most data-driven "civic apps" report problems. What if they facilitated civic engagement instead?

Christopher Le Dantec/Cycle Atlanta/Georgia Tech

In the 1960s, RAND Corporation researcher Harold Sackman advanced some of the earliest arguments for real-time data in the public interest: “It is commonplace to point out that computers make it possible to collect, organize, and process vast amounts of data quickly and reliably in real time experiments that were beyond the ken of the precomputer era.” Sackman’s assessment of computing in the late 1960s was meant to argue that computation would allow—and had already begun facilitating—large-scale, data-based experimentation with social programs. His was a pragmatist’s view. The combination of scientific method and open real-time control systems, or what we would now call Big Data and Cloud Computing, could be productively brought to bear on for benefit of the public good.

Nearly 50 years later, we are faced with whole new classes of devices and services that operate with very much the same goals in mind. By building sensing capabilities into everything from our phones to our refrigerators, our eye-wear to our thermostats, we occupy a world ever more sensed and sensible to computational analysis and management. Turning our selves into data, structuring our lives around devices that communicate with us through our smartphones, and adopting services that tailor themselves to our presumptive needs comes with incremental conveniences for the individual but substantial consequences for the community.

A subset of mobile computing apps and services bring this conflict into focus. These tools combine mobile computing devices that collect real- or recent-time data, with new forms of community participation, civic engagement, and local governance. Examples of these kinds apps and services include SeeClickFix, PublicStuff, and Street Bump—all of which exist as systems that mediate feedback about local urban issues and help communities mobilize to address those issues: SeeClickFix enables community members to setup neighborhood watch groups that record and respond to issues and interface with local officials when needed; PublicStuff acts as social-media enabled middleware allowing citizens to directly submit issues to local government; Street Bump automates pothole and pavement issue detection by leveraging the sensing capabilities of modern smartphones. These services hope to improve the public good by means of new forms of public participation, data collection, and data analysis. They enable citizens to collect facts about their community—untrimmed trees, broken streetlights, and potholes, for example—and help report those facts to local governments for more efficient repair and management.

These services encourage a specific type of public participation, one that reinforces the idea of cities as service providers and of citizenship as a kind of light-weight consumer feedback activity.

Computing has always promised efficiencies—of productivity, of data—and services like SeeClickFix and its ilk strive to apply those promises to local government. It is through these efficiencies that they are able to elide the messiness and contention of citizenship that play out through more traditional modes of public engagement and political discourse. Of course, the messiness of politics turns out to be unavoidable, and not so easily tamed by the rationality of computation. We are simply left with digital versions of the questions that have always followed democracy: Which citizens participate? Who benefits?

What might it look like to incorporate more active and political forms of participation in an app- and data-driven world? In my research group at Georgia Tech, we have developed a smartphone app that interfaces with a larger regional planning project. The app, called Cycle Atlanta, enables cyclists to record their ride data—where they’ve gone, why they went there, what kind of cyclist they are—in an effort to collect more knowledge about cycling in the city. But more importantly, rather than just using data to identify faults in need of mending, Cycle Atlanta aggregates information with the explicit goal of helping the Atlanta cycling community advocate for particular infrastructural reforms.

To put the difference into relief: apps like Street Bump or SeeClickFix are designed primarily toward individuals sensing the world. They are intended to collect raw data about the city by asserting facts like, “There is a pothole here.” The social and political consequences of such assertions have nothing to do with the data itself, but are instead linked to the socio-economic contours of the populations who would use their smartphone to detect potholes. This is a standard critique of smartphone-based civic participation: they may sense the urban environment, but they only sense the parts of the urban environment encountered by the citizens who would be inclined to use their smartphone to do so. It is not the data itself that is enacting a political reality, but the conditions under which those data are being collected.

Cycle Atlanta uses those same world-sensing capabilities to gather data about where cyclists are riding, and it succumbs to the same critiques with respect to socio-economic implications for who is and who is not counted or sensed—a critique we are working to address. However, where Cycle Atlanta differs is in the way it is positioned as a tool for civic engagement. It is not simply sensing and reporting conditions, enlisting individuals in a kind of customer-service feedback loop. Instead, the app has been intentionally and strategically deployed under the aegis of collecting data to inform planners where bike facilities should be built. The data do not just reflect the world as it is, but instead make claims about how the world ought to be. The recorded data of many cyclists on a given Atlanta road is not validation of adequate bike lanes and safe facilities, but an activists’ claim of, “We are here!”

Cycle Atlanta/Georgia Tech Participatory Publics Lab

By recording their rides and sharing that data with the City, Atlanta cyclists are participating in a new form of civic advocacy. This new civic work is accomplished through the rhetoric of collected data, and often, in lieu of attending public meetings. On first blush, this circumstance may upset the average citizen. How can data take the place of public discourse? But in reality, public discourse is often relatively meager, and citizens might participate more when they do so in aggregated data. At a recent public design charrette organized to discuss planned bike lanes, about 20 members of the public turned up to discuss and advocate for particular changes. By contrast, over 1,500 Atlanta cyclists have contributed data about their rides through the smartphone apps and these data are advocating on behalf of those cyclists for facilities even though those individuals may never attend a public meeting on the topic.

In any civic app or service, the data collected are never raw, they are always the product of social production that include the choices of what to data to collect, how to collect it, and the motivations for why individuals might participate in that collection. Apps like Street Bump use data in the hopes of improving the city as a service by increasing our ability to know about the condition of those services. But the Cycle Atlanta app uses data to change how communities interact with local policy making and planning activities. This is not merely a difference in degree, but a difference in kind. Data from Cycle Atlanta is not reporting the condition of present services provided by the city, but is being submitted and collected to support public debate. The app and the data collected are meant to be an alternative to the public meeting—where citizens could once only advocate for particular decisions at the microphone, they can now advocate through digital participation with the added persuasive force of the data they produced to support that argument.

The information historian Geof Bowker has observed, “If data are so central to our lives and our planet, then we need to understand just what they are and what they are doing.” In the case of Cycle Atlanta, data form an explicit argument cyclists make about their current way of navigating the city. As a form of public rhetoric, the data our app creates do not simply report the state of the world, but act as political surrogates for a community advocating for its interests.

The strength of communities is not the efficiency of their services, but the degree to which citizens participate in the creation of those communities. Cities are composed of many different communities, with many different ideas about how to best to support their needs and desires. That a city has whatever services it has is the product of negotiations and political battles that take place over time. When we configure digital civic participation as a way to improve the city as a service, we fail to recognize that citizens are not consumers of cities, but producers of cities.

Within the context of the Cycle Atlanta project—from the app that collects data to the tools and analyses that planners use to interpret that data—there is an opportunity to support novel forms of civic participation and to develop novel institutional responses that understand data as knowing about the city in a particular way. App-driven digital democracy need not default to treating city as service and citizen as consumer. Instead, civic tools can produce new forms of civic participation wherein the app and the data become bullhorn and soapbox.

This post is based on research from the Intel Science and Technology Center for Social Computing