Figure III.1 Street sign in Cambridge, MA, requesting citizen participation in reporting maintenance problems with urban infrastructure.
Figure III.1 Street sign in Cambridge, MA, requesting citizen participation in reporting maintenance problems with urban infrastructure.
Self-described as “investigative crowdsourcing,” the German collective GuttenPlag Wiki produced extensive evidence of plagiarism in the dissertation of then-Minister of Defense Karl-Theodor zu Guttenberg (Anonymous and Kotynek 2013, 76). Interacting exclusively through pseudonyms, the group set up an online infrastructure to coordinate tasks and verify findings, establishing internal self-governance through techniques that exposed false findings by participants acting out of malice or overzealousness.
GuttenPlag is one of many examples of citizens organizing around a specific purpose and developing tools meant to hold a government accountable. Mechanisms for accountability are a central part of every democratic government that has to maintain legitimacy in the eyes of citizens. As political scientist Andreas Schedler puts it, “The great difficulty lies in this: you must first enable the government to control the governed; and in the next place oblige it to control itself” (1999). Accountability obliges one party to inform another about past or future actions and decisions, to justify these actions, and to accept sanctions in the case of violations (ibid., 17).
From the perspectives of both the system and the individual, accountability can be seen as the legibility of governance, made visible by transparency measures such as the U.S. Freedom of Information Act. As I show in chapter 5, official instruments of visibility often are not enough to establish the accountability necessary for a democratic society to function. When citizens perceive that government actions lack responsiveness or clarity, they may attempt to establish these values through official channels or by their own means, invoking rowdy shame campaigns or developing the tools and procedures that the government cannot or will not provide.
The GuttenPlag Wiki example illustrates the multifaceted roles that data-centric practices and visual representations play in community-driven initiatives. Citizen groups utilize visual representations for purposes that include recruiting volunteers, coordinating tasks, collecting data, analyzing evidence, and managing discussions. At the beginning of the GuttenPlag project, the text analysis and visualization tools were makeshift creations. I often describe such things as “dirty visualizations”:1 rich, cluttered, and less polished than professional design products, but essential artifacts of the operational process, containing many traces of the collaborative work that are no longer present in the final product.
As the GuttenPlag analysis neared completion, a more polished form of visualization became necessary for distilling the results for public discourse and building coalitions with journalists. Information designer Gregor Aisch created an elegant and complex visualization, documenting all identified instances of plagiarism in the dissertation.2 This required the careful rewording of statements to make the findings quotable and to abstain from hyperbolic language that would compromise the group’s role as a research platform.
The methods used in GuttenPlag are also applicable in other domains, including environmental monitoring. Traditionally, citizen-led initiatives depend on a well-organized community of volunteers who can make a considerable commitment in terms of time, knowledge, and resources. What can be effective for addressing local waste issues does not necessarily work at a larger scale when a large number of observations is necessary. Crowdsourcing initiatives try to overcome these limitations by reducing the necessary commitment to participation. The flood-mapping initiative Peta Jakarta accomplishes this by harvesting the Twitter network—popular among the residents of Jakarta—for reports of flooding and then invites the senders to contribute their observations to a shared real-time map.3 International organizations including UN agencies are experimenting with crowdsourcing platforms to collect information for disaster and crisis response (Meier 2015) or to report illegal dumpsites (Fathih 2015).
Crowdsourcing can excel when measurement is a simple affair, as in the case of identifying and reporting software bugs by the users of the program. In the words of open source “evangelist” Eric Raymond, “Given enough eyeballs, all bugs are shallow” (Raymond 1999). Reaching the necessary number of “eyeballs” is not trivial, though, and it is the main reason why the majority of crowdsourcing projects fail. Crowdsourced monitoring initiatives also struggle with data quality issues, with data being notoriously biased, unreliable, and unevenly distributed.
In classic crowdsourcing, a participant solves an atomistic problem without necessarily seeing the larger purpose of the task, the contributions of other users, or how the collected data are used. Such a design may simplify participation for both contributors and users, but it also implies lower accountability on both sides. Environmental sensing projects frequently address this tradeoff by using a hybrid design that combines aspects of traditional community-oriented projects with the simplicity of crowdsourcing apps. Calibrating the visibility of users and their work both internally and externally is an ongoing process that will be further discussed in the third part of this book.
The design of civic interfaces plays a critical role in communicating any type of information. Given the complexity of infrastructure governance, interface design deserves particular attention for a reporting system in which citizens interact with governments and service providers. Using two case studies that look at mobile apps used within the Boston metropolitan area, chapter 6 examines how interface design affects the reporting of infrastructure issues such as potholes, broken streetlights, and graffiti. As I will show, interface design not only affects how problems are reported, but it also alters the perceptions of what constitutes an issue in the eyes of both citizens and the city.