> From: list-managers-owner@greatcircle.com
> [mailto:list-managers-owner@greatcircle.com] On Behalf Of Tom Neff
>
> --On Tuesday, February 25, 2003 7:04 AM -0800 "Roger B.A.
> Klorese" <rogerk@queernet.org> wrote:
> > Nick Simicich wrote:
> >> One thing about Bayesian spam filters (I use and contribute to
> >> bogofilter, which has performance as a goal, see
> >> http://bogofilter.sourceforge.net) is that they are
> typically tunable.
> >
> > So what?
> >
> > a) That takes work. I don't want to work at it.
> > b) That occurs after false positives have happened. It's
> > unacceptable to me that they *ever* happen.
>
> Well, then Roger is in a real bind, because there is
> absolutely no spam prevention measure that has been, is
> being, or could be implemented on any level whatsoever - from
> draconian intergalactic legislation to hiring the neighbor's
> kid to check your Inbox - that is incapable of generating a
> false positive.
Exactly.
And we've been told by major users of ours that one false positive is
worse to them than 100,000 pieces of spam in their mailboxes.
So we go on as before.
Follow-Ups:
References:
|
|
