Thus spake Barry Wainwright <[EMAIL PROTECTED]>, circa 6/16/2004 1:28 AM:
> Individually, [Bayesian poisoning] won't have any apparent effect, but
> because of the way a bayesian filter works, if it is trained on messages
> like this there will be a slow accumulated detrimental effect which will
> lessen the effectiveness of the filter.

I'm not qualified to comment on this except to say that better programmers
than I find this controversial.

Some of you may find this page interesting:

Sophos white papers about spam <http://www.sophos.com/spaminfo/whitepapers/>

     peter

-- 
Help the environment! Become a vegetarian! <http://www.vegpledge.com/>

-- 
To unsubscribe:                     
<mailto:[EMAIL PROTECTED]>
archives:       
<http://www.mail-archive.com/entourage-talk%40lists.letterrip.com/>
old-archive:       
<http://www.mail-archive.com/entourage-talk%40lists.boingo.com/>

Reply via email to