From: Forrest J. Cavalier III (mibsoft@epix.net)
Date: Tue Jun 25 2002 - 07:37:15 CDT
> To: USEFOR <usenet-format@rkive.landfield.com>
> Subject: Re: Subject header statistics
> From: Russ Allbery <rra@stanford.edu>
> Organization: The Eyrie
> And I agree. As far as I'm concerned, false detection is no longer an
> issue, assuming that Andrew's results are representative (and I have no
> reason to believe that they're not).
Whoa. There were 4 false positives out of 49.
We don't know anything about those 49. If they were all from
the same newsgroup, 8% error rate is pretty bad.