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

[LC Comment ONT] Comments on the relationships between OM and RDF/OWL

From: Ivan Herman <ivan@w3.org>
Date: Tue, 13 Jul 2010 15:28:26 +0200
Cc: W3C T and S Domain <t-and-s@w3.org>
To: public-media-annotation@w3.org
Message-Id: <89952FB1-26E3-47C2-9936-FE65C74E3DE3@w3.org>
Dear all,

I was looking at the MO document

http://www.w3.org/TR/2010/WD-mediaont-10-20100608/

From a Semantic Web point of view. I understand that the ontology you define is not exclusively for its usage on the Semantic Web, ie, it is not only meant to be used in RDF. However, the current document makes it very difficult to understand how this vocabulary _could_ be used on the Semantic Web if this is what one wants.  

The 'mo' terms are defined in 

http://www.w3.org/TR/2010/WD-mediaont-10-20100608/#core-property-definitions

and you define your own syntax to describe each term. It is, however, not clear (to me...) how that syntax translates into RDF if I want to use it this way. Indeed, if I look at ma:identifier, it is defined as { (identifier:URI), (type:String)? }. But the text does not explain whether I have to use

<...> mo:identifier <a-unique-uri>; 
     mo:identifier "somethingelse" .

or whether I have to involve an intermediate node...

To make it more complex, what should I do with ma:location, defined by:

{ (name:(URI|String))?, ((longitude:Float), (latitude:Float), (altitude:Float), (coordinateSystem:String)?)?, }

does it mean that I have something like

ma:location [
  ma:name <uri...> ;
  ma:longitude 1.2344 ;
  ma:latitude  4.56789 ;
  ...
  ]

or something else?

Note that the namespace document at http://www.w3.org/ns/ma-ont does not refer to any RDF or OWL version of the ontology. Ie, on the semantic Web, I cannot follow my nose in finding out anything about that vocabulary in terms of RDF or OWL.

I do believe an explicit bridge from your document to the Semantic Web is needed, in the form of a clear translation of your syntax to RDF, corresponding examples, and a proper RDF Schema (or OWL, if necessary) for the vocabulary. 

I know that the document does include the following remark:

[[[
This specification provides a simple text description and definition of the relationships. An implementation of the vocabulary in RDF [RDF] will be provided as an example in the appendix of this specification.
]]]

but that appendix is missing. That would be acceptable for a Working Draft; I do not think a missing appendix of that sort is acceptable for a Last Call document. It is also not clear whether that appendix would be normative or not; I believe it should.

Finally, looking at the charter of the Working Group[1], the charter says, among other things:

[[[
Ultimately, users of the ontology should also be able to take advantage of Semantic Web technologies, such as the SPARQL Query Language for RDF. 
]]]

ie, my criticism is also based on the charter obligation of the group...

Sincerely

Ivan

[1] http://www.w3.org/2008/01/media-annotations-wg.html

----
Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf







Received on Tuesday, 13 July 2010 13:28:31 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 13 July 2010 13:28:31 GMT