I find that Bayesian is great, but that I get the best results by
relying on many of the other checks. I loved the scoring that was
introduced to reduce the false positives resulting from Bayesian
mistakes.

Just my opinion across a few domains and sites.

-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Jeroen van
Aart
Sent: Friday, 14 September 2007 5:27 AM
To: Questions and Answers for users of ASSP Anti-Spam SMTP Proxy
Subject: Re: [Assp-user] Next official stable release?

Greg Wright wrote:
> I personally would prefer a more complex, but capable software package
> than one that wasn't updated to the newest threats, or better ways.
> Certainly the logging is improving too, for the better.

Thing is the vast majority of spam emails are filtered out by the 
bayesian filter. At least here, about 90% or more. So adding lots of new

features is nice, but won't do much decreasing spam. It's not a bad idea

to stick with an older version you know works for you.

Greetings,
Jeroen

------------------------------------------------------------------------
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Assp-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/assp-user

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Assp-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/assp-user

Reply via email to