Civic Tech in a Time of Technopessimism

“When I look back at 2010 me, I think I had this naïve idea that tech would save government,” confessed Jen Pahlka, the founder of Code for America, a civic-tech nonprofit.

In its early days, Code for America’s mission was to persuade people at the big tech companies to take a year or two away from their Silicon Valley gigs to serve the public, building technologies for all levels of government. A lot of the time, they built apps. In her 2012 ted Talk, Pahlka referenced a Boston city app that allowed citizens to “adopt” the responsibility for digging out a fire hydrant after a snow storm.

But the more time Pahlka and her colleagues spent with public servants, who she called “some of the most dedicated, brilliant, creative people, just working with incredible constraints,” they came to think differently about the role their technologies could play. They no longer focus solely on building apps that make a government service more efficient, but on changing the systems to which their code is attached, she said.

“When we started it was just about technology,” Pahlka said on Monday at the Aspen Ideas Festival, which is co-hosted by the Aspen Institute and The Atlantic. “Now it’s really about the outcomes of the operations”—outcomes like reducing recidivism or increasing the percentage of people who are eligible for food stamps who actually receive them. And improving government can make tremendous change. “The math shows that at least in safety-net services, being 10 percent more effective would be as impactful as doubling all philanthropic spending,” she said.

Pahlka wasn’t alone in her initial technological optimism. In Mark Zuckerberg’s 2012 letter describing Facebook’s IPO, he said the company’s tools for sharing “could lead to more direct empowerment of people, more accountability for officials, and better solutions to some of the biggest problems of our time.” For most tech companies, that triumphalism continued all the way through the 2016 election, when revelations about misinformation and foreign meddling forced an industry-wide reckoning with its own power.

But it’s not as if digital tools and services have become less important in the broader world. Nor does the national political situation obviate the needs of people in local communities all across the country. Code for America now has about 75 full-time employees and 22,000 active volunteers, and in the areas where Code for America has dedicated the most resources, there is huge room for improvement.

“If you are vulnerable to that cycle of poverty and incarceration, government services are awful to use and they don’t work very well,” Pahlka said. “It doesn’t have to work that way.”

Take probation. “Succeeding on probation is like jumping through 20 successive hoops. Most of us couldn’t do it,” Pahlka said. “Most people end up back in jail but because of a technical violation,” like missing one court date or scheduled drug test out of many.

“We help them overcome those barriers. We collect an enormous amount of data about why people go back into jail for a technical violation,” she said. “Then we work with the system to say, ‘Here are some things we can do not just to help one person, but so everyone can better succeed on probation.’”

She used the example of a volunteer group, Code for Tulsa in Oklahoma. The volunteers had heard from a public defender that “almost everyone” there was held in jail as they awaited trial for fear they wouldn’t show up to court.

“This community group said, ‘We can give you a texting tool that allows you to text with people and it’s been shown to decrease the rates of failure to appear,’” Pahlka said. “And it has helped Tulsa to start to change that policy and begin releasing people who have been held pretrial because they have this tool and they know that more people will show up.”

Her point was: It’s not the technology that’s significant—a texting tool is not a complex technical artifact—but the tool can change the way the system works.

Here in 2018, it’s possible that you’ve noticed that tech did not save government. But some parents who have been accused of crimes in Tulsa, Oklahoma, are now spending the night at home with their kids instead of in jail. Or to take another major Code for America initiative, a bunch of California counties have now made it easier to apply for food stamps.

Neither of these efforts is likely to be hailed as technology saving government, but maybe those big abstractions were part of the problem. By zeroing in on the people who need the most help and figuring out how to get it to them, Code for America offers another way for coders and designers to react to the current political moment.

Center the people and the outcomes, Pahlka argued, not the technology.