You'll likely find that those words wouldn't be considered "interesting tokens" - and if they do, they will also be considered interesting tokens for all the ham you receive discussing these topics. The bayesian engine doesn't simply grab words; it grabs tokens, and it grabs them in some really (to a human eye) bizarre contexts.
Also, out of curiosity: do you find that the spamassassin-talk emails with attached spams score high enough to meet your auto-learn threshold? -tom > -----Original Message----- > From: Carlo Wood [mailto:[EMAIL PROTECTED] > Sent: Monday, September 01, 2003 7:25 PM > To: Simon Byrnand > Cc: [EMAIL PROTECTED] > Subject: Re: [SAtalk] Bayes and whitelisting > [ ] > > Take for example this mailinglist, this very mail, it is full of words > like "whitelist", "SpamAssassin", "autolearnt", "score", "man pages" > etc. If you included a SPAM as example (quite possible on this list, > and the reason why I whitelist it) then I still don't want it to be > autolearnt: that would mean that the mentioned words get tagged as > spammy, and they are not. ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk