W3C home > Mailing lists > Public > public-media-annotation@w3.org > February 2009

RE: What is needed to move forward

From: Evain, Jean-Pierre <evain@ebu.ch>
Date: Mon, 9 Feb 2009 12:14:50 +0100
Message-ID: <14AE8514098875488F9FEACD90C747A20459DD@gnvasmail1a.gva.ebu.ch>
To: "Felix Sasaki" <fsasaki@w3.org>
Cc: <public-media-annotation@w3.org>

Dear Felix,

Thanks for the feedback.  

As you know, although I have now gathered some experience in traditional
metadata and schema definition, I am a newcomer in all this 'ontology'

With this limited  knowledge and (non-academic) background, I have sometimes
my own views on the answers to several of the questions I raised... but not
always.  I have tried to write the whole TV-Anytime in RDF-OWL.  This is
certainly quite imperfect (this has been a source of lots of still
unanswered questions on RDF/OWL modelling)  but if a contribution like this
could help I am happy to share it, although I don't want to be seen as
pushing anything (just hoping for feedback). Of course the complexity of
this model has nothing to do with our mapping to XMP.

I have another one on NewsML-G2, which I'll rewrite reusing the approach I
followed for TVA. On this one, I could see the benefits of RDF because the
model is 'simple' and generating instances is easy.  But it is a completely
different matter with TVA and the 'industrialisation' of generating masses
of metadata in this format raises a number of issues (it was so simple with
pure XML). So I hope the promises of semantic web will hold.



-----Original Message-----
From: Felix Sasaki [mailto:fsasaki@w3.org] 
Sent: lundi, 9. février 2009 11:53
To: Evain, Jean-Pierre
Cc: public-media-annotation@w3.org
Subject: Re: What is needed to move forward

Thank you very much for this mail, Jean-Pierre.

For me,

The matrix itself and how mappings have been made would be worth going
through term by term to check its validity.

Would be the next step we should go - if no concrete, example based 
answers to your questions show up.


Evain, Jean-Pierre wrote:
> Dear Felix,
> It is going to be difficult to expose all our respective views on this via
> e-mail.
> My position is very simple, I am going to work on an ontology for
> Audio-visual content AND services. The more I can share and exchange a
> maximum of knowledge on this with the MAWG colleagues, the better. But I
> won't be religious about it.
> I think the discussion of last week with Yves was covering part of the
> overall picture that would need to be clarified like:
> - what is the role of structured XML schemas (what do they do that e.g.
> RDF/OWL doesn't or maybe less adequately like cardinality and type or ID
> management)
> - if I go for an RDF/OWL model, what is its role? How different is it from
> dumb RDF description (why is it NOT rdfising existing XML schemas? What
> makes its value and who will integrate it to exploit metadata instances
> (what would search engines do with it)? 
> - How do I generate instances? From where (transformation from structured
> metadata instances or from a database)? What are the tools to generate
> templates from complex models?
> - etc. Where are tools? Why will/should large communities ignore the
> model and why it makes sense because they don't need it to generate
> metadata?
> As concerns the discussion on SKOS, I have enough data models to deal with
> to possibly write one RDF model for each and built relationships using
> That is an option.  However, I think that reconciling the data models is
> more critically important.
> Another point on SKOS, I can make SKOS transformations from mpeg7/TVA/DVB
> like classifications schemes. However, if you look at it carefully you
> see that this can be done in different ways.  Would you all use AltLabel
> the termID?  This would be worth proposing an harmonised representation.
> then, what do we do with it? We point to term within the CS with a URI
> a # (my preference as the user falls back into the schema if available as
> web resource), or a /, or a '.'? Or do we have a proposal to make to point
> to THE term?
> Quite a lot of work to be done.
> If I look at what we have done with the mapping on XMP:
> - I have the equivalent with the EBUCOre (close to the PBCore), which is a
> sort of refinement of DC that our XMP matrix proposes.
> - The IPTC is doing exactly the same work on XMP (less surprisingly)
> - The matrix itself and how mappings have been made would be worth going
> through term by term to check its validity.
> I would therefore suggest that we dedicate part of the next physical
> going through these issues asking everyone of us to come with a
> presentation. 
> This partly explains why I kept a low profile but it seems we have now
> reached a new level of maturity in our discussion and we shouldn't miss
> turn.
> Best regards,
> Jean-Pierre
> -----Original Message-----
> From: public-media-annotation-request@w3.org
> [mailto:public-media-annotation-request@w3.org] On Behalf Of Felix Sasaki
> Sent: lundi, 9. février 2009 04:17
> To: public-media-annotation@w3.org
> Subject: What is needed to move forward
> Hello all,
> there may be the impression that I do not want a semantic web based
> approach for our ontology. This is not the case, and I very much hope that
> one slice of requirement 11
> http://www.w3.org/TR/2009/WD-media-annot-reqs-20090119/#req-r11
> will be an RDF-based ontology. However, I am very worried with approving
> such an approach *at the moment* for various reasons:
> 1) We have in my view an unclear requirement "allow for several
> abstraction layers like FRBR", without a clear scope description and a
> clear relation to existing formats (see separate thread)
> 2) we have no restriction of the expressive power of an implementation.
> Without such a restriction I am worried about feature creep and as a
> result too much complexity in the ontology.
> 3) Of coures 1) and 2) are chicken-and-egg problems: How to decide about
> them if we don't have proposals on the table? We are missing just these:
> small, but concrete proposals for the "semantic web" conformance slice.
> For the conformance slice "prose" we have
> - the table
> - as a proposal how to relate that to the API the draft at
> (think that each row of the table becomase a subsection 4.2.x)
> - an implementation which makes use of such a proposal, and solves the
> granularity problem Joakim mentioned, see the "dateGeneral vs. pubdate"
> example at
> l
> It would be great to
> - Understand what (if any) problems see people with the "prose" proposal
> as it is described above, and
> - Have somebody creating even a toy implementatin of the ontology and the
> API, the ontology replying to 1), 2) and 3) above. I know about the action
> item for Pierre-Antoine and the SKOS example from Veronique, but I would
> like to see an integrated example, as we have it for the "prose" approach
> already, to know if people only think about SKOS or OWL (which part?), if
> SKOS when which part etc.
> Felix
> -----------------------------------------
> **************************************************
> This email and any files transmitted with it 
> are confidential and intended solely for the 
> use of the individual or entity to whom they
> are addressed. 
> If you have received this email in error, 
> please notify the system manager.
> This footnote also confirms that this email 
> message has been swept by the mailgateway
> **************************************************
Received on Monday, 9 February 2009 11:15:39 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:17:33 UTC