Hello, as I was kindly notified by Don Hammond that the script was
mentioned on the mailing list as having some bugs, I'd like to announce
that most of the bugs pointed out have just been fixed. This includes
most of the perlisms like \d and {m,n}.
This script is actively developed as I simply use it to protect my
mailbox. It has efficiency close to 90%, which both means that it loses
about 10% of spam as well as it has about 10% of false positives. If
you have anny suggestions, that could improve this score, please notify
me at <kravietz AT aba.krakow.pl>. I will appreciate any feedback,
especially analysis of the scoring from "VERBOSE=yes" on mails that were
wrongly classified.
--
Pawel Krawczyk
Krakow, POLAND
http://ipsec.pl/
_______________________________________________
procmail mailing list
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail