[openpgp] Re: Fwd: I-D list for Open Specification for Pretty Good Privacy notification: Changes to draft-gallagher-openpgp-code-point-exhaustion

Hi Heiko & all,

The persistent symmetric keys draft currently proposes to use the public persistent key algorithm ID range 0b10000000-0b10001111 (128-143) for symmetric algorithms, to make it easy to check whether a given algorithm is symmetric.

We could of course shift this over, to e.g. 0b01000000-0b01001111 (64-79). Though, then we're putting them much closer to the currently used algorithms. Whether that's a good or a bad thing I don't know. Perhaps it doesn't matter much.

So, if the WG is in favor of that I can make that change, of course. But, I just personally think that reserving half of every OpenPGP registry for future expansion purposes is unnecessary; reserving a single value (e.g. 255) should be enough.

Once we fill all the IANA registries with "128-255: Reserved", I imagine in might require a lot of arguing to revert that, so I would prefer that we do that arguing and agree on a strategy for future expansion now, if we think it might be needed at some point :)

Best,
Daniel

On Thursday, March 20th, 2025 at 15:44, Heiko Schäfer <heiko.schaefer@posteo.de> wrote:

> Hello,
>
> I'm sure much discussion would be needed to positively decide in favor of the scheme proposed in draft-gallagher-openpgp-code-point-exhaustion.
>
> However, as I understand it, the only immediate question that this draft is effectively asking is: "do we want to reserve bit 8, for now?"
>
> I have not seen any compelling argument against this proposition, so far.
>
> So as a defensive stance, I'm in favor of "reserving bit 8" for all code points where this bit is still unused. At least until there is a compelling argument why the currently vacant second half of a particular code point space is required for some proposed new feature.
>
> Thanks,
> Heiko
>
> On 3/18/25 5:44 PM, Andrew Gallagher wrote:
>
>> Hi, all.
>>
>> Apologies for uploading this so close to the meeting date, but it’s been sitting in my draft documents for some time now and I want to bring it to the list before we make a final decision about code point allocation for PQC and persistent symmetric algorithms.
>>
>> tl;dr: please keep code points >=128 free, because if at some point in the future we needed to extend the algorithm registry to more than a single octet, having these code points available would let us define a UTF8-like self-synchronising encoding that would be fully backwards compatible with all existing wire formats.
>>
>> Yes, I said “all”. Please read the document, and feel free to ask questions! ;-)
>>
>> Thanks,
>> A
>>
>>> Begin forwarded message:
>>>
>>> From: IETF Secretariat [<ietf-secretariat-reply@ietf.org>](mailto:ietf-secretariat-reply@ietf.org)
>>>
>>> Subject: I-D list for Open Specification for Pretty Good Privacy notification: Changes to draft-gallagher-openpgp-code-point-exhaustion
>>>
>>> Date: 18 March 2025 at 16:33:13 GMT
>>> To: [<andrewg@andrewg.com>](mailto:andrewg@andrewg.com)
>>>
>>> Hello,
>>>
>>> This is a notification from the I-D list for Open Specification for Pretty Good Privacy.
>>>
>>> Document: draft-gallagher-openpgp-code-point-exhaustion,
>>> https://datatracker.ietf.org/doc/draft-gallagher-openpgp-code-point-exhaustion/
>>>
>>> Change by Andrew Gallagher on 2025-03-18 09:33 PDT:
>>>
>>> Changed document external resources from: None to:
>>>
>>> gitlab_repo https://gitlab.com/andrewgdotcom/openpgp-code-point-exhaustion
>>> mailing_list https://www.ietf.org/mailman/listinfo/openpgp
>>>
>>> Best regards,
>>>
>>> The Datatracker Internet-Draft tracking service
>>> (for the IETF Secretariat)
>>
>> _______________________________________________
>> openpgp mailing list --
>> openpgp@ietf.org
>> To unsubscribe send an email to
>> openpgp-leave@ietf.org