Re: [openpgp] Marker packet for OpenPGP-NG (was: Confirming open questions discussed at IETF 114)

[Paul Wouters <[e-mail]>]

On Wed, 12 Oct 2022, Werner Koch wrote:

> Given that the crypto-refresh I-D heavily deviates from OpenPGP as
> specified in RFC-2440 and RFC-4880 and deployed practise I have two
> suggestions to avoid confusion:

I am confused what confusion there would be? See dkg's reply to Kai
a few days ago. The versions for keys and packets are clear and
unambiguous.

If you want to implement something different from OpenPGP, all the
IANA registries at https://www.iana.org/assignments/pgp-parameters/pgp-parameters.xhtml
have options for Private Use values that can be used for such a
purpose.

> 1. All data including key packets shall make use of a marker packet
>
>   5.8.  Marker Packet (Tag 10)
>
>   This packet was used by certain PGP versions and flagged as obsolete
>   in RFC-4880.  With this specification it is repurposed to mark the
>   data and keys constructed as specified by this specification and its
>   successors.

I would not be in favour of this, as its main use seems to be for supporting
forks of the IETF OpenPGP specification.

> 2. Clearly identify the new specification as a new version of OpenPGP
>   with only limited backward compatibility.  For example
>     OpenPGP-NG or OpenPGP/2 or ModernPGP or IETFPGP
>   or maybe just PGP (if Broadcom allows for such a use)

I do not believe this is neccessary. Again, see dkg's reply to Kai's
email a few days ago.

If gnupg has deployed versions of somrthing incompatible with OpenPGP as
per 4880, and has done so by using unassigned values from the above IANA
registries, please let us know so that crypto-refresh can skip those
numbers and mark those numbers in the IANA registry as RESERVED. In that
case, it would be recommended that your implementation starts migrating
to Private Use numbers within those IANA registries.

Paul