-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Alex Pleiner writes: > Fellow Spamfighter, > > while preparing some slides for an anti-spam workshop I reread Paul > Graham's "Will Filters Kill Spam" at > <http://www.paulgraham.com/wfks.html>. > > Wouldn't the URIs mentioned in Spam be good keys for some kind of > auto-whitelisting with a similar mechanism as for AWL? > > While the from address is frequently forged, the URIs listed in the mail > body should not. > > I like Paul's idea to punish the spammers by auto-retrieving those URIs > but I can imagine lots of collateral damage (spammers include links to > microsoft.com). Auto-whitelisting those URIs and using this to > lower/increase the final score would be different. > > Maybe the bayes filter can take URIs into account, but only as part of > the whole body. > > Am I wrong on this? Is this a practical idea? Do I miss something? They already are in SpamAssassin 2.60. They're tokenized heavily, and it gives appreciably good results in terms of improved accuracy. - --j. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.2 (GNU/Linux) Comment: Exmh CVS iD8DBQE/mCkIQTcbUG5Y7woRAjHLAKDuP21hLB7OGz3WwVZrIAwDLKBRGwCfSKO6 XGoEDuexQNBFJ8xCd8g02gs= =GoZs -----END PGP SIGNATURE----- ------------------------------------------------------- This SF.net email is sponsored by: The SF.net Donation Program. Do you like what SourceForge.net is doing for the Open Source Community? Make a contribution, and help us add new features and functionality. Click here: http://sourceforge.net/donate/ _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
