W3C home > Mailing lists > Public > public-media-annotation@w3.org > October 2010

Re: Response to your LC Comment -2393 on Media Ontology spec

From: Thierry MICHEL <tmichel@w3.org>
Date: Fri, 15 Oct 2010 10:47:03 +0200
Message-ID: <4CB81507.2050101@w3.org>
To: James Salsman <jsalsman@talknicer.com>
CC: Silvia Pfeiffer <silviapfeiffer1@gmail.com>, "public-media-annotation@w3.org" <public-media-annotation@w3.org>
Resent to MAWG mailing list for archive purpose.




Le 15/10/2010 08:25, Thierry MICHEL a écrit :
> James,
>
>
> I said that your two codecs Speex and VP8 are not appropriate to be
> included in our *multimedia metadata formats mapping tables*.
>
> Note as a side remark, I have asked you to provide these two tables.
> So far I have not received this information.
> You forwarded this request to jean-marc Valin who responded " I don't
> have time to figure it out". When do you plan to send these 2 mapping
> tables ?
>
> I also I said that we could provide a WebM *multimedia container format
> mapping table*. And I also asked you to provide this table based on this
> template, filled with appropriate info.
> http://dev.w3.org/2008/video/mediaann/mediaont-1.0/mediaont-1.0.html#iddiv4x2x27165
>
>
> Please send us this info if you want it to be included in the spec.
>
> About your formal objection, if you refuse to cancel it, we will
> continue our publication track as is and will discuss it with the W3C
> Director, as required by the W3C process. We don't want to delay the
> Ontology Spec.
>
> Could you please also respond to LC Comment -2395 on Media API spec
>
>
> Thierry
>
>
>
>
>
> 2- that you agree to the MAWG proposal to include a
>  >>>> WebM multimedia container format mapping table
>  >>>>
>  >>>> as you had referred to
>  >>>> http://www.webmproject.org
>
> Le 15/10/2010 07:36, James Salsman a écrit :
>> Silvia,
>>
>> Thierry specifically told me that Speex wasn't approved because it was
>> not a metadata format.
>>
>> Are you saying that Speex is already included in the ontology under ogg?
>>
>> Sincerely,
>> James Salsman
>>
>>
>> On Thu, Oct 14, 2010 at 10:07 PM, Silvia Pfeiffer
>> <silviapfeiffer1@gmail.com> wrote:
>>> James,
>>>
>>> Nobody says that media formats do not contain metadata. But you are
>>> referring to codec data here and not to media fromats. Codec data never
>>> stands on its own and can thus not expose metadata. It always exposes
>>> it as
>>> part of a container. Thus, Ogg Speex already is covered. And you have to
>>> regard metadata for VP8 in the context of WebM.
>>>
>>> Thus, when Thierry asks for a spec for WebM metadata, that will cover
>>> the
>>> VP8 metadata.
>>>
>>> Regards,
>>> Silvia.
>>>
>>>
>>> On Fri, Oct 15, 2010 at 3:07 PM, James Salsman<jsalsman@talknicer.com>
>>> wrote:
>>>>
>>>> Dear Thierry,
>>>>
>>>> I most certainly do not cancel my objection.
>>>>
>>>> Please send me the names and email addresses of all persons who
>>>> purport to claim that media format definitions do not contain
>>>> metadata.
>>>>
>>>> I have asked the VP8 and Speex media authorities to complete your
>>>> ontology forms.
>>>>
>>>> Sincerely,
>>>> James Salsman
>>>>
>>>>
>>>> On Thu, Oct 14, 2010 at 4:15 AM, Thierry MICHEL<tmichel@w3.org> wrote:
>>>>> James,
>>>>>
>>>>>
>>>>> We need to wrap up this issue in order to allow the MAWG to move
>>>>> forward
>>>>> the
>>>>> Media Ontology spec.
>>>>>
>>>>>
>>>>> Could you please respond to this email and say:
>>>>>
>>>>> 1- that you cancel your formal objection.
>>>>> 2- that you agree to the MAWG proposal to include a
>>>>> WebM multimedia container format mapping table
>>>>>
>>>>> as you had referred to
>>>>> http://www.webmproject.org
>>>>>
>>>>>
>>>>> 3- Do you plan to send us the WebM multimedia container format mapping
>>>>> table, whith the appropriate info ?
>>>>>
>>>>> http://dev.w3.org/2008/video/mediaann/mediaont-1.0/mediaont-1.0.html#iddiv4x2x27165
>>>>>
>>>>>
>>>>> Best,
>>>>>
>>>>> Thierry
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Le 12/10/2010 08:54, Thierry MICHEL a écrit :
>>>>>>
>>>>>> Silvia,
>>>>>>
>>>>>> This is exactly what I have told James, in my earlier email in
>>>>>> response
>>>>>> to his last call comment.
>>>>>>
>>>>>>
>>>>>> http://lists.w3.org/Archives/Public/public-media-annotation/2010Sep/0225.html
>>>>>>
>>>>>>
>>>>>> extract:
