Hi Scott,

XML schema/namespace registration is not required however, so choice b) is 
still possible correct?
We are deciding about registration rules for EPP extensions, and I think a toll 
gate of draft being WG adopted to registrier it with IETF namespace may be the 
right balance.

Viele Grüße / Kind regards
Pawel Kowalik 

> On 22. Oct 2025, at 16:57, Hollenbeck, Scott <[email protected]> wrote:
> 
> 
>> 
>> -----Original Message-----
>> From: Pawel Kowalik <[email protected]>
>> Sent: Tuesday, October 21, 2025 12:31 PM
>> To: Andy Newton <[email protected]>; Gould, James <[email protected]>;
>> Hollenbeck, Scott <[email protected]>; draft-ietf-regext-ext-registry-
>> [email protected]; [email protected]
>> Subject: [EXTERNAL] Re: [regext] Re: WG Last Call: 
>> draft-ietf-regext-ext-registry-
>> epp-00 (Ends 2025-10-27)
>> 
>> Hi,
>> 
>> I think it's ok to say MUST but same time not apply this retrospectively to
>> already registered extensions. Similar approach as proposed in the RDAP
>> extensions draft. Fixing those sins of the past will cause more problem than 
>> it
>> solves.
>> 
>> Also I see additional question: what makes an "IETF specification"?
>> Taking this literarily one would not be allowed to register an extension 
>> with IETF
>> namespace until it becomes RFC. Taking into account extension development,
>> some deployments happen already in the mean time. So the choice is to:
>> 
>> a) have different namespace during development -> downside is change of
>> namespace in already existing deployments and clients after it becomes RFC.
>> 
>> b) say that drafts under IETF change control (so from the point of WG
>> adoption) can use IETF namespace, which is more pragmatic approach imho
>> and solves the issue with abandoned drafts, as those even if abandoned still
>> remain under IETF change control
> 
> [SAH] Thanks for the input, Pawel. I agree that I'd rather not see any change 
> applied retroactively. With respect to choice "b" above, the issue is that 
> RFC 3688 says this: "NOTE: in order for a URN of this type to be assigned, 
> the item being registered MUST have been through the IETF consensus process.  
> Basically, this means that it must be documented in a RFC." Drafts aren't 
> RFCs, so I doubt that a request to register a value in the IETF namespace 
> with an I-D reference specification will be approved.
> 
> Scott

_______________________________________________
regext mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to