I looked at Spam Bayes manager today and saw a message that struck me as amusing.
"Database 160 good and 3832 spam. Warning you have much more spam than ham - Spambayes works best with approximately even numbers of ham and spam."
Now don't get me wrong, I'm sure that statement is quite accurate, but if I had even an equal # of ham and spam coming in every day, I probably wouldn't need SpamBayes. Sheesh.... <grin>
(Spam Bayes works very well for helping manage my mail box. A few more things fall into potential spam than I'd like, but it's still the best I've worked with)