I have a question.
Why aren't Bayesian filters being used to minimize blog comment spam? And why are the 17 people who worry about tag spam (or "spag") not thinking Bayesian?
There's probably a good reason, but I can't think of it. If you know why, please let me know (in my non-filtered comments).
[ Ridiculous aside: Paul Graham wrote some influential stuff called "A Plan for Spam". Graham rhymes with spam... Instead of being eponymous, it's epono-rhyme-onomous... ]