On Sat 16/May/2026 19:27:05 +0200 John R Levine wrote:
On Sat, 16 May 2026, Alessandro Vesely wrote:
Mainly to understand whether the ARC experiment revealed a missing
authorization semantic.
ARC can preserve authentication results across intermediaries, but that
is not quite the same thing as establishing whether a mediation path is
acceptable.
It can't even do that, since you can't tell whether links in the ARC chain
beyond the most recent are real or fake.
I agree that, operationally, the receiver can only act on the most recent ARC
set.
That's why DKIM2 puts a signature on each hop, so the recipient can verify
them all.
I'm still not seeing any use in distracting DKIM2 with ARC stuff, nor in
spending more time trying to resuscitate ARC.
The question of authorization semantics for indirect mail flows concerns the
design assumptions behind mediation handling in general, not necessarily
limited to ARC as deployed.
Seeking a correct diagnosis does not imply an attempt to resuscitate ARC, and
may still be useful independently of the particular mechanism used to
authenticate the chain.
Best
Ale
--
_______________________________________________
dmarc mailing list -- [email protected]
To unsubscribe send an email to [email protected]