Re: [MMUSIC] IANA registration of SDP attributes

[Flemming Andreasen <[e-mail]>]

On 3/29/16 1:48 PM, Paul Kyzivat wrote:
> On 3/29/16 1:28 AM, Christian Groves wrote:
>> Hello Paul,
>>
>> Please see below.
>>
>> Regards, Christian
>>
>> On 24/03/2016 8:59 AM, Paul Kyzivat wrote:
>>> On 3/22/16 3:11 PM, Flemming Andreasen wrote:
>>>
>>>> Do we have a volunteer to drive the discussion there ?
>>>
>>> OK. I'll volunteer.
>>>
>>> The starting point for this discussion is the current text of section
>>> 8.2.4 of rfc4566bis-16.
>>>
>>> Looking at that, one thing leaps out at me:
>>>
>>> There is a long list of "stuff" that must be provided when defining a
>>> new attribute. IMO it is all *good* stuff, that should be provided.
>>> But it also strikes me that it will be difficult for IANA to evaluate
>>> the adequacy of this stuff itself. So we might want to require an
>>> expert review as a gate to registration.
>> [CNG] What's the role of the SDP directorate in relation to this? I
>> guess they would check any RFC defined attributes but that non-RFCs
>> would also need to be checked? How much of this should be visible on the
>> IANA registry page?
>
> The SDP directorate will only be invoked for ietf documents. The 
> policy for attributes is Specification Required, so it might be a 
> document from another SDO, ore even no SDO. In those cases there would 
> be no SDP directorate review.
>
> After refreshing my memory about the definition of Specification 
> Required, I see that it implies Expert Review. (Flemming is currently 
> listed as the expert.) 
Right - I'm currently the designated expert and hence review all 
external registration requests (i.e. from outside IETF).

> That may be good enough, though we might want to beef up the 
> requirements in section 8.2.4 of 4566bis.
>
> The hard part is seeing to any new actions when *existing* attributes 
> are used in new ways. For that I think the SDP Directorate is all we 
> have to check for that sort of thing. I wonder if there ought to be a 
> checklist of things for SDP Directorate reviews.
>
There probably should, but it shouldn't be any different than for any 
other SDP extension (syntactic and/or semantic) and hence we should have 
that in 4566bis.

> And of course *that* also will only work for IETF documents. If 
> somebody outside the IETF defines a new use for an existing attribute 
> then we have no mechanism to notice that and trigger an update to the 
> registry.
>
True

-- Flemming