There’s a big myth that the “wisdom of crowds” just happens. I studied this in the aformentioned Talking Points Meme article. The mining of the document dump didn’t happen as many people think it did. Read this note from Marshall. After Mike Allen at the Politico broke the news about the 18 day gap, Marshall hunted around the thread and realized that somebody else had discovered that overnight. But, due to the unscaleability of single-threaded comment displays, nobody noticed it, least of all Paul Kiel. The first thing that Kiel did in the morning of March 20. 2007 was read the newspapers to see what they’d found. (Another

But DMN doesn’t even have a discussion thread. So there’s no way for people to work together once they find something.

Here’s what they should do.

One, people needs to go through the source material and tag the data — which are handwritten notes, which are typed notes, which are print clippings (along date, from, to, etc.). This can be done in parallel as long as the researchers follow a common taxonomy.

Two, you need to OCR the handwritten and typed notes. You can outsource this abroad, or you can do it in a distributed fashion via ReCAPTCHA. A.H. Belo (owner of the DMN) could make a token donation to CMU to get these at the front of the queue. (Also note: if you want to preach the gospel of distributed knowledge capturing, you must install ReCAPTCHA on your blog without delay.)

From there, if you have all of the primary source material in plaintext and metadata-tagged, it becomes much easier for a historian to grok through.