[Spambayes] Just a thought...

Rick Scherle rscherle at bravomarketing.com
Tue Dec 6 19:51:21 CET 2005

First of all, I really like this piece of code. I've been interested in
Bayesian filtering ever since Graham wrote his paper, and your
implementation is very useful.


SpamBayes mostly works on my email, but occasionally an important
message from one of my business partners gets sent to the questionable
mail folder. I don't want this to ever happen, not even one time. 


There are some people with whom I correspond that are completely trusted
by me, regardless of the content they send. Yes, it is possible that
they could be impersonated. Yes, it is possible that a virus could take
possession of their laptop, but these are the risks you run in a
relationship. (I'm not going to give my girlfriend a weekly blood test
before I sleep with her.)


In my experience, it is far too easy to get dogmatic about a philosophy
and forget what it is that we were trying to do in the first place. And
so, when people ask for white lists or black lists, the immediate
reaction is that this is not within the philosophy of Bayesian
filtering. Well, true enough. But they are SO easy to implement, and
they could be an adjunctive technology which moves your software from
98% to 100%; at least for me.


All the best, and keep up the good work.


--Rick Scherle

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20051206/46e3a3d1/attachment.html

More information about the SpamBayes mailing list