[Mimedefang] SpamAssassin 2.40 experiences?

[Stephane Lentz]

On Wed, Sep 04, 2002 at 01:26:40PM -0400, David F. Skoll wrote:
> On Wed, 4 Sep 2002, Stephane Lentz wrote:
> 
> > David, maybe 2.41-cvs has better weights yet.
> 
> I'm getting nervous... the weights have changed a lot between 2.31
> and 2.40.  I know they use a genetic algorithm to determine the
> weights.  The problem with that is that it can perturb the weights
> chaotically, and there's no guarantee you'll come up with an optimal
> solution, or even be able to tell how far you are from optimal.
> 
> ...

On the mailscanner list, someone just mentionned something related 
to Bayesian filter :

<<
Date: Wed, 4 Sep 2002 14:13:54 -0400
From: Matt Kettler <mkettler at EVI-INC.COM>
Subject: Re: Feature request!
To: MAILSCANNER at JISCMAIL.AC.UK

As a side-note, the spamassasin team is already working on adding a
Bayesian filter to SA. But that probably won't happen till 2.5x at least.

http://www.hughes-family.org/bugzilla/show_bug.cgi?id=813
>>

regards, 

SL/
---
Stephane Lentz / Alcanet International - Internet Services