Re: Problems with Bayesian filtering

On Feb 26 2004, Craig Hughes wrote:
> You're talking about situations only in which you don't either already 
> know for sure which MUA's being used, or aren't being told either by 
> the user, or by the MUA, which MUA's being used.  Obviously, if you 
> don't know, then you don't know.  But if you *do* know, then you can 
> certainly do better than just guessing.

Actually, I was partly thinking of situations where mail is filtered 
on a company wide server, whith different users using different MUAs,
sometimes several depending upon where they are logging in from.

It doesn't matter really. I'm all for filter writers doing whatever
they feel is necessary to pick up spam. If that means working out how
PINE handles messages, then so be it.

-- 
Laird Breyer.