Tony L. Svanstrom [mailto:tony(_at_)svanstrom(_dot_)com] wrote on January 04,
2003:
(Can't write to the list, feel free to reply to it though)
[Paul Chvostek had written:]
PC> > > > * ! ^Message-Id:
<[a-z0-9!#%*+/=?^_{|}~.\$\'\`-]+(_at_)[a-z0-9!#%*+/=?^_{|}~.\$\'\`-]+>$
So what if the Message-Id:-header is RFC-valid or not (for
incoming e-mails)?
Unless part of many rules (a la SpamAssassin) you can't use
this information for anything useful, since it isn't a sure spam-sign.
If you limit your available options to black-and-white analysis, then,
yes, a standard that is broken often in legitimate mail and by popular
clients is not helpful.
However, if -- as you implied with your leading phrasal modifier
("Unless . . .") -- I find it very useful to accumulate data about,
well, for lack of better wording, "icky behavior" in a piece of
mail, and then to apply that data to put me over or pull me back
away from the proverbial edge of a decision-tree concerning just
how spammy the mail actually is.
So I look for the legality of the Message-ID. If it's out of conformity
with the RFCs, I look further at *how* it's out of conformance. If
it's "evul" in ways common to spammers, and if there are other spammy
indicia in the message, then its spam calculus tips it over the edge.
If the manner of non-conformance for the Message-ID is less "egregious,"
however, then I look some more inside my rc.
--
"Weltbedenkend, örtlich lenkend!"
-- Original von Dallman Ross
_______________________________________________
procmail mailing list
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail