Quoting Bart Schaefer <[EMAIL PROTECTED]>: > It's usually easier to promptly re-learn a false negative as spam than > it > is to re-learn a false positive as ham, because FNs probably go right > into > your mailbox while FPs are dropped in a quarantine (or worse). Unless > you're not paying attention, a flood of obvious FNs is not going to > "poison" the Bayes database for very long.
Yes, except I neglected to mention I use a common Bayes database for all users (duh, I should have realised this before...). More on this in a sec. > Also, I think you seriously misjudge the difficulty of pumping enough > bad > data into a Bayes database to get something misclassified. Perhaps. Also, I didn't take into account that I run a common bayes database for the whole mailserver, which I know is not the recommended way to run bayes. Running a setup such as mine means this "vulnerability" is magnified (how much I am not qualified to say; I'm just starting to learn more about the inner workings of bayes now) since more of this "bayes poision" can hit the database before the server admin can realize and counteract. I'm very much convinced now that if one runs a common bayes database between all users, one should *not* run autolearning at all (for this reason and others), and instead use manual training of the database on a regular basis. > If the Habeas headers still concern you, use bayes_ignore_header for > them, > don't spend your time manually deleting them. It wasn't the habeas headers getting into the bayes database that concerned me so much, rather it was the contents of the spam itself with all its spammy tokens. Thank you for your very enlightening response. DaC ------------------------------------------------------- This SF.net email is sponsored by: Perforce Software. Perforce is the Fast Software Configuration Management System offering advanced branching capabilities and atomic changes on 50+ platforms. Free Eval! http://www.perforce.com/perforce/loadprog.html _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
