Tuesday night, while on the train from DC to New York, I got an email from Jay directing me to Dan Froomkin’s post about the House Oversight Committee’s online transcripts. I’ve spent the last year tearing my hair out over the lack of such Congressional records, and Jay knows this, so he asked me for my reactions to Dan’s discovery.

First, this development is big news, even when we disregard the citizen journalism implications. Our government’s information services are decrepit, and the fact that citizens do not already have access to all Congressional records is shameful. So the Oversight Committee’s action is a big leap forward for government transparency and the citizenry’s right to know.

Now onto matters of bigger concern to this site. This news clearly bodes well for the citizen journalism movement. The opportunity reminds me of what Talking Points Memo did during the US Attorneys scandal. They had their readers pore over 3,000 pages of email to find news. That’s a lot of reading for a lone journalist. It was like looking for a needle in a haystack, but with so many people doing the search, the job became much easier. The researchers’ discovery made national headlines.

If what Dan says is true—if oversight hearings are seldom covered by journalists—then the Committee is offering citizen journalists about 5 potential coups per month, each one requiring 100-200 pages of reading. That’s much less demanding that the DoJ emails. And because these transcripts are from oversight hearings, they come gift-wrapped with a label: “Scandal lies within.” We might be searching for needles, but our haystack is full of them.

The transcripts are also easy to read. I know from experience how important readability is to an effort like this. For the past year, I’ve been trying to build a site that asks the public read through and decrypt legislation. A big problem is trying to whet people’s appetites for legalese. But transcripts of bureaucrats getting grilled by politicians—especially firebrands like Chairman Waxman—is much more tantalizing. An effort to collaboratively read these transcripts should be a much easier sell than what I’ve been trying to pull off. I’m excited (and a bit envious).

So, Jay asked me, what infrastructure do we use to manage this effort?

This is a special opportunity, but we don’t need anything special to do it. TPM readers parsed those 3,000 pages using only a single blog post and comments. As for coordination, there wasn’t any. The blog owners literally said, “Here are the emails, go read them,” with no further guidance. They didn’t track which pages had and had not been read. (Those 3,000 pages were doubtless read more than once. But rereading can’t hurt.) TPM didn’t know what would be found, so they gave their readers no restrictions on the format of their reports, letting them post their findings as comments.

These committee transcripts are only 100-200 pages, much shorter than the 3,000 pages of emails TPM parsed. And just like TPM, we don’t know what our researchers will discover, so they should be just as unrestricted in their reporting format as TPM’s researchers were.* This project won’t need any dedicated tools.

What it does need is one pair of diligent eyes. This opportunity is different from TPM’s in one regard: it is not a one-and-done project. The committee is posting about five transcripts per month. What we need is someone to keep reminding us about these documents. We never would have found them without Dan—raise your hand if you frequent the Web sites of House committees—and, like TPM, he has made a request for volunteer readers. But is one appeal enough? Months from now, will those of you who were intrigued by Dan’s initial post still be checking the Oversight Committee’s site for new transcripts? If we want to take full advantage of this, we’ll need a designated monitor. This monitor would keep an eye on the committee’s site and broadcast alerts when new transcripts are posted. I’m sure this blog would have no problem hosting such announcements and the resulting research dump.

*Big footnote: Such loose guidance could leave us swimming in comments, unable to make sense of them all. It sounds like this is what happened with NewAssignment.Net’s Bill and Hillary project: “Given HOW MUCH research and data we’ve compiled and produced, we set back our publishing date because there was no way that Daniel Nichanian, who has volunteered to write the piece, could go over everything in the original time frame.”

My question is: why make Daniel do such a thing? After the research is done, if the remainder of the work is still too much for one person, why not WRITE it collaboratively as well? I asked David this. He responded that collaborative writing has no regard for voice—a fair point, and something every Wikipedia reader knows to be true. But Wikipedia articles are still cohesive distillations of disparate sources. Why not distill our collective research notes in a similar manner? If we want distributed research to scale, we should try anything that makes volunteers’ jobs easier. If readers’ notes on these transcripts get out of control, let’s try wiki-fying them before we task the writer. I volunteer myself to coordinate that effort.