16.2 not as effective

David Relson relson at osagesoftware.com
Tue Jan 20 17:50:52 CET 2004


On Tue, 20 Jan 2004 10:35:55 -0600
Bill McClain <wmcclain at salamander.com> wrote:

> On Tue, 20 Jan 2004 08:24:52 -0800
> Dan Singletary <dvsing at sonicspike.net> wrote:
> 
> > Has anyone noticed that bogofilter 16.2 is just not as effective
> > against 
> >   catching spam as previous versions?
> 
> Since I first used bogotune in 0.15.11, bogofilter is catching 98% of
> my spam, and that has not changed in recent updates. Before that: not
> sure. I was averaging 60% success with the default parameters.
> 
> -Bill

Bill,

Good to hear it's going well.  As some statistics, bogofilter has caught
6132 spam (so far, this year).  There have been 122 unsures of which 89
were spam and 33 were ham).  There have also been 3 false negatives. 
I'm unaware of any false positives (though I've not looked all that
closely at the 6132 spam).

About 50 of the unsures(spam) are from 2 gnu.org mailing lists that get
spammed regularly.  I'm staying on those lists as an experiment to see
how long it takes for the hammish mailing list header tokens to be
outweighed by the spammish message content.

David




More information about the Bogofilter mailing list