Re: [MMUSIC] Draft new version: draft-ietf-mmusic-dtls-sdp-09

[Christer Holmberg <[e-mail]>]

Hi,

>>>>>> So, do we still need some changes in draft-dtls-sdp?
>>>>>>
>>>>>> Perhaps we should completely remove section 3.4 (“Multiple SDP 
>>>>>> fingerprint attributes”), and cover it in the 4572 update?
>>>>>
>>>>> Every place in the document that mentions "fingerprint" needs to be 
>>>>> adjusted so it *allows* multiple fingerprints and makes sense if 
>>>>> there are more than one. (You got some but not all of them.)
>>>>
>>>> That change I have done, and it will be in the next version of the document.
>>>>
>>>>> Hopefully anything that deals with the semantics of having multiple fingerprints can be delegated to 4572bis. That includes what it means if a fingerprint is added, removed, changed.
>>>>
>>>> So, you're ok with removing section 3.4 from draft-dtls-sdp?
>>>
>>> Well, as long as "fingerprint value change" is used elsewhere in the 
>>> document it needs to be defined *somewhere*. What 3.4 currently says 
>>> is part of what bothers me, because it only makes sense if you make certain assumptions about the semantics of multiple fingerprints.
>>>
>>> If it is defined in draft-4572-update then that can be referenced.
>>
>> If we do it in draft-4572-update, then we only need to talk about "if the fingerprint changes" and  "inserting one or more fingerprint attributes" in draft-dtls-sdp, without going into the semantics.
>>
>> I did a pre-10 version of the draft, and pushed it to Github:
>>
>> https://github.com/cdh4u/draft-dtls-sdp
>
> OK, that seems to do it, as long as the 4572 update sufficiently explains multiple fingerprints.

Ok, so I'll submit the -10 version, and I think we can finalize the WGLC. Then we can focus on the 4572 updates.

Regards,

Christer