> Any thoughts about having some sort of RBL-check vet the incoming > mail against the RBLs before SpamBayes gets a look at it?
Most blacklist checks are really better done server-side, at the time of mail delivery. At that time, you have access to much more reliable information about where the email is coming from (after that time, you really only have whatever the spammer put in the email and a few headers added along the way). However, doing some DNS-based checks (e.g. SURBL) does appear to help. I have code that does various lookups, which at some point I'll get around to adding as an experimental option to the tokenizer (possibly only when messages are unsure, like the 'slurping' options, since these are pretty time-consuming). Some of the work on these was done in work-time, so I'll have to clear with that employer that it's ok to add it, but it should be ok. > I'm using the SpamBayes Outlook plug-in. After long use I'm still > getting too much UNSURE stuff which is invariably straight spam. If you send us the clues (SpamBayes->Show spam clues for this message) then we may be able to offer advice. =Tony.Meyer -- Please always include the list (spambayes at python.org) in your replies (reply-all), and please don't send me personal mail about SpamBayes. http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this. _______________________________________________ [email protected] http://mail.python.org/mailman/listinfo/spambayes Check the FAQ before asking: http://spambayes.sf.net/faq.html
