Re: [rtcweb] Barry Leiba's No Objection on draft-ietf-rtcweb-rtp-usage-24: (with COMMENT)

[Magnus Westerlund <[e-mail]>]

Barry Leiba skrev den 2015-06-10 18:07:
>> However, my personal view is that IETF need to be able to reference concept
>> and basic definitions from informational documents without making it
>> normative references. Otherwise we will have a lot of things in the downref
>> registry and not be able to find when we end up in actual real issues with
>> protocol mechanisms and their source and maturity. That is why I intended to
>> leave draft-ietf-rtcweb-overview as an informational reference.
>
> Well, I think normative references are all those that you *need* in
> order to understand the document.  If you don't read rtcweb-overview
> and understand the terminology in it... would you be able to properly
> understand this document?

So the overview is the entry point into the whole set of RTCWEB WG 
documents. It gives high level understanding of the relation of all the 
components. The RTP usage I think only one real include, which is the 
definition of WebRTC endpoint.

>
> That said, I'll leave it to your judgment here, and won't argue the
> point further.

I will change this to a normative reference, mostly because I have a 
faulty memory and overview is actually targeted to become an 
applicability statement, i.e. standards track. Thus, this whole circus 
around downrefs don't apply.

 From my point of view the overview is important, but it is only this 
definition you truly need to understand to implement correctly the RTP 
stack. However, you clearly need to understand where the piece fits.

>
> [And as a side point, I'm working with some other folks on an update
> to BCP 97.  The update would *not* require that these sorts of
> documents go into the downref registry, but would instead give
> judgment to the IESG on the right way to handle this while making sure
> that normative references were sufficiently mature.  Because I agree
> that the downref registry was never *meant* for documents that are
> only used for terminology.]

I think that is mostly good. But, we clearly have had a shift in 
interpretation of what a normative reference is from when I started 
being on the IESG 2006 (When the IESG statement was written) and now.

I hope this means that also the expectations are a bit more explicitly 
written up on what an normative reference is. Thus avoiding discussions 
due to shifts in interpretation.

Cheers

Magnus Westerlund

----------------------------------------------------------------------
Services, Media and Network features, Ericsson Research EAB/TXM
----------------------------------------------------------------------
Ericsson AB                 | Phone  [phone]
Färögatan 6                 | Mobile [phone]
SE-164 80 Stockholm, Sweden | mailto: [e-mail]
----------------------------------------------------------------------