Re: [Full-Disclosure] Spam with PGP

["Jonathan A. Zdziarski" <jonathan@nuclearelephant.com>]

So essentially you are agreeing with me; that heuristic-based filters
are obsolete.  This is evidenced by the fact that you're interested in
turning SA into more or less a Bayesian filter.

>From what others have told me, it looks like SA devel is actually going
in that direction too - which is good.  My only qualm now is that if SA
wants to be a Bayesian filter, go ahead and do it and get rid of all the
rulesets so they can focus their attention on enhancing the Bayesian.

On Tue, 2003-10-07 at 16:14, Shawn McMahon wrote:
> On Tue, Oct 07, 2003 at 02:45:43PM -0400, Jonathan A. Zdziarski said:
> > percentage of the actual final score.  Because true Bayesian filtering
> > performs a huge majority of the same tests that SA performs, SA's own
> > ruleset easily waters down any bayesian findings whenever there are
> > opposing values between the two.  For example, a pine MUA...SA thinks a
> 
> Those values are fully configurable.  You're confusing obsolescent
> design with a PEBKAC.
> 

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.netsys.com/full-disclosure-charter.html