On Saturday 09 August 2003 4:37 pm, Uncle Jim wrote:

> > Speaking of which, has anyone noticed that the May release of
> > SpamAssassin doesn't catch as much as the releases from last fall?  Is
> > this because they wanted to rely more on the Beysian filter?

> No, I've been quite happy with version 2.55.  Very little spam gets through
> and when I look at the spam which does, the Bayesian test usually shows at
> least 80%.  It took quite a bit of training before it started using the
> Bayesian filter.

I think that's the thing, I haven't been able to do the Bayesian training yet, 
and I'm getting loads of crap to do it with.

How imporant is it to have a "ham" - which I assume means genuine messages - 
database as opposed to just a spam collection?