Re: [core] John Scudder's Discuss on draft-ietf-core-oscore-edhoc-10: (with DISCUSS)

[Carsten Bormann <[e-mail]>]

Hi John,

> On 2. Apr 2024, at 23:22, John Scudder via Datatracker <[e-mail]> wrote:
> 
> Regarding the IANA section, I see that the "OSCORE Security Context Parameters"
> registry has a range whose policy is given as "Standards Action With Expert
> Review" so one might argue this ship has sailed... but the more I dig into that
> registry and RFC 9203, the less I understand, so I think it's worth clarifying
> this now instead of propagating confusion. (To continue the "ship has sailed"
> metaphor, I am concerned the ship is lost at sea and we just haven't realized
> it yet.)

Thank you for looking more closely — registration policies are often not receiving the attention they need.
You may want to consult the thread at 

<https://mailarchive.ietf.org/arch/msg/ace/YZrR_wnixKPxmvrbRluKeVnmmbM>

…for a previous discussion of the “Standards Action With Expert Review” label (in the ACE WG).

This form of policy label is standard fare in the CoRE working group.
ISTR that in ancient history there was a case when IESG approved a document with a registration where the actually registered value was arithmetically unsound (there was a reason we wanted experts to look at registration requests), and, ever since, we have been using this combination label liberally.

Should RFC 8126 have picked that up?  Probably.
But then “with” has some quite obvious semantics to it.
(E.g., in fries with ketchup, the fries actually need to be fried.)

Grüße, Carsten