Re: [Cellar] ATRAC1 codec ID for Matroska

[Michael Richardson <[e-mail]>]

Steve Lhomme <[e-mail]> wrote:
    >> We don't have to move the list.  We can, in the IANA Considerations ask
    >> IANA
    >> to insert the contents of section XYZ into the initial registry.

    > You mean in the IANA Considerations of Matroska mention there will be a
    > registry for codecs (and tags) ?

Let me start again now that I've realized that both documents are in the same
repo.
   ps: I suggest in chapter-codecs that the IANA Considerations and Security
   Considerations be move to the end of the document.

On Matroska document:
a) 27.2 does not provide an updating formula for adding new Chapter Codec IDs.
b) says, "to be defined", and "More codec IDs" can be added later.

What it says now:

5.1.7.1.4.15.  ChapProcessCodecID Element

   id / type / default:  0x6955 / uinteger / 0

   path:  \Segment\Chapters\EditionEntry\+ChapterAtom\ChapProcess\ChapPr
      ocessCodecID

   minOccurs / maxOccurs:  1 / 1

   definition:  Contains the type of the codec used for the processing.
      A value of 0 means native Matroska processing (to be defined), a
      value of 1 means the DVD command set is used; see Section 20.3 on
      DVD menus.  More codec IDs can be added later.

I think that the above section should say instead:

   definition:  Contains the type of the codec used for the processing.
   A value of 0 means native Matroska processing, which is defined in
   [chapter-codecs].
   A value of 1 means the DVD command set is used; see Section 20.3 on
   DVD menus.

    >> >> I don't see ATRAC3 in the document, but maybe I'm looking in the wrong
    >> >> place.
    >> > Indeed, it's missing. That document may not be exhaustive for all
    >> the
    >> > existing codecs in the wild.
    >> >> Please, could the WG clarify where the right registry would be?
    >> Let's talk about what the IANA Considerations are for adding items.
    >> I think that probably Specifications Required would be fine.

    > Specifications Required seems appropriate for codec. They are highly
    > technical parts that do need some (lengthy) explanation. Only "raw" data
    > (no codec) might not need an expert review. Although you still need to tell
    > the width, sign, endianess of data.

I wonder if the entire ChapProcessCodecID element shouldn't just move to
chapter-codecs document?  That does leave it undefined, but also moves it
closer to where it is normatively defined.


--
Michael Richardson <[e-mail]>   . o O ( IPv6 IøT consulting )
           Sandelman Software Works Inc, Ottawa and Worldwide