On 21/05/2026 17:25, John Levine wrote:
A separate issue that they should pay attention to but usually don't
is whether list mail purporting to be from a subscriber actually is
from that person.
Indeed. One of the longstanding puzzles is why mailing lists generally
make so little use of domain-level authentication signals, even when
those are available.
A separate question is how mail systems decide to accept mail from list hosts.
The mail systems I know do not try to assess a remote site's authentication
practices. They look to see whether it's sending mail its users want, based
on the usual spam filtering signals.
I think this is precisely where DMARC changed the traditional model.
Historically, receivers could evaluate list traffic largely on the basis
of reputation and user acceptance. DMARC, however, forces policy
decisions to depend to some extent on sender-domain authentication
semantics as well. A receiver enforcing DMARC may reject list traffic
despite the list itself having an acceptable reputation.
ARC explicitly allows a receiver to override an author's domain DMARC
policy as a matter of local policy, but it gives the mailing list no
reliable way to know in advance whether a particular receiver will
accept such traffic. In practice, that uncertainty is what drives many
lists toward From: rewriting or other transformations. A widely
deployed authorization mechanism, whether OAuth-based or otherwise,
might have reduced the need to overload author-domain authentication
semantics in this way.
Best
Ale
--
_______________________________________________
dmarc mailing list -- [email protected]
To unsubscribe send an email to [email protected]