Minutes interim-2020-cellar-01: Tue 21:00
minutes-interim-2020-cellar-01-202001282100-00

CELLAR -- DRAFT AGENDA for Virtual Interim Meeting
January  28, 2020      19:00 UTC
                       21:00 Amsterdam
                       15:00 NYC
                       12:00 San Francisco

PRESENT:
        1. Michael Richardson
        1. Spencer Dawkins
        1. Steve Lhomme
        1. Michael Niedermayer
        1. Jerome Martinez
        1. Dave Rice
7. Martin Below

REGRETS:
        *

INFO:
   https://datatracker.ietf.org/meeting/interim-2020-cellar-01/session/cellar
   https://datatracker.ietf.org/doc/agenda-interim-2020-cellar-01-cellar-01/

WEB CONFERENCE:
   https://whereby.com/cellar-interim
   THERE IS NO TELEPHONE DIALIN (You can try this at any time.)
   These notes at: https://github.com/cellar-wg/chair-notes

1. Note Well.  https://www.ietf.org/about/note-well/
2. Accept draft minutes from December 10 meeting (attached below)
        no objections or changes noted.

3. Logistics for Meeting.
   2a) Etherpad for notes
       https://etherpad.ietf.org/p/notes-cellar-virtual?useMonospaceFont=true

   2b) APPEAR.IN is now called "whereby.com"
       https://whereby.com/cellar-interim

   2c) Roll call

4. Welcome to Spencer Dawkins.

5. WG status update
   * EBML -- version 17 was posted 2020-01-27
      https://datatracker.ietf.org/doc/draft-ietf-cellar-ebml/writeup/
      https://datatracker.ietf.org/doc/draft-ietf-cellar-ebml/ballot/

    -17 believed to address all DISCUSS comments now.

   * FFV1 -- version 12 was posted 2020-1-28. Going to Adam Roach responsible AD
                                   doing review.

    - some issues made. Work is half-done.
    - Jerome will take the next move to close some of the To do items.

6. Work on Matroska issues.

** Start New Chapters Pull Request.
-> Conclusion that splitting chapters off into new document is not worth the
effort.

*** does not validate yet, but working on it.  More Text to explain what
Chapters do. *** pull request was too big, and has been split off into smaller
chunks to absorb.

Discussion: do we need some diagrams, etc. for the next meeting?
No: just go through some issues and deal with some concrete pull requests.
Updated, issue #354, that was lingering for some time because of EBML path
format, which is now clear in -17 of EBML, so now we can merge it.  It changes
half of the lines in the document, so need to do this now.

7. Any other business.

Next meeting: 2020-02-25.

---- PREVIOUS NOTES

  CELLAR -- DRAFT AGENDA for Virtual Interim Meeting
  December 10, 2019      19:00 UTC
                       21:00 Amsterdam
                       15:00 NYC
                       12:00 San Francisco

REGRETS:
     1) Reto Kromer

PRESENT:
    1) Michael Richardson
    2) Benjamin Turkus
    3) Dave Rice
    4) Martin Below
    5) Michael Niedermayer
    6) Peter Bubestinger
    7) Steve Lhomme
    8) Jerome Martinez
    9) Dante Bromkovsky

INFO:
   https://datatracker.ietf.org/meeting/interim-2019-cellar-10/session/cellar
   https://datatracker.ietf.org/doc/agenda-interim-2019-cellar-10-sessa/

WEB CONFERENCE:
   https://whereby.com/cellar-interim
   THERE IS NO TELEPHONE DIALIN (You can try this at any time.)
   These notes at: https://github.com/cellar-wg/chair-notes

1. Note Well.  https://www.ietf.org/about/note-well/
2. Accept draft minutes from October 22 meeting (attached below)
        no objections or changes noted.

3. Logistics for Meeting.
   2a) Etherpad for notes
       https://etherpad.ietf.org/p/notes-cellar-virtual?useMonospaceFont=true

   2b) APPEAR.IN is not called "whereby.com"
       https://whereby.com/cellar-interim

   2c) Roll call

4. Establish meeting schedule for 2020!
     Fourth tuesday of the month.  Starting January 28, 2020.
     No meeting in July.
     September meeting might be 22nd, at conference "No Time to Wait". (web
     page for the last one: https://mediaarea.net/NoTimeToWait4 , in Amsterdam
     2020 September) November meeting moved to Tuesday, December 1.

5. WG status update
   * EBML -- version 14 was posted 2019-12-02, on IESG telechat for 2019-12-05
               --- deferred as document was too long for some IESG membes to
               get a handle on
   * FFV1 -- version 11 was posted 2019-10-23, waiting for AD writeup.

Dave Rice reports that changes that occured today was as a result of a review;
some figures and tables are now cross-referenced more clearly. The same was
done to ffv1.  Some help is needed to get appropriate captions on the tables.
Robert Sparks asks that new versions not be updated until there are replies and
instructions. ACTION: MCR to followup with Robert Sparks on IANA issue.
  -> EBML is a "general purpose audio/video container"  Steve to fix #304.

6. Work on Matroska issues.

Split the main Matroska document to take the Chapters in another document?
Only elements required for proper playback should be in the main document.
Document 166 pages today.
Chapters is 6 pages, section 11.  --- but the section is not finished.
Core document 9.3.4 "Tracks" is pages 45->106.

MCR: hears support for splitting the chapter off, but asks if there is really
that much savings. Steve: we could make the document smaller by grouping things
that have only one parent or one child, with some better transformation of the
formatting.

The chapters section is not finished, and there are lot of incomplete sections
which will make the section significantly larger, and so splitting it off into
another document may still be justified. It makes no sense to work on it now,
but make the split off now. Steve will give it a try and see what the savings
is, and if we have a lot references that break, but does not think that this is
the case. Dave Rice is hesistant, not sure he sees enough of an advantage. Tags
and Meta-data are developed asynchronously, but the chapters are not going to
evolve in place. This issue will be deferred.

Peter B:
        ffv1 support in davinci / black-magic.  Would rather do this after the
        document is published. What is status? did Derek's feedback about the
        implementation in GO make it back into the document. DR: Yes, this went
        back into the document about three months ago.  This resulted in a few
        issues being created, and the review was very helpful.

Noting that there are three decoders: ffmpeg, Derek-GO, and Jerome's decoder

7. Any other business.

7.1  Side Data Format for MKV
        Benjamin: side data format for MKV!
        Dave Rice: it is merged into Matroska, can add side data to frames.
        Three competing things how to encode time code into the file. Related
        discussion at
        https://mailarchive.ietf.org/arch/browse/cellar/?q=matroska%20and%20side%20data%20vs%20timecode.
        1) use side data. A debated proposal is at
        https://github.com/cellar-wg/matroska-specification/pull/348. (needsd
        to mentioned in main document) A sample of matroska with side data
        timecode is at
        https://github.com/Matroska-Org/matroska-test-files/pull/5. 2)
        meta-data tag or segment info(needs to be in main document) 3) seperate
        track (codec-like document)

NEXT meeting is 2020-01-28,