Hacking Institutional Corruption

by Lilia Kilburn, re-blogged with the author's permission from the MIT Center for Civic Media

Midway through the first full day of Hacking iCorruption, the hackathon that Civic co-hosted with Harvard University’s Edmond J. Safra Center for Ethics this weekend, a participant and Fellow at the Center for Ethics approached Stephanie Dant, Assistant Director of the Center, and gushed, “This is amazing! I asked them”—she gestured at a team of coders—“if they could build it, and they said, ‘Of course.’” The Fellow was agape, in contrast to her coders: “They were deadpan! Like it was nothing!”

For the past five years, under the leadership of Professor Lawrence Lessig, the Center for Ethics has taken the investigation of institutional corruption as its aim. Approximately 180 Fellows have joined Lessig in the aim of producing research and scholarship on the topic—much of which is chronicled on the Center’s recently launched archive of its work. [Full disclosure: I used to work at the Center for Ethics.] For many of them, the work has been public-facing from the outset, and their impact has been seen everywhere from the front page of the New York Times (investigative journalism Fellow Brooke Williams’ article “Foreign Powers Buy Influence at Think Tanks,” which led to a Congressional Truth in Testimony rule) to a series on poverty in The Atlantic. The Center's exploration of institutional corruption in campaign finance has also developed the driving philosophy behind Lessig's own extra-curricular campaigns to solve the issue of money in politics, including the Mayday PAC, the New Hampshire Rebellion, and Rootstrikers. For these Fellows, Hacking iCorruption has been an opportunity to continue to seek and build public platforms for their work—as with Williams, whose readers had implored her to make the data included in her Times article searchable, a task for which she’d lacked the tech wizardry until the hackathon. With the help of her team, she produced a database, thinktankdonors.org, that visualizes contributions. And it’s available on mobile devices, meaning that those in attendance at congressional hearings can quickly check the site to “find potential conflicts of interest on the go.” For others, like the Fellow who bounded over to Dant on Saturday, the hackathon afforded the chance to make their own research and the data behind it more public-facing than they’d ever thought possible.

 on Imgur

Over the course of the weekend, a photographer, Ali (to credit for the image above), and an anthropologist studying hackathons, Audrey, darted about documenting the event. One Fellow arrived on Sunday morning to find a trove of scraped data left for her by an anonymous data fairy. On Sunday afternoon, everyone gathered for the teams’ final presentations, MCed by Danny Miller and presided over by a panel of nine judges, including Civic’s Matt Carroll, who paused his live blogging of the event over at Medium to adjudicate. Across four categories, the ten Hacking iCorruption teams sought to address a wide range of questions:

Law & Governance

How can we improve the annual ethics training for congressional staff? Team Schoolhouse Ethics, led by Fellow Robert Lucas, wants to “replac[e] a proprietary inflexible system with an open-source, easily scalable MOOC…emphasizing learning rather than the recitation of rules.” For the hackathon, they built an EdX course that begins to do just that.

How can we visualize the impact of decisions like Citizens United on campaign contributions over time?Team CampaignCon designed a workflow that downloads and scrapes new FEC filings weekly to detect both long-term shifts as well as suspicious short-term deletions, which Fellow Paul Jorgensen has been tracking. One of their searchable visualizations is available here. (About their name, team representative Bruce Skarin said, “‘Con’ could stand for Congress, or contributions, or it could just stand for “con job.” He then cited Walter Kirn, “The reason con artists get away with what they get away with is their victims are ashamed of their own blindness and their own gullibility, and they tend to just quietly go away”—somewhat of a countermanifesto for the team.)

How can we accelerate citizen participation in legislation, and measure their representation by elected officials? Team BillFinity sought to produce metrics for representation. “By pinpointing constituent ideology on social and fiscal axes, this information can be compared with that of the representative, determining the quality of their representation,” they write. (Some teams are still working on getting their work live; I’ll add their links when I have them.)

Finance & Economics

How can we identify and track dysfunctional community development authorities? MuniMining, led by Fellow Mary Bathory Vidaver, sought to free data around municipal bonds, which is siloed in various government websites including that of the Municipal Securities Rulemaking Board, so that citizens can better monitor CDAs that appear to be behaving corruptly.

