Re: [openpgp] PQC encryption algorithm selection

[Daniel Kahn Gillmor <[e-mail]>]

On Wed 2024-02-07 17:27:33 +0000, Daniel Huigens wrote:
> I'd assume the requirements are specific to the document,

The IANA registries at https://www.iana.org/assignments/openpgp are not
document-specific.  The document will eventually need to provide
explicit, unambiguous guidance for how IANA should update those tables.

Those tables currently DO NOT have a column listing MUST/SHOULD/MAY for
asymmetric algorithms.

> meaning that if draft-personX-openpgp-algorithmY says that it MUST be
> implemented, that only means that in order for an implementation to
> claim it implements *that draft* it must implement that algorithm, not
> that it must do so in order to claim it implements OpenPGP, right?

If the draft is "OpenPGP", and it obsoletes all previous OpenPGP drafts,
then these are the same thing.  That was true for
draft-ietf-openpgp-crypto-refresh, but this draft won't do that.

If we want to be able to talk about whether some implementation is
conformant to "RFC XXXX" (assuming we can progress this draft all the
way to an RFC) then MUST and SHOULD need to be in the text of the draft,
but not in the registry tables.

I don't think this draft is the right place to add MUST/SHOULD to
existing IANA registry tables.

         --dkg