well, yes, that's the point. solutions like the ones you used to use do not scale to large volumes of spam. Bayesian filters do - in fact, the more spam they process, the more accurately they work. SpamAssassin is also not just a Bayesian - it's got a pretty sensible set of default rules, but over time it'll build a personalized database for you.
i'd suggest taking a little time to read over some of the stuff i mentioned - at the very least, it'll make you feel geekier :)
(no subject)
Date: 2003-11-07 09:55 am (UTC)Then the pile grew too large.
meh.
(no subject)
Date: 2003-11-07 10:54 am (UTC)well, yes, that's the point. solutions like the ones you used to use do not scale to large volumes of spam. Bayesian filters do - in fact, the more spam they process, the more accurately they work. SpamAssassin is also not just a Bayesian - it's got a pretty sensible set of default rules, but over time it'll build a personalized database for you.
i'd suggest taking a little time to read over some of the stuff i mentioned - at the very least, it'll make you feel geekier :)
-steve