Matthew,
ME> Have you considered and rejected or not considered adding the NBB idea
ME> that I threw out a while back to the CSV spec?
...
ME> Take CSV, and add a new requirement: mail that has failed (as in
ME> there is a MARID record AND the sending IP isn't there AND there's no
ME> ?all) a 2821.FROM check MUST NOT be bounced;
CSV provides a system for obtaining information about a sending SMTP
client. It makes no attempt to dictate how the receiving SMTP server
should use that information.
This is not an accident.
For one thing, that would cross the line into a much fuzzier area, as
some of the discussions about this already demonstrated.
For another, my own experience is that specifications that are too
broad get complicated, take a long time, and ultimately are not very
useful.
ME> instead it MUST either be
ME> refused at SMTP time, or accepted and destroyed. In other words, DON'T
ME> require SRS, but DO require that mail that goes via non-SRS systems not
ME> lead to bounces to systems that didn't originate the original message.
SRS? you mean re-writing stuff? I did not think that CSV said
anything at all about address re-writing. At least, I sure hope we
didn't...
d/
--
Dave Crocker <mailto:dcrocker(_at_)brandenburg(_dot_)com>
Brandenburg InternetWorking <http://www.brandenburg.com>
Sunnyvale, CA USA <tel:+1.408.246.8253>, <fax:+1.866.358.5301>