W3C home > Mailing lists > Public > public-media-annotation@w3.org > July 2011

Re: RE : Minutes of the MAWG telecon july 18th 2011 and ACTION for editors.

From: Silvia Pfeiffer <silviapfeiffer1@gmail.com>
Date: Wed, 27 Jul 2011 09:07:38 +1000
Message-ID: <CAHp8n2kncO7_uix9Za2doGJUVHzPza3WDtoaAkw9_4KdRTD82A@mail.gmail.com>
To: Thierry Michel <tmichel@w3.org>
Cc: Felix Sasaki <felix.sasaki@dfki.de>, "Evain, Jean-Pierre" <evain@ebu.ch>, "public-media-annotation@w3.org" <public-media-annotation@w3.org>
On Wed, Jul 27, 2011 at 1:47 AM, Thierry Michel <tmichel@w3.org> wrote:
>
>> On Tue, Jul 26, 2011 at 5:46 PM, Thierry Michel <tmichel@w3.org> wrote:
>>> Felix,
>>>
>>> I agree that the deadline Aug 7th is not the greatest, regarding the
>>> summer holiday. (but remember that these files were requested more than
>>> 5
>>> months ago).
>>
>> That is an unfair statement. The files were requested and provided. It
>> is this continued request to re-edit and re-edit the files that I
>> don't understand.
>
> Do you at least agree that we must provide RDF file that are valid against
> our Ontology ? Because this is our goal.


Mine validated when I sent them through, see
http://lists.w3.org/Archives/Public/public-media-annotation/2011Jun/0058.html
.


>>> Now why Aug 7th ? Just to match the specification SoTD:
>>>
>>> http://www.w3.org/TR/2011/CR-mediaont-10-20110707/
>>>
>>> "This specification will remain a Candidate Recommendation until at
>>> least
>>> 01 August 2011 and until the API for Media Resource 1.0 specification
>>> enters Candidate Recommendation."
>>>
>>> And the API for Media Resource 1.0 specification LC review ends  07
>>> August
>>> 2011.
>>> see:
>>>    http://www.w3.org/TR/2011/WD-mediaont-api-1.0-20110712
>>>
>>> The idea was to collect all files at that date. Then we will have to
>>> check
>>> again all the RDF files: We must make sure that not only the RDF files
>>> are
>>> valid (against the RDF validator) but also they validate against the
>>> Media
>>> Ontology (for this we use Protégé). If anyone has a better idea idea,
>>> the
>>> Group will be happy to do otherwise. Thanks to Werner and Martin to
>>> provide this "how to use" guide to validate the RDF files with Protégé.
>>>
>>> We can not move to PR with RDF files in a testsuite that do not validate
>>> the Media Ontology scheme.
>>>
>>> I am fine with extending the deadline, but I would not want to delay
>>> eternally this spec. At some point if there are RDF files that do
>>> fulfill
>>> the guidelines, we will have to remove the corresponding mapping Format
>>> from the Ontology spec (according to the CR exit criteria).
>>>
>>> I personnally have no real RDF expertise. But I think the guidelines are
>>> very helpfull (thank for all the people who have given input here). And
>>> it
>>> has helped me a lot to do my format. Also looking at other correct file
>>> is
>>> a lot of help.
>>>
>>> There are a  few lines mandatory like the following to add to the file
>>> to
>>> allow validation. Also providing in parallel the HTML column of the
>>> "RDF-tested" column is affordable to everyone.
>>>
>>> <?xml version="1.0" encoding="UTF-8"?>
>>> <!-- created by hand by Thierry MICHEL (W3C) for Dublin Core example
>>> file
>>> : version 4 -->
>>> <rdf:RDF xmlns="http://www.w3.org/ns/ma-ont#"
>>> xmlns:ma="http://www.w3.org/ns/ma-ont#"
>>> xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
>>> xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#"
>>> xmlns:owl="http://www.w3.org/2002/07/owl#"
>>> xmlns:xsd="http://www.w3.org/2001/XMLSchema#"
>>> xmlns:more="........." >
>>> <owl:Ontology>
>>>    <owl:imports rdf:resource="http://www.w3.org/ns/ma-ont"/>
>>> </owl:Ontology>
>>
>>
>> It seems to me there are 25 formats and thus 25 files to edit.
>
> Right. And each have an editor.
>
> If this
>> is all the change you need, why not just go ahead and do it? It will
>> be done within an hour or two rather than having to bug every editor,
>> which will take you weeks.
>
> This piece of code only allows to validate the RDF files with Protégé
> (Just like you would add a DTD or a Shema link to validate an XML file).
>
>
>>
>>> Then one has to make sure the file validates against the Ontology using
>>> Protégé.
>>
>> And that's where I decided that I wasn't able to help you. In one of
>> your emails there is a list of 3 or 4 pieces of software that editors
>> are being asked to install and test their files on. To me it seems
>> that the approach of distributing out the editorship to this many
>> people does not scale. Why would everyone need to install all this
>> software and go through all these processes when one central person
>> that understands it can do it within a short amount of time?
>
> When all the files are provided. We decided that we would assign
> reviwer(s) to make sure the files do validate. Meanwhile it is up to the
> editors to provide complete and conformant files to the guideline.

OK, maybe text files can be extended to cover all of the metadata that
you are asking for. That's not so simple for binary media files. I've
done what I can to put metadata in there. Then I wrote the RDF files
that represent the content of the media files and made sure they
validated, see http://lists.w3.org/Archives/Public/public-media-annotation/2011Jun/0058.html
.


>> Just the amount of time that it takes me to understand what you are
>> asking me to do would probably be shorter if you just did it yourself.
>
> I could of course run your files in Protégé (though I haver no more
> expertise than you have with this software.
>
> But more is that one needs to chech is the vocabulary used is OK with the
> one of the Ontology.

If that is about vocabulary used in the RDF files, then it's a matter
of making all the files consistent, which is better done by a single
author than by everyone who provided the format-specific files.


> I have no RDF and OWL expertise  nor expertise for every formats provided
> (Only editors do) to check all the RDF files.

Who are those editors then that are supposed to have RDF and OWL
expertise. I'm certainly not one of them. I only have expertise in
encoding the audio and video files, which I did, and whose metadata
content that I encoded I fully described in RDF files that I authored
manually by inspecting how others had done theirs (see
http://lists.w3.org/Archives/Public/public-media-annotation/2011Jun/0002.html).
To make those RDF files persistent across the collection isn't my
expertise and I'm sure to get it wrong, which will just end in more
cycles going back and forth.


Regards,
Silvia.
Received on Tuesday, 26 July 2011 23:08:26 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 26 July 2011 23:08:27 GMT