Hello Alexander, On 21 April 2005 at 10:38:07 GMT +0200 (which was 10:38:07 where I live), Alexander S. Kunz wrote and made these valuable points on the subject of "Bayesit - how does it mark a message as spam?":
> On Thu, 21 Apr 2005 09:57:53 +0200, Nick Dutton <[EMAIL PROTECTED]> wrote: >> It's slow during mail retrieval, which is OK because it's doing lots >> of "thinking" (pathetic fallacy?) > K9 does the same "thinking" and its blazingly fast during mail retrieval > (hardly noticable there's anything in-between TB and the POP server). > PopFile is a Perl application. And Perl is the "C64 Basic v2" where other > programming languages are Assembler. :-) K9, POPFile, Spambayes and what more products might be available, it's more a matter of personal experiences and how you look at the product from a cosmetic point of view. I agree with you that POPFile (using Perl, install file over 5MB) and Spambayes (using Python, install file over 4MB) are rather large compared to the tiny install file of K9 (116KB). From an installation point of view BayesIT and Bayes Filter ar easier as they are just plug-ins but my experiences with the TB plug-ins are not very positive...they make the behavior of TB sometimes very "strange" and unpredictable. Since I stopped using plug-ins, I've far less problems with TB. I've used K9, Spambayes and POPFile and my impression is that both Spambayes and POPFile "learn" quicker than K9.....but before we start a new string about this...it's my personal observation....so don't shoot me for that :-) -- Best regards, Dick ___________________________________________________________ I'm not afraid to die. I just don't want to be there when it happens. Using The Bat! v3.0.9.17 Return on Windows XP 5.1 Build 2600 Service Pack 1 Privacy is your right...only if you defend it: http://www.metropipe.net/landing.cgi?id=cloggy ________________________________________________ Current version is 3.0.1.33 | 'Using TBUDL' information: http://www.silverstones.com/thebat/TBUDLInfo.html