Interview with Matthew Burton on Mozilla Foundation, but for Government

Hi Dave, thanks for bringing me in. Let's get to it!

A few government employees said it was a morale boost, so that was fantastic. At the end of the article, I asked people who were interested in two aspects of the article to write to me and voice their interest. I've gotten a good response to that request.

Finally, some people, both in comments and via email, disagreed with my arguments and/or the ideas. What's great, though, is that all such criticisms were constructive. There was zero bickering or ad hominem attacks, which, having blogged in the past about government and politics, was a huge surprise for me.

Overwhelmingly positive, especially from those outside the government. I've gotten great feedback from everyone from government reformers to everyday programmers to founding fathers of the open source movement.

My friends in the government, who are primarily in the Intelligence Community, have stopped short of a full-out endorsement of it. But they are definitely open to using more open source software...as long as it comes from a trusted source. That's one of the reasons why I want to create an organization with a formal government relationship.

Hi Mark. State and local governments could absolutely participate. There are a few reasons why most of my words about it have focused on the federal level:

-That's where I have experience
-That's where all of the high-profile government software screw-ups have happened. In other words, their projects waste more money than state ones, and using such examples is a better way to convince people of the need for this project.
-Their work affects people nationwide--worldwide, even. So it's easier to attract possible programmers if they know their work will have a larger impact.

On the other hand, it might be easier to attract local programmers for local projects. Local governments will probably have less red tape than the Feds as well. So yes, I am definitely open to working with non-federal agencies.

Two things:

-First, I'm looking for funding to help me develop the project and promote the idea.

-Second, I'm researching what sort of legal entity the organization should be, how it should be funded, etc. This seems trivial, but when dealing with the government procurement process, your corporate structure could be the difference maker. Government procurement is a very murky world. I'm sure that even if a government office would love to work with us, their ability to do so could hinge on the most arcane of legal matters ("Oh, sorry...you didn't check Box 47f. Deal's off."). Therefore, I'm reaching out to people with government contracting and procurement experience to find out how to do get this project off on the right foot.

Josh,

you've struck at the heart of the matter. The two challenges you mention are precisely my motivations. I want to open the government's eyes to a new and better way of doing things, and I want to decrease the influence of the military-industrial complex and keep them from pilfering taxpayer dollars.

I have an answer for the first one. A necessary part of marketing the project will be talking to government managers and teaching them the facts about open source. I expect to encounter lots of skeptics who believe open source is low quality, is unsafe, etc. They need someone to explain the concept to them, so this project would be a good way to do that. Even if a certain person isn't willing to do a project with us, at least we have an opportunity to teach them the facts about open source.

Another way to fight it is in how we develop. Lots of open source projects allow for anonymous contributions, and government friends have told me this makes them skittish. So a low identity-based barrier to entry might be necessary if you want to contribute code to government projects.

As for the second challenge, I don't expect our financial resources to compete with the likes of SAIC, Booz Allen, Lockheed Martin, etc. But we can provide some things they can't--or, at least, aren't very good at: Our programmers will come from a bigger and better talent pool. The Release Early, Release Often philosophy will let us respond to user feedback as we develop each project. Change requests can be made and deployed within hours instead of months. And then there's the whole "free" thing.

I realize, however, this "muscle" you speak of might be used subversively. If that happens, I have no idea how I'll react to it. But I do know that if it does happen, it means this project has arrived.

Dan,

you mentioned a stellar prospect for a pilot project. Voting reform would definitely attract hacktivists. There are a few movements afoot to create an open source voting mechanism, but I don't know how far along they are, whether they're simply pushing for it versus actually building it, etc. If it were adopted, it would do wonders to fight the open source stigma that Josh mentioned: if the government trusts it to run our elections, they can probably trust it for other things as well.

That said, any project that addresses a visible public concern would be a good pilot. Two possibilities:

-anything that helps an agency improve its FOIA process
-a better way to manage Homeland Security's No Fly List (which topped 1 million people last week)

I used to be a government employee, and, like many organizations, EVERYTHING revolved around our systems. Searching for information, finding people, emailing them, finding out who knows what, distributing knowledge...all of that relied on a good network and good software.

We, however, barely got by using a bad network and bad software. As I wrote in the Man essay, almost every government employee does their job with a computer. And if the computers they're using aren't fit for their jobs, they can't be expected to do those jobs well. That's why improving systems will improve government.

In the long run, I also hope that this project will teach government managers that good results need not be expensive. This lesson would do wonders for public confidence and pocketbooks.

Yeah, I haven't really come up with a good name yet. I'm not the branding type. So if anyone has any ideas, I'd love to hear them.

(govzilla.org appears to have been parked by spammers posing as government employee benefits providers.)

ReadableLaws was my Masters thesis from ITP. It's a Wikipedia-like site that lets people read laws in plain English, along with an explanation of their practical implications.

It was born from an idea proposed by Jay Rosen: wouldn't it be great if we actually read the laws passed by Congress? This idea intrigued me, because it's so simple, so essential, and yet so infuriating that it's not already happening: bills are way too long and complex for regular people to understand, and the senators and representatives who vote on them often don't read them either.

There are a few other Congressional information sites out there, some of which have much more visibility than ReadableLaws. So lately I've been talking with those people about integrating the RL concept into their sites.

Speechology is about two things: building a video archive of campaign ads and debates, and subjecting those videos to public scrutiny. Campaign ads are full of quotes and headlines taken out of context. Debates are supposed to be about informing voters about candidates' records and viewpoints. But instead, candidates usually simply ignore the question and reframe it in order to attack their opponents.

These videos deserve scrutiny. At Speechology, you can not only see every TV-based presidential campaign ad and debate, but you can find out truthful they are as well. We do this by letting users post their own research and analyses.

Dan Phiffer and I launched the project about a month ago, and it's going pretty well so far. We got a small grant from the Sunlight Foundation to help us out. We also want to expand beyond the presidential race and archive Congressional campaign videos as well.

Aside from the ones I mentioned in response to Dan's earlier question, I've thought about making some simple things to help government employees communicate and manage projects. Think of it as a 37Signals for the government. You'd think that they would have good tools for simple things like chat, but they don't.

I'll add it to the list. Before proposing this idea, I knew of a few embarrassing failures, but ever since mentioning the idea, I've learned of several others. In the coming weeks, I need to hide in the library and read up on them. Let's talk about the census mishap offline sometime soon.

Analysis of Competing Hypotheses is an analytical technique developed by Richards Heuer, a CIA veteran. The technique helps you consider many different hypotheses and track them against many pieces of evidence, which is helpful not just for intelligence analysts, but for anyone whose job it is to solve problems.

Heuer found me and asked me to build a Web-based platform for this technique, so that multiple people could share information, collaboratively solve a problem, and hammer out their differences.

We've been developing it for the past year as a pilot program for the Intelligence Community's much talked-about A-Space project. It will be ready for public (and corporate) use in a matter of weeks.

Thanks, Dave. I appreciate the opportunity to get the word out.