On Fri, 10 May 2013 15:34:13 -0600 Bob Proulx <b...@proulx.com> wrote:
> The weasel words "agrees substantially" is telling. If it isn't 100% > with no false positives then at least one of those messages does not > agree. That would be the evidence requested. > I am not saying that your technique isn't useful. It is very > pragmatic. I am sure it is very effective. I would probably do that > myself. But it isn't 100%. Nothing is 100%. Even personal Bayes databases are not 100%. > And would you suggest distributing your well-averaged database to > people who install SpamAssassin to as to seed their Bayes? We have a distribution mechanism built into our software. > I think having users start with a blank slate and then start learning > from their own messages makes the most sense. Maybe. But I know that our (commercial) customers expect high catch rates out of the box, and we get that with our shared Bayes database. > And users can always learn from their current mailbox of past > messages so it isn't much hardship. Right; pretend you're a salesperson trying to sell an anti-spam product. "Oh, you just have to go through your old mailbox and classify a few hundred messages by hand... then the system will work great!" No sale. Regards, David.
