On Mon 31/Mar/2025 18:37:58 +0200 Al Iverson wrote:
On Mon, Mar 31, 2025 at 11:30 AM Murray S. Kucherawy <superu...@gmail.com>
wrote:
On Mon, Mar 31, 2025 at 1:56 AM Alessandro Vesely <ves...@tana.it> wrote:
There is room for a lot of compatibility. If we don't change the
canonicalizations, a DKIM1 verifier will be able to verify a DKIM2 signature,
limited to DKIM1 semantics. [...]
I can't tell if this sentence confuses me, or is expressly contrary to what we
probably want here.
That is, it's possible we specifically do not want a DKIM verifier to be able
to claim success over a DKIM2 signature, accidentally or otherwise.
I think I don't want an existing DKIM verifier to be able to provide
some sort of result for a DKIM2 signature. I foresee much confusion
resulting from that.
I don't see the reason why it would be bad or confusing to be compatible.
Obviously, in the sunrise period, senders engaging with DKIM2 will sign both
DKIM1 and DKIM2. The same happened with DomainKeys. Then, after a reasonable
period of time the old thing went away.
DKIM2 specifies some points that may require MTA changes. It may take longer
to get them. Meanwhile, DKIM1 verifiers could be organized to cope with DKIM2
signatures, still limited to DKIM1 semantics. That means, for example, to
accept v=2 or understand h=, but not much else. If that were possible, DKIM2
signing would be sufficient. Otherwise, double signing would persist and it
might not be clear who wins. To wit, I've been double signing RSA and Ed25519
for a while. Guess which one I gave up on.
Best
Ale
--
_______________________________________________
Ietf-dkim mailing list -- ietf-dkim@ietf.org
To unsubscribe send an email to ietf-dkim-le...@ietf.org
