With Kickstarter Funding, FOIA Machine Wants to Help Fix Public Records
BY Miranda Neubauer | Friday, July 19 2013
FOIA Machine, a platform that aims to streamline the process of tracking of filing and tracking public record requests, has raised more than $29,000 on Kickstarter — exceeding its funding goal by more than $10,000.
Backed by the Center for Investigative Reporting, the project grew out of a 2012 effort called BirdDog. With the help of $47,000 in seed funding from the John S. and James L. Knight Foundation's Knight Prototype Fund, the project expanded from collecting statistics on government response times to Freedom of Information Act requests into actually generating such requests. It has been tested internally by reporters at CIR, and once it is opened up to public use, it will be hosted by Investigative Reporters and Editors, a professional organization. Shane Shifflett, a developer on the FOIA Machine project and a data developer at the Huffington Post, says the team is now figuring out what features to add given their additional funding.
"We're all totally blown away. Nobody on the team anticipated this kind of support at all. We're just kind of awestruck and trying to figure out what to do next," said Shifflett. Development went dormant after the project's initial funding ran dry. New invitations to use the platform should start coming in September, and a public launch by Christmas.
While other transparency advocates welcomed the success of a new tool, they expressed some concern that FOIA Machine might duplicate existing efforts.
"I think it's amazing that so many people are excited about public records and for better tools to access government information," said Michael Morisy, co-founder of Muckrock, an existing platform in the business of helping users file public records requests.
Muckrock had offered to work with FOIA Machine two years ago, and the MuckRock team was "disappointed" that the FOIA Machine team opted to divide resources rather than working to create one "great" platform, Morisy said.
"Competition is great and giving people more options is good, but it doesn't seem like they are doing anything new or different. It's wasted energy that could be used towards something that is new or different," he said. "It's kind of hard to talk about FOIA Machine since nothing is actually public yet, and they have been talking about the site for a long time."
"Access to information (ATI) laws are among of the most underused tools for citizens to hold government accountable, so it's great to see such support for FOIA Machine. I am as surprised as its creators that it's been funded so quickly," wrote James McKinney, technical lead of OpenNorth, a Canadian open government group, in an e-mail.
"Is it possible for it to take a lot of pain out of the FOIA process? Yes, it's possible. Will it deliver? We'll have to wait and see. The creators of all existing tools have realized that the devil's in the details when it comes to ATI laws. Existing tools handle these particularities by scoping their project to a single jurisdiction, like a country," he wrote. "For example, Frag den Staat, a German ATI website, needs to contend with privacy laws that make an ATI platform much more difficult to run; they've had to implement specific workarounds and redaction tools to comply. So, I'm skeptical FOIA Machine will be able to offer a smooth process across countries as they promise."
"From an open source point of view, it seems like a waste of time to rebuild existing functionality," he wrote.
Unlike Alaveteli, which is more Europe-focused, FOIA Machine is concentrated on the U.S., Shifflett said. And unlike Muckrock, which is backed by a user-pays business model, FOIA Machine would remain free. FOIA Machine's source code will be released, he added, so that news organizations could develop their own versions of the platform for their own use.
The FOIA Machine platform will have tools to help the community engage with each other and help users craft FOIA requests, he said. An API will give users access to a database of contacts and laws, and any FOIA requests made publicly available, which could be used as templates for requests in other areas.
Once development is done, he said, the team expects to have enough money to host the platform for several years and anticipates interest from other organizations once the Kickstarter funding runs out.
Tom Lee, director of Sunlight Labs, said his "greatest hope" was that FOIA Machine would be able to complement the successful work done by Muckrock. But any project that hopes to help people get answers to their FOIA requests faces a backlog of other such requests already piling up inside of government, he said.
Online platforms could help address that issue by making it easy find already submitted requests that are similar, he said. While having more different platforms helps more people become engaged in the transparency process, it "makes it harder to solve that second problem, since responses get splintered across [platforms]."
Lee said a solution would be the development of a protocol for sharing that data across different platforms, as has been discussed in a Sunlight Labs discussion thread, to "create a network effect across all of them."
Update: Shane Shifflett emails —
I have no comment on what took place before I was involved with the project but we're open to collaborating with projects and will have an API to make accessible requests, contacts and laws for anyone to benefit from the work shared by FOIAMachine users ... Software projects are obviously complicated endeavors. Lots of factors influenced our decision to roll our own vs using existing code. That comes down to what languages and frameworks the team was familiar with and the way existing code was organized given the things we wanted to build. These were options we considered while balancing our need to get the job done given our resources.
This post has been corrected. It previously misspelled the name of a FOIA Machine developer, Shane Shifflett.