Graham's spam filter

[Erik Max Francis]

Heiko Wundram wrote:

> Another thing that made me consider a client/server based solution is
> the fact that you can then build a central probabilities database;
> this
> (I think) solves many concerns that people have raised about training
> the algorithm.

But this doesn't sound so appealing when one of the main features of
Graham's method is that it can be specialized by each user over time. 
You'd start with some basic representation of typical good and bad
emails, but over time the filters could come better.  A client/server
solution suggests a single, monolithic database, which doesn't extend
well to this idea.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ There is nothing so subject to the inconstancy of fortune as war.
\__/ Miguel de Cervantes
    Church / http://www.alcyone.com/pyos/church/
 A lambda calculus explorer in Python.