[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: lists.debian.org vs google groups

On Thu, Apr 06, 2006 at 11:36:05AM -0400, Roberto C. Sanchez wrote:
> Quoting Pascal Hakim <pasc@redellipse.net>:
> >Even if we assume that I fell asleep on the page down key while counting
> >4., and guess that I missed half, we're still talking about blocking
> >over 800 valid messages.
> >
> >25/37700 works out to be 0.066% of spam not being blocked. It's still
> >annoying of course, as the metric to use is the number of spam messages
> >that make it through rather than the percentage that make it through.
> >SNR and all that.
> >
> That is crazy.  How is the spam filter that good?  On my own server I 
> block a ton of stuff at SMTP time, but an annoyingly high amount of 
> stuff still gets past SpamAssassin.  Is the filter processing all this 
> mail regularly trained?  I know that we can now report mail in the 
> archive as spam, e.g., if it actually gets through.  Has that 
> contributed to an increased level of effectiveness?

The filters have been around for quite a while. There's no systematic
training of filters apart from SpamAssassin AWL stuff.

We haven't integrated the mail marked as spam from the archives into
our filtering yet, so that has obviously had no effect.

Pascal Hakim                                          0403 411 672
Do Not Bend

Reply to: