As someone who has recently been the victim of an unnoticed false positive, I heartily endorse this. As well, I'd recommend taking down the score for FORGED_HOTMAIL_RCVD, since the test is implemented using a hardcoded IP address and currently triggers on legitimate Hotmail email (see http://issues.apache.org/SpamAssassin/show_bug.cgi?id=5316 ).
Some quick suggestions, then: score DNS_FROM_RFC_POST 0.5 score FORGED_HOTMAIL_RCVD 0.5 As per Adam's suggestion, if no one complains, I'll implement this in a few days. (Franklin, you might have a better idea of the numbers Yahoo mail gets for the RFC-Ignorant lists -- please chime in on what might be better numbers for some of the other rules.) Jeremy Adam Chlipala wrote: > Franklin Gordon Bynum wrote: > >> Could we at least reduce the score for these lists? Especially the >> postmaster score that's outright excessive at 1.7. >> >> I'll figure out how if people are interested in improving our spam >> filter. I sure am. >> >> > You'll find that we already have some custom weights set in > /etc/spamassassin/local.cf. If you suggest some additions to that file > on this mailing list, I think the reasonable procedure would be to wait > a few days for anyone to object, then, if no one does, add your lines to > the file. > > _______________________________________________ > HCoop-Discuss mailing list > [email protected] > http://hcoop.net/cgi-bin/mailman/listinfo/hcoop-discuss _______________________________________________ HCoop-Discuss mailing list [email protected] http://hcoop.net/cgi-bin/mailman/listinfo/hcoop-discuss
