medium and long term trends?

[Stroller]

On Jan 2, 2004, at 6:22 pm, Bopolissimus Platypus wrote:

> i don't understand the math (of graham's original suggestion, or robson
> and fisher's enhancements) enough to have truly a  informed opinion on
> how the bayesian/graham/robson/fisher approaches will work out in
> the medium and long term...

Why not take a look at "A Plan for Spam" & "Better Bayesian Filtering" 
at <http://www.paulgraham.com/antispam.html>. They explain the basic 
math quite well - anything you don't get feel free to ask again here. 
I'm not familiar myself with the Robson/Fisher methods, but Graham's 
articles should be a good outline.

Stroller.