Re: [openpgp] Weird OIDs in the 4880bis draft

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

On Fri 2023-02-24 19:22:04 +0000, Stephen Farrell wrote:

> It looks to me like we do seem to have consensus to
> include this late change.
>
> So if the people active on gitlab can get any remaining
> clean-up done to the MR and then let the editor know it's
> ready to merge, that'd be great.

I think it's already merged (though the test vectors for the sample v6
certificate, v6 secret key, and the hashed data of the cert's internal
v6 signatures will need to be replaced at some point).  Thanks, Paul!

Still outstanding to merge are three requests that we've reached group
consensus on in the interim and approvals and no objections on the list:

!238 - move SKESK and PKESK from v5 to v6 (includes updated SKESK test vectors)
!220 - use v6 signature trailer size from 8 to 4 octets (avoid hashdata alias with v3)
!219 - bind signature salt size to hash function

and two reference updates/simplifications (neither of which change
anything in terms of what's on the wire):

!216 - update SEC-1 reference
!205 - clarify CFB mode as used in OpenPGP

Beyond that, we have two outstanding questions for the group:

!233 - Forbid generation of ElGamal v5 (now v6) PKESK
!223 - remove session key checksum and padding for v6 ECDH

I'll start a separate thread on both of these.

Once these are resolved, i think we should roll out a new version.

If anyone can offer cleanup of the sample v6 secret key, certificate,
and hashed signature data before the new verson lands, that would be
great!  Note that !219 and !220 will both affect all three samples.

     --dkg