...was gonna do something similar with the remaining forums. Wrote some scripts to download the pages from webarchive - probably hitting them a quarter of million times in a span of a few weeks. Unfortunately, the quality of their archives for the ezboard forums is very uneven - spread out over all sorts of dates, lots of dead links, and incosistency in whether it was a single message or a span of messages. I got most of the forum titles reconstructed, but the messages themselves require quite a bit more work to extract.

I need to dig through the Icon code again. Icon is one of the languages I've been meaning to explore (found the Griswold book at the local discount bookstore not long ago). Probably a good sort of project for learning the language, since it's a string matching sort of problem.