How can we reduce barriers to boycotting? Team WeCott, (with the “we” in the logo visually borrowed from the opening line of the Constitution) aims to “crowdsource ethical consumerism” by producing a platform that reduces the transaction costs of participating in a boycott. WeCott makes it easier for people to discuss alternatives to a boycotted business, publicly pledge the amount they will boycott each week, invite others to participate, and see the boycotts their friends are engaging in.

How can we reform the London Inter-Bank Offer Rate (LIBOR)? Team LIBOR REPAIR’s Fellow Katherine Silz-Carson told those gathered for the final presentations that “many of you in this room” are likely affected, through student loans and mortgages, by banks’ strategic manipulation of LIBOR. By developing a benchmark interest rate based on public data that is of comparable term and risk category, the team hopes to create a non-gameable demand-based solution to the problem.

Medicine & Public Health

How can we hack the structure of scientific articles such that information about funding sources and conflicts of interest aren’t hidden at the bottom of articles, but are available up front? Fellow Christopher Robertson spearheaded Team Unearth’s construction of a Chrome extension for PubMed that does just that, thus making it easier for physicians to make informed decisions about what drugs to prescribe their patients.

How can we rank pharma and biotech companies, as well as the medicines and vaccines they create, on critical ethics, public health, and human rights issues? Team OpenPharma, building on Fellow Jennifer Miller’s work, built an index to rank companies on their transparency in disclosing clinical trial information. Some companies make as little as 21% of their clinical trial data publicly available, but GlaxoSmithKline, the winner in OpenPharma’s ranking, does so with 100%.

Nonprofits & Academia

How can we allow professors to certify publicly that they are committed to not allowing their outside engagements (like Board of Directorships or consulting arrangements) to introduce bias into their work?Building on Center staffers Sujay Tyle and Szelena Gray’s ongoing work in the academic independence project at the Center for Ethics, Nikin Tharan, a first-year computer science whiz at Northeastern, stayed up all night to build ProfessorCert, a site that grants certification licenses to academics that can then be embedded into research papers and personal profiles to flag to readers their commitment to unbiased research.

(Williams’ aforementioned project--asking “How can the public find out which countries have contributed to major US think tanks?”--was in this category as well.)

After a deliberation made extra-dramatic by its visually transparent staging in one of the Media Lab’s see-through classrooms, the judges announced the winners. Open Think Tanks and CampaignCon tied for third place. WeCott took second place, and Unearth was in the top spot. In his closing remarks, William English, Research Director at the Center for Ethics, said that although the Center had planned to invite the winners to present at its Ending Institutional Corruption Conference on May 1st-2nd—the culmination of its half-decade of work on the topic—the Center had been so impressed by the outcome of the event that they now wanted to extend the invitation to everyone. “We want to make sure there is a space for all of these projects in May,” he said.

Hacking iCorruption represents yet another variation on the hackathon model and a contribution to the ongoing conversation happening at Civic about how to make hackathons sustainable and inclusive in their approach to problem-solving. It built on the Center for Ethics’ sustained work on institutional corruption, yet made room for first-timers, who, in Carroll’s estimation, comprised more than half of those present. Center for Ethics fellows suggested ideas worth exploring and offered datasets, then worked closely with groups interested in their topic--the emergence of something like a Pro-Am format. How might this example bear on the question of sustainability? As Carroll observed, “These people have been working deeply on these issues for so long. They have a strong incentive to keep these projects alive.”

As the event died down, participants stopped by to thank Katy Evans Pritchard, the Program Coordinator at the Center, and one of the organizers of the event. One volunteer joked, “I’ll send you the bill.” Evans Pritchard to one Fellow, and prizewinner: “You won!” “We won,” he said, gesturing to his coders. “Really, I just made some spreadsheets”—a common refrain in the hackathon’s encounters between academics and coders. One volunteer tried to gift another a container of leftover bulghur salad, but she refrained (“Bulghur and I don’t get along”) so he ambled off to find a new home for it. And long after most participants and all leftovers had left the building, Williams and several of her team members remained gathered around a table, their own nascent think tank thinking through the future of their collaboration.