The just-arrived copy of Communications of the ACM has an article on the development of e-mail spam and methods that are used to fight it. It's interesting from the perspective of the various machine learning techniques they describe - and how spammers respond to each tactic.
Spam and the ongoing battle for the inbox by Joshua Goodman and David Heckerman of Microsoft Research and Gordon V. Cormack of the University of Waterloo. The article is behind ACM's portal, so you will have to wait for it to show up on one of the authors' web sites. Or visit your local academic library and read it there (WorldCat lookup). The authors have written extensively on the topic.
Even as spammers and phishers try evermore sophisticated techniques to get past filters and into users' mailboxes, anti-spam researchers have managed to stay several steps ahead, so far.
Interesting to see the claim that the forces for anti-spam are more-or-less keeping ahead of the spammers. This would be a good thing, given the dire statistics they provide: upwards of a billion spam emails per day come into Microsoft's Hotmail servers.
Yes, I am writing this at 10 pm on Saturday. It's well below freezing, and I'm sitting next to the fireplace.