From: Jeroen v. A. <kr...@at...> - 2007-11-30 21:31:41
|
bytehd wrote: > This is dumb design. > Fritz and John have it right: > do the easy, simple-to-execute (helo checks come to mind) checking > then, after all the OBVIOUS things are done, run Bayes. I agree it seems the right thing to do. But I can imagine the filtering happening before the bayesian filter may become less and less effective. If spammers actually got wiser and decided to have their software obey common standards. Many of those filters do seem to depend on spammers' software to be broken. Or did I miss something? Regards, Jeroen |