>>>>>> "To focus on your request to include VP8 and speex in our
>>>>>> Ontology, the
>>>>>> mapping tables are intended for multimedia *metadata formats*.
>>>>>> - VP8 is an open video compression format (codec).
>>>>>> - Speex is a free audio codec for Free Speech.
>>>>>> These are not *metadata formats*, and therefore not meant to be
>>>>>> listed
>>>>>> in our mapping tables".
>>>>>>
>>>>>> As James raised an objection if this 2 formats were not included, I
>>>>>> have
>>>>>> asked him to provide the 2 mapping tables for Speex and VP8, and see
>>>>>> what he could provide ...
>>>>>>
>>>>>> And, as we had concluded in the MAWG that on the container carries
>>>>>> metadata, I have in a couple of emails asked James to provide a
>>>>>> mapping
>>>>>> table for the media container formats WebM.
>>>>>>
>>>>>> an pointed to the template to use
>>>>>>
>>>>>>
>>>>>> http://www.w3.org/2008/WebVideo/Annotations/drafts/ontology10/CR/mapping_table-php.html
>>>>>>
>>>>>>
>>>>>>
>>>>>> I am still waiting for James to provide this information if he
>>>>>> wants to
>>>>>> see it in the spec. A place holder for this containet is already
>>>>>> created
>>>>>>
>>>>>>
>>>>>>
>>>>>> http://dev.w3.org/2008/video/mediaann/mediaont-1.0/mediaont-1.0.html#iddiv4x2x27165
>>>>>>
>>>>>>
>>>>>>
>>>>>> Cheers,
>>>>>>
>>>>>> Thierry.
>>>>>>
>>>>>>
>>>>>>
>>>>>> Le 12/10/2010 03:43, Silvia Pfeiffer a écrit :
>>>>>>>
>>>>>>> James, Thierry,
>>>>>>>
>>>>>>> Speex and VP8 are codecs and don't by themselves have accessible
>>>>>>> metadata.
>>>>>>> They, however, are not distributed as codec data, but rather inside
>>>>>>> media
>>>>>>> containers.
>>>>>>>
>>>>>>> When we talk about Speex in files, it comes in an Ogg container
>>>>>>> and as
>>>>>>> such
>>>>>>>
>>>>>>>
>>>>>>> http://dev.w3.org/2008/video/mediaann/mediaont-1.0/mediaont-1.0.html#iddiv4x2x16776should
>>>>>>>
>>>>>>>
>>>>>>> satisfy your need for a mapping table.
>>>>>>>
>>>>>>> When we talk about VP8 in files, it comes in a WebM container with
>>>>>>> Vorbis.
>>>>>>> Somebody needs to fill in
>>>>>>>
>>>>>>>
>>>>>>> http://dev.w3.org/2008/video/mediaann/mediaont-1.0/mediaont-1.0.html#iddiv4x2x27165!
>>>>>>>
>>>>>>>
>>>>>>> If nobody is doing it, let me know and I'll make the time to dig it
>>>>>>> out.
>>>>>>>
>>>>>>> Cheers,
>>>>>>> Silvia.
>>>>>>>
>>>>>>>
>>>>>>> On Tue, Oct 12, 2010 at 3:54 AM, Thierry MICHEL<tmichel@w3.org>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> Dear James,
>>>>>>>>
>>>>>>>> In your latest email to the MAWG response to your comment, you seem
>>>>>>>> to
>>>>>>>> disagree with the MAWG proposal.
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> http://lists.w3.org/Archives/Public/public-media-annotation/2010Sep/0245.html
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> "I don't agree with the statement that speex or VP8 are not
>>>>>>>> metadata
>>>>>>>> formats. They are audio and audiovisual media formats which contain
>>>>>>>> both metadata and data in their specifications.
>>>>>>>> This disagreement can be resolved if speex and VP8 are included in
>>>>>>>> the
>>>>>>>> media ontology".
>>>>>>>>
>>>>>>>> The group is currently discussing your objection. Speex is an audio
>>>>>>>> format, and the Ontology spec is mainly incorporating images and
>>>>>>>> video
>>>>>>>> metadata formats.
>>>>>>>>
>>>>>>>> To move forward, could you please send the MAWG the 2 mapping
>>>>>>>> tables
>>>>>>>> for
>>>>>>>> Speex and VP8.
>>>>>>>>
>>>>>>>> Please use as template the following tables with the exact name of
>>>>>>>> the
>>>>>>>> property being mapped to, like ma:identifier, ma:title, etc.
>>>>>>>> and the same columns: Relation, How to do the mapping, Datatype,
>>>>>>>> Required
>>>>>>>> vs Optional,XPath
>>>>>>>>
>>>>>>>> templates:
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> http://dev.w3.org/2008/video/mediaann/mediaont-1.0/mediaont-1.0.html#metadata-mapping-table
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> Could you please send these ASAP as we are about to release a new
>>>>>>>> publication of the Media Ontology spec
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> Best regards,
>>>>>>>> Thierry Michel
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>
>>>
>>>
Received on Friday, 15 October 2010 08:47:44 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 15 October 2010 08:47:46 GMT