Nix said:
> On Fri, 06 Jun 2003, Kai Schaetzl uttered the following:
> > Jonathan Vanasco wrote on Fri, 6 Jun 2003 11:21:43 -0400:
> >
> >> what about the random stuff?
> >
> > I think this is specifically included for confusing Bayes.
>
> No, spams with piles of ordinary email pasted onto the
On Fri, 06 Jun 2003, Kai Schaetzl uttered the following:
> Jonathan Vanasco wrote on Fri, 6 Jun 2003 11:21:43 -0400:
>
>> what about the random stuff?
>
> I think this is specifically included for confusing Bayes.
No, spams with piles of ordinary email pasted onto the bottom are for
that. The ra
Jonathan Vanasco wrote on Fri, 6 Jun 2003 11:21:43 -0400:
> what about the random stuff?
>
I think this is specifically included for confusing Bayes. I don't know if
it works. I just see that we catch 99,9% of the spam at the moment (after
adding some custom rules) and none of those get thru if
some spam has gotten though lately with characteristics like the
following:
1)
EMVJJIUHXRHCKNQIULFTQGTMGLWWSNDDRJVGKMDGXSKATVUPOITCMYWOTFHMRJDQKJDPVTHC
OTGVCU-3809950
2) a{LOCAL_PART}o{LOCAL_PART}e
3) d y e w e e v l p t u?
w c y c y s j t k s d?
4) Unsubs.cribe
5) kcv yuekaxgymqfur b
