> I think we will definitely need to think through this issue as > we work on the multi-alg-rules draft. PQC signature deployment > will very likely necessitate algorithm downgrade protection > rules in DNSSEC, but I don't think those rules should take the > form proposed in jabley draft, which unilaterally privileges > PQC only algorithms. I prefer a more general signaling scheme > of the form you have proposed, though I was thinking of classes > of algorithms (PQC and Classical) rather than individual > algorithms. [...]
I fully agree. I tried to come up with something that is simple and has clear semantics. Signalling classes of algorithms is clearly better for the use cases of downgrade protection and hybrid. But the devil is in the details. Can we define 'this zone is signed with a PQC algorithm' in a way that does not lead to failures if the zone is signed (or some signers sign the zone) with an algorithm that is not supported by the validator? _______________________________________________ DNSOP mailing list -- [email protected] To unsubscribe send an email to [email protected]
