[FFmpeg-devel] [libav-devel] FFV1/MKV IETF Session followup

Dave Rice dave at dericed.com
Fri Jul 24 20:23:52 CEST 2015

Hi all,
Further reporting on the IETF happenings below.

> On Jul 22, 2015, at 7:44 PM, Dave Rice <dave at dericed.com> wrote:
> Hi all,
> An HTML5 presentation of today's IETF meeting on FFV1 and Matroska is available at http://recordings.conf.meetecho.com/Playout/watch.jsp?recording=IETF93_DISPATCH&chapter=chapter_2 <http://recordings.conf.meetecho.com/Playout/watch.jsp?recording=IETF93_DISPATCH&chapter=chapter_2>. The FFV1 and Matroska portion of the session is from ~9:00 - ~40:30 and open discussion starts at 14:40. At this point, for those who are interested in participating in further IETF discussions on this topic I encourage you to sign up on the IETF Dispatch mailing list at https://www.ietf.org/mailman/listinfo/dispatch <https://www.ietf.org/mailman/listinfo/dispatch>. Thanks much to those who joined the discussion via jabber and/or meetecho. At this point, we're very interested in comments, questions and feedback on this session and strategies for moving forward. One suggestion from Ted Hardy at ~31:00 that I found helpful was that the communities that develop the Matroska and FFV1 specifications should not s
> imply throw the specifications "over the wall" into IETF, but to move specification development itself to an IETF context in a way that brings our discussions currently in ffmpeg-devel and matroska-devel and integrates them into an IETF context.
> Session minutes by Jean Mahoney (copied from https://mailarchive.ietf.org/arch/msg/dispatch/zHSrtPuSZTxeZxIKzoaT6zuei34 <https://mailarchive.ietf.org/arch/msg/dispatch/zHSrtPuSZTxeZxIKzoaT6zuei34>):
>> FFV1 and Matroska ________________________________________________
>> Presentation: 
>> https://www.ietf.org/proceedings/93/slides/slides-93-dispatch-6.pdf
>> Presenter:  Tessa Fallon, Emmanuel
>> References:
>> - FFV1 Video Specification: https://mediaarea.net/temp/ffv1.html
>> - Github: http://matroska.org/technical/specs/index.html
>> Robert Sparks and Roni Even asked what was Internet-specific about this 
>> work. Jerome Martinez pointed out that Matroska was used with VP8, VP9 
>> and WebM. Steve Lhome said that Matroska was designed for streaming in 
>> networks and that Opus could be stored in Matroska for streaming. 
>> Matroska is supported in Chrome, Firefox and MS Edge.
>> Jerome Martinez said that, while the main purpose of Matroska was 
>> storage, it could be used for transport and that they were looking for 
>> transparency, open source and openness for their specification. They 
>> didn't take it to SMPTE because it's paywalled.
>> Tessa said that although the specifications are already available and 
>> the work is complete, the specifications would benefit from the IETF 
>> review process. Ben Campbell asked if it was ok if IETF take over change 
>> control, and Tessa said that was understood. Ted Hardy said that the 
>> community would need to participate in the IETF or the effort would 
>> fail. Tessa said that she hoped the community and IETF would come 
>> together. Steve Lhome, an original author of Matroska, said he would 
>> continue to participate where ever the work was going to happen.
>> Cullen and Dave Rice pointed out the needs for lossless video. Cullen 
>> felt the specifications didn't support interoperability as they were 
>> currently written, but didn't see the effort as a huge leap for the IETF.
>> Steve Bozko and ??? voiced concerns about the ongoing maintenance 
>> aspects of the work.
>> By raising hands, two people at the meeting showed interest in 
>> contributing. 8-10 people indicated that they were willing to review 
>> documents.
>> Richard Barnes pointed out that netvc people were not in the room and 
>> that they may be interested.
>> ACTION: Dispatch chairs to take the discussion to the list, contact the 
>> netvc list.
> Session notes from Cullen Jennings (copied from https://mailarchive.ietf.org/arch/msg/dispatch/9NGD2XHj9z_UExrd0shx_hsZ5fk <https://mailarchive.ietf.org/arch/msg/dispatch/9NGD2XHj9z_UExrd0shx_hsZ5fk>)
>> Matroska / ffv1  - start 9:18
>> Q. around what is internet standard 
>> - more IETF has right openness 
>> Q. do these need to be done together
>> - could be separated but both useful
>> Q. why not at SMPTE
>> - specs there are behind a paywall
>> Q. does development community meet in person
>> - mostly not
>> Q. what is the support in development community to bring this
>> - support from both
>> Q. Will they join actively 
>> - depends on outcome of if the IETF wants, both communities want to have these 
>> Mary -  clear there is interest in the room
>> Q. Do we have people people willing to review this work 
>> - we had more than a handful of people 
>> - need to get the 
>> * Action Chairs - ping the netvc and codec list 

Tessa Fallon, who chaired the IETF session on Matroska and FFV1 send to notes to relay to the lists:

Reporting back from IETF 93 in Prague on behalf of PREFORMA:

A rather lively session with pushback against our proposal (if you watched the video--well, rest assured it was just as much fun to live it as it was to watch it). However, it appears this type of exchange is par for the course with many IETF meetings. Ultimately 8-10 people showed support by indicating their willingness to review and assist with the specifications (note that this would not include actively contributing to writing/development).  I am told that this is more support than that had by some currently chartered work, which is enormously encouraging.

So to re-iterate: there is significant interest within the IETF for doing technical review of the specifications. What would need to happen for this to be successful is for developers on this list to be willing to participate on the relevant IETF lists to both:
1)show support for review to happen
2)make sure that review is in line with the most recent technical developments

There isn't any formal IETF membership process, and IETF participants are those who choose to participate on the lists and at meetings. Most IETF work takes place on the lists, as the meetings are far-flung (though remote participation is possible and highly encouraged). Thus by signing up for the IETF lists, you effectively have a voice in the IETF.

Currently, FFV1/Matroska discussions take place on this list:
https://www.ietf.org/mailman/listinfo/dispatch <https://www.ietf.org/mailman/listinfo/dispatch>

If you're interested in getting the specifications reviewed by the IETF, please sign up and participate. This won't move forward without support from the developer community--if the developers aren't interested in standardizing the specifications, neither is the IETF.

Post-IETF, we'll be talking with the leaders of related groups within the IETF about how to move this forward.  Enthusiasm for this proposal is very high amongst those who work with a/v and related codecs at the IETF, and folks are willing to help guide us through the next steps.

Best Regards,
Dave Rice

More information about the ffmpeg-devel mailing list