Todd Hubler
Congregating in the back room of Michelangelo’s coffee shop a few Sunday mornings per month, a small group of Madison programmers is trying to save the world. Or at least help make Madison’s budget process run a little smoother.
They are members of Civic Hacking Madison, a Meetup.com group started in 2013 by Erik Paulson — a UW-Madison computer sciences graduate student and former member of the city’s Digital Technology Committee — and a handful of other local developers. This particular faction aims to build a tool that will enable Madison’s Common Council to prioritize items during the city budgeting process.
Software product manager Lucas Dailey, who works at Propeller Health, launched this project earlier this year in response to a “little-known but very important problem in legislative budgeting” that he observed as a Common Council alder.
In the budgeting process, Dailey says, the council debates and votes on items sequentially throughout a single session. An unintended consequence is that sometimes, by the end of the night, “you run out of money.” This runs the risk of leaving items toward the end of the list unfunded.
“City budgets are, to some degree, determined by the order in which they are discussed,” he explains, “which is kind of an undemocratic and really poor way to do it.”
When Dailey was on the council, he wanted to find a computerized tool that would allow members to rank the proposed amendments in order of preference, and calculate a ranking of their collective preferences from highest to lowest priority. Fellow alders were receptive to the idea, but Dailey’s search for such a tool was fruitless.
So, after leaving his post in May 2015, Dailey reached out to local programmers to see if they could build one.
Civic hackers are computer programmers, designers, engineers, and other community members who use technology to develop creative solutions to civic problems: from apps that crowdsource data on faulty emergency sirens to data rescue-thons that archive federal data at risk of being deleted or altered by the Trump administration.
While the movement has yet to gain major traction in Madison there are signs that it is growing. Civic Hacking Madison is one of a few groups that have cropped up over the past few years. Its projects, such as improving the digital displays at city bus-stop kiosks, are free-form: run by whoever has the desire (and time) to contribute.
Currently, the team has around 10 core members, including Dailey, Andrew Shell, Carl Schroedl, Nathan Eckberg, Tyler Harter, Ken Krebs, Jordan Weaver and Sebastian Petak. Most are developers at startups in town, but they contribute a variety of skill sets and backgrounds.
Seated around two tables pushed together at Michelangelo’s, they discuss logistics for the budget tool — sharing progress reports of research and development from outside meetings, discussing how to tackle various problems, and suggesting additional features, such as multi-step authentication to improve security, or the option for users to update their preferences continuously. They usually spend the latter half of these meetings working, their oversized laptops occupying most of the table space, and use GitHub (a development platform) and Slack (a group messaging service) to communicate after hours.
The group will likely draw from existing tools and apps, rather than coding the tool from scratch.
“As coders, we get excited about coding something, but it might not be the best or most efficient thing to do,” says Schroedl, who offered to learn a little Ruby (a programming language) to help adapt an open-source tool on GitHub.
Early on, the group settled on a mathematical method that would order the preferences of council members: it’s a type of preferential voting (any kind of voting that requires people to rank their choices) called ranked pairs. “I like ranked pairs because it looks at the greatest margin of victory,” Dailey says. “It asks: ‘Which of these choices most clearly represents the voice of the people?’ ... It looks at three or more options that are about tied and breaks the tie.
“I think ranked pairs is the best for democratic governments,” he adds. “It’s the most fair and clear for the general public. People can broadly understand it.”
The group will spend time focusing on the look and feel of the tool and its overall functionality for users.
Shell wants the user interface to have a “rockstar” quality. “If it’s even a little confusing to use, people won’t use it.”
The group hopes to present a finalized, open-source version of its tool to the city in time to be used in the fall budgeting process. If successful, they plan to showcase it to national tech groups, so it can be further improved and used more widely. The idea of ranked pairs could also be applied to elections, says Dailey, who envisions a “post-partisan” future.
Generally speaking, Dailey says, civic hacking tends to attract people who are “technocratic.” Rather than advocating for a single political issue, they are trying to make the overall system better.
“Looking at these systems, you start to feel like we don’t live in a democracy,” he says. “This is the only truly democratic way for people to make a decision on something that isn’t yes or no.”
