Re: [payload] New version of draft-ietf-payload-vp8

[Colin Perkins <[e-mail]>]

> On 22 Sep 2015, at 18:32, Harald Alvestrand <[e-mail]> wrote:
> 
> On 09/22/2015 04:40 PM, Ben Campbell wrote:
>> Hi Harald,
>> 
>> I will check with Alissa to see if her concerns have been addressed.
>> 
>> Has there been closure on Elwyn's Gen-ART telechat review and on
>> Colin's comments on the payload list? I was under the impression that
>> at least Colin's comments might result in an update (although we can
>> handle minor changes with RFC editor notes)
>> 
>> Also, I had some comments a while back to which I do not recall seeing
>> a response:
>> 
>>> -- section 3, Payload Type: "... the VP8 RTP payload profile MUST
>>> assign a dynamic payload type number ..."
>>> 
>>> If I read correctly, Colin Perkins objected to this change. Did you
>>> consider his objection?
> I asked in response if we can delete the whole thing about payload type
> assignment, since it's not a proper concern of the payload type. I did
> not get a response, so I thought it couldn’t be important.

I thought the question was for others in the working group. My opinion is that we do need text here, and I proposed something I thought appropriate, although I’m open to other suggestions. The current text does need to be changed, however.

Colin





>>> 
>>> -- 4.2
>>> 
>>> I didn't see anything addressing Elwyn's Gen-ART (last call) review
>>> question about: “What happens if L=1 but both T=0 and K=0 so that
>>> there is no TID value present? Or indeed if T=0 but K=1 so that the
>>> TID field is there but 'MUST be ignored by the receiver'  (definition
>>> of TID field)?” Did I miss something?
> 
> On this matter, we thought the text was clear that those streams would
> be illegal, and we don't specify the processing of illegal streams. But
> if it isn't clear that it's illegal, we may want to reconsider.
>> 
>> Thanks!
>> 
>> Ben.
>> 
>> 
>> On 10 Sep 2015, at 11:24, Harald Alvestrand wrote:
>> 
>>> On 09/10/2015 03:28 AM, Colin Perkins wrote:
>>>>> On 9 Sep 2015, at 23:12, Ben Campbell <[e-mail]> wrote:
>>>>> 
>>>>> On 9 Sep 2015, at 16:01, Henrik Lundin wrote:
>>>>> 
>>>>>> We have made no technical changes to the document, but have added
>>>>>> a number of grammar updates and clarifications. Some of the
>>>>>> important clarifications are as follows:
>>>>> Payload participants: Please note that there are in fact changes to
>>>>> 2119 language in this version. I won't quibble about whether or not
>>>>> those count as "technical" changes, but please take a moment to
>>>>> review them. If people object to any of them please say so asap ,
>>>>> as this draft will likely be on the agenda for the IESG telechat
>>>>> next week.
>>>> The new text about the Payload Type in Section 4.1 says:
>>>> 
>>>> Payload type (PT):  In line with the policy in Section 3 of
>>>>  [RFC3551], applications using the VP8 RTP payload profile MUST
>>>>  assign a dynamic payload type number to be used in each RTP
>>>>  session and provide a mechanism to indicate the mapping.  See
>>>>  Section 6.2 for the mechanism to be used with the Session
>>>>  Description Protocol (SDP) [RFC4566].
>>>> 
>>>> This would be okay, except that RTP profiles that don’t derive from
>>>> RFC 3551 could be defined. A better wording might be:
>>>> 
>>>> Payload Type (PT): The assignment of an RTP payload type for this
>>>> packet format is outside the scope of this document and will not be
>>>> specified here.  It is expected that the RTP profile for a particular
>>>> class of applications will assign a payload type for this encoding,
>>>> or if that is not done, then a payload type in the dynamic range
>>>> SHALL be chosen by means of an out-of-band signaling protocol. See
>>>> Section 6.2 for the mechanism to be used with the Session Description
>>>> Protocol (SDP) [RFC4566].
>>>> 
>>>> 
>>>> In Section 4.2, rather than add a note that “this X bit is not to be
>>>> confused with the X bit in the RTP header”, would it not make sense
>>>> to rename this X bit?
>>> 
>>> We considered this option, but decided that having a large existing body
>>> of documentation outside of the IETF that refers to the "X bit" and
>>> having a payload type registration that called it something else, it ws
>>> better to keep the name and just make a note that it's different.
>>> 
>>>> Similarly for the M bit in the extension data fields, and the P bit
>>>> in the payload header (which isn’t mentioned). Or, at least,
>>>> annotate each mention of these fields to say which X, M, or P bit it
>>>> required (e.g., “the P bit in the payload header” or “the P bit in
>>>> the RTP header” rather than “the P bit”).
>>> 
>>> This can be done, and probably should be done for any reference that
>>> occurs outside of the section describing the VP8 payload descriptor
>>> (section 4.2) - I couldn't find any such references.
>>> 
>>> There is no P bit in the VP8 payload descriptor, but there is one in the
>>> VP8 payload header, which is why we missed warning about the name
>>> collision there. It's never referred to in text.
>>> 
>>> 
>>>> 
>>>> Colin
>>>> 
>>>> 
>>>> 
>>>> 
>>>> 
>>>> 
>>>> 
>>> 
>>> 
>>> -- 
>>> Surveillance is pervasive. Go Dark.
>>> 
>>> 
>>> _______________________________________________
>>> payload mailing list
>>> [e-mail]
>>> https://www.ietf.org/mailman/listinfo/payload
> 
> 
> -- 
> Surveillance is pervasive. Go Dark.
> 
> _______________________________________________
> payload mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/payload



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