On Jun 21, 2004, at 12:18 AM, Paul Howarth wrote:
What is happening to Chris is that his emails get a "neutral" result
from SPF checks as a result of the "?all" in the pobox.com SPF record.
In theory, this should be treated the same as a "none" result. However,
in a scoring-based scheme, mail purportedly coming from a domain that
has published SPF records but not getting a "pass" is considerably more
likely to be spam than mail coming from a domain that has not published
SPF records at all. [...]
This sort of thing *must* be happening elsewhere too though.
Well, yes and no. With the exception of aol, earthlink and amazon, the
bulk of the "pass"es that I see are from spammers. So I fully expect
that my statistical filtering (Bayes with autolearn through
spamassassin) is learning to treat passes as weakly spammish. A
trainable statistical system will end up doing the right thing. The
problem is where humans presume that a "pass" means non-spam and
manually add such a rule.
-j
--
Jeffrey Goldberg http://www.goldmark.org/jeff/