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

Re: [FYI] [W3C MAWG] Example of property mapping using semantic technologies

From: Veronique Malaise <vmalaise@few.vu.nl>
Date: Thu, 19 Mar 2009 13:54:47 +0100
Cc: Pierre-Antoine Champin <pchampin@liris.cnrs.fr>, Tobias Bürger <tobias.buerger@sti2.at>, public-media-annotation@w3.org
Message-Id: <0E1F0459-A1BC-459F-8AF5-740EE88F884E@few.vu.nl>
To: Felix Sasaki <felix.sasaki@fh-potsdam.de>
Hi Felix, P-A, all,

Here is (in attachment) an updated version of a mapping example using  
SKOS constructs [1]. The mapping properties in SKOS right now are:
skos:mappingRelation
skos:closeMatch
skos:exactMatch
skos:broadMatch
skos:narrowMatch
skos:relatedMatch
and it seems  that there is no fundamental problem for applying them  
to properties and not skos:Concepts. The update takes the remarks of P- 
A into account (at least in the syntactic point of view, on the  
semantic point of view the mapping proposed is now a  
skos:relatedMapping, which should give rise to less problems), and of  
Felix: the skos properties are now more diverse and more correct too :)

Best,
Véronique

[1] http://www.w3.org/TR/2009/CR-skos-reference-20090317/#mapping

On Mar 18, 2009, at 5:24 PM, Felix Sasaki wrote:

> Hello Veronique, all,
>
> some general questions: what relations in SKOS do you expect to be  
> necessary for the table? broadMatch, closeMatch, exactMatch, ...?  
> Could you give examples for each of the necesary relations?
>
> Felix
>
> 2009/3/18 Veronique Malaise <vmalaise@few.vu.nl>
> Hi!
>
> I used this controversial mapping example to show that we could not  
> use owl:equivalentProperty between the properties (this is an  
> extreme case, but in my opinion even very closely related properties  
> should not be stated as equivalent); in dcterms there are better  
> mappings than with dc:date anyway, so the whole "mapping proposal"  
> is subject to debate: the whole idea was to show an example of a  
> syntax displaying relations in skos between pairs of properties. The  
> "real" file will be based on the mapping table after the reviewing  
> phase. But I agree with your comment and with the "borderline-ness"  
> of this mapping proposition.
> And you are indeed right about the 2nd rdfs:comment, thanks for  
> correcting it!
>
> Best,
> Véronique
>
>
> On Mar 18, 2009, at 2:01 PM, Pierre-Antoine Champin wrote:
>
> Tobias Bürger wrote:
> 1/ I do not agree about the mapping between xmp:CreatorTool (a *tool*)
>   and dc:creator (an *agent*).
>
> We had a discussion about this actually, too. The official defintion  
> of
> dc:creator is "Examples of a Creator include a person, an  
> organization,
> or a service. Typically, the name of a Creator should be used to
> indicate the entity." [2] So a creator can be a service. It is  
> debateble
> if this includes a tool, too.
>
> About dc:Creator, since
> 1/ the DC spec calls it a service rather than a software, and
> 2/ the other two examples (person, organization) are clearly agents,
>
> I tend to interpret "service" here not as *any* software, but as  
> having
> some "agentive quality".
>
> For example, a webcam publishing photos on the web every 10 minutes,  
> is
> making it "on its own", in a sense. Although one could attribute those
> photos to the person/organization that owns the webcam, it may seem  
> more
> relevant to state that the webcam (or the software running it) creates
> the photos.
>
> But this is, in my view, very different from stating that  
> "photoshop" or
> "the gimp" created a photo that I edited with them.
>
>
> Note that I have no definite optinion on whether the software running
> the webcam is an appropriate value for xmp:CreatorTool, though... :)
>
>  pa
>
>
>
>
Received on Thursday, 19 March 2009 12:56:37 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 19 March 2009 12:56:38 GMT