Re: [payload] Request for consideration: draft-westin-payload-vp8-01

On 03/03/11 15:35, Roni Even wrote:
> Hi Harald,
> In general there is no problem to accept a payload specification as a WG
> document as long as there is just one proposal. In this case I do not think
> there will be a second proposal since this is unusual for non standard
> codecs.
>
> So we can ask for a milestone.
>
> As for progressing the work I did a very quick look and have a couple of
> comments
>
> 1. The media subtype registration in section 6 is not following the
> registration template (see in rtp-howto or in one of the payload RFC e.g.
> RFC5391  section 5.1. You can also look at a video codec like the updated
> H.264 payload http://tools.ietf.org/html/draft-ietf-avt-rtp-rfc3984bis-12
> which is in the RFC editor queue at the moment.)
I'll get that updated. Most of the fields are boilerplate.
> 2. The draft does not have a required congestion control section see
> http://tools.ietf.org/html/draft-ietf-avt-rtp-howto-06 for payload
> specification structure.
Checking: Is this currently draft-ietf-payload-rtp-howto-00?
Patrik did a pass at updating after being pointed at the document as 
part of Magnus' review, that's the diff between the -00 and the -01 version.
> 3. I noticed the division of the data to prediction mode parameters and
> motion vectors partition and DCT/WHT coefficients partitions. Which enables
> better resilience to loss according to the draft. In IP network the typical
> loss is packet loss and not part of a packet so maybe in the congestion
> control section you can provide some guideline about a recommendation of how
> to build packets that can be removed.
Area of active research. I'll get it considered; the comment was part of 
a consideration of using variable-protection FEC mechanisms, which was 
subsequently removed (we'll get back to that at a later stage).
> 4. In the example in section 6.4 I think that you should nave avpf and not
> avp as the profile since you recommend using the feedback mechanism
It's only an example, but I agree.
>
> Going forward  you can update the individual draft regardless of when it
> becomes a WG item
Thanks!
> Regards
> Roni Even
>
>
>
>> -----Original Message-----
>> From: payload-bounces@ietf.org [mailto:payload-bounces@ietf.org] On
>> Behalf Of Harald Alvestrand
>> Sent: Wednesday, March 02, 2011 3:36 PM
>> To: payload@ietf.org
>> Subject: [payload] Request for consideration: draft-westin-payload-vp8-
>> 01
>>
>> Hello,
>>
>> I'm not sure what the formal procedure should be, so I'm just sending
>> email...
>> I think that draft-westin-payload-vp8-01, describing an RTP
>> encapsulation for VP8, is ready for IETF review and possible approval.
>>
>> What steps do we need to take in order to get it on the proper agendas?
>>
>>                    Harald Alvestrand
>>
>> _______________________________________________
>> payload mailing list
>> payload@ietf.org
>> https://www.ietf.org/mailman/listinfo/payload
>