On Tue 02/Jun/2026 15:27:28 +0200 Tero Kivinen wrote:
Alessandro Vesely writes:
Actually ARC provides the authentication information that can be used
to create authorization in the final recipients email filters.
The authorization data that can be derived from authentication cannot
cover indirect mail flows.
I think I disagree on that. I myself do trust iki.fi because I know
the other adminstrators of that service. I know iki.fi will be doing
its best to correctly generate ARC-Authentication-Results headers, and
I know I can trust them as much as I can trust my local mail service
providers Authentication-Results. This trust only applies directly to
me, and that authentication information from iki.fi will be derived to
authorization information by me when I say that trust those headers.
This boils down to the definitions of authentication and authorization.
Authentication can confirm that the message passed through iki.fi.
Authorization (and any trust decisions derived from it) are then applied
according to policies that existed beforehand. In other words, successful
authentication does not establish authorization; it merely provides the
identity information needed to evaluate it.
[...]
In addition, this is only one side of the authorization
relationship: ietf.org has no way to know which recipients are willing
to accept list-modified messages despite a resulting DMARC failure, and
therefore must still munge From: header fields for some domains.
Ietf.org does not have to do that, and I think it is bad idea that it
does. This mungling was caused because people started complaining about
DMARC, and the correct solution should have been to implement ARC in
ietf.org mailing lists and users who subscribe the ietf.org mailing
list should just configure to trust ARC-Authentication-Results of the
ietf.org for ietf mailing lists.
It's that "should just configure". Until the list has some way to know whether
a recipient has configured trust for the list's ARC assertions, it cannot rely
on that trust being present and therefore cannot generally avoid From: munging.
Without a mechanism to communicate or confirm the recipient's authorization
policy, the mailing-list problem remains unsolved.
This is independent of the authentication technology used. Whether the signal
comes from SPF, DKIM, ARC, a hypothetical DKIM2, PGP, or S/MIME, authentication
alone provides no information about which authenticated identities the
recipient is willing to authorize.
Or ietf.org should reject all subscription requests where there is
dmarc p=reject...
This was considered too.
Ietf decided to do bit more user friendly thing and decided that
mangling headers is causing least problems for people...
It is the simplest non-cooperative option, numbered 1.3 in
https://wiki.asrg.sp.am/wiki/Mitigating_DMARC_damage_to_third_party_mail
(Note that /Authorization approaches/ on that page somewhat blurs the
distinction, w.r.t. the two-sided nature of authorization we are discussing.)
Best
Ale
--
_______________________________________________
dmarc mailing list -- [email protected]
To unsubscribe send an email to [email protected]