> > What really pissed me off the other day was that I got some > > spam which got a low bayes score (and hence -5) but a high razor > > score (and hence +2 or so) so I ended up getting it... > > Now that is an interesting dilema. Since I don't run Bayes I don't run into > that situation. Your situation is like asking SA to score your experience > and score someone else's experience. If they do not match then they can be > cancelled.
spamassassin works _so_ well for me, but because I got _three_ (count 'em) spams in the last week I got all mad about spam again and decided I wanted more. The reason bayes scored low was because of all this w o r d s p a c i n g" which counts as l o t s o f t o k e n s presumably. But Razor was saying "yup, that's 100% spam...". I guess eventually my bayes will learn that email containing words like " p e n i s " is probably spam...but maybe it won't :-) I am a mathematician and my emails are full of people saying "let p be a prime number and let s be a complex number..." so probably single-letters look like ham :-) Roll on multiple-token identifiers! Kevin ------------------------------------------------------- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01 _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
