Re: [dtn] Marking RFC5050 as Obsolete?

[Colin Perkins <[e-mail]>]

> On 23 Sep 2019, at 11:28, Carsten Bormann <[e-mail]> wrote:
> On Sep 23, 2019, at 10:16, [e-mail] <lloyd.wood=[e-mail]> wrote:
>> 
>> Explicitly obsolete RFC5050? Do you really want to do this?
> 
> Yes.
> 
> Generally, when experimental protocols are replaced by standards-track protocols, we mark the RFC documenting the experimental protocol as obsoleted by the RFC documenting the standards-track protocol.  E.g., see RFC 3940(*) and RFC 5740.  That, of course, doesn’t mean the *experiment* is “obsolete”, the experiment is done and was a great success, and has led to the standards-track protocol.
> 
> The fact that the obsoleted protocol specification was documented in a different process (here, in the IRTF stream) is not really relevant.  Of course, obsoleting the document in the RFC series doesn’t mean that other SDOs (here: CCSDS) magically also consider it obsoleted; that may be work that needs to be done.
> 
> I’m CCing the IRTF chair for another opinion on whether there is a problem with a IETF specification obsoleting an experimental IRTF protocol specification.  To me, it is obvious that the DTN WG was set up to do exactly that, and there were a few years to stop this process if this outcome wasn’t desired.


The RFC stream managers and RFC Editor had a discussion about this recently. It doesn’t seem problematic for a standards track RFC published on the IETF stream to obsolete an experimental RFC published by the IRTF, provided there’s coordination between the RFC stream managers and agreement that such a status change is appropriate. I’d expect the IESG to check with the IRTF Chair and IRSG before approving such a draft, for example.

I’ve not read the documents in question, so I won’t comment on this specific case. 

Colin



-- 
Colin Perkins
https://csperkins.org/