W3C home > Mailing lists > Public > public-swd-wg@w3.org > June 2007

Semantic relation BroaderPartitive/ NarrowerPartitiv

From: Dichev, Christo <dichevc@wssu.edu>
Date: Thu, 28 Jun 2007 19:38:57 -0400
Message-ID: <32B20B2BFD3B344D9F4D7E73CFAA3E3B065099BD@wssuex01>
To: <public-swd-wg@w3.org>

Hi,

We are in a process of implementing Topic Maps to RDF mapping this summer as a plug-in to our TM4L editor  (http://compsci.wssu.edu/iis/nsdl/download.html) TM4L supports four predefined associations: Type-Instance, Superclass-Subclass, Whole-Part and Related. In this mapping project we try to use well recognized vocabularies. There is a pretty good match between the TM4L predefined associations and the SKOS basic semantic relations. One of my concerns is the semantic relation BroaderPartitive/ NarrowerPartitive. At the end of the last week I wrote a private email to Alistair to share my concern (see below). He responded to my question and concerns but also advised me to report the problem to (public-swd-wg@w3.org). 

I think that partitive relations are very important in classifying and organizing the world important not to be addressed. For example, in learning context,  concepts are typically organized in a class-subclass taxonomy. However in terms of learning collections they are organized into chapter-subchapters, sections-subsections, also into Spring-2007, Fall-2007,... etc. This is just one example of a general relation with significant practical implications: any composite item is defined in terms of part-whole, even basic elements like molecules-atoms-electrons.

I am aware of the problems with part-hole. However I feel that some restricted version(s) should be included in SKOS. It is better to have a few standard definitions rather than unpredictable amount of user specifications.

For example I would prefer to map 

            Topic1		--> whole-part	 --> Topic2
                                 *
          Topic1	    --> skos:narrowerPartitive      -->  Topic2

 Instead of using something like the following:


<owl:TransitiveProperty rdf:about=" http://purl.org/dc/terms/isPartOf"/ <http://purl.org/dc/terms/isPartOf> >
    <owl:inverseOf rdf:resource=" http://purl.org/dc/terms/hasPart">
    <rdfs:range rdf:resource="http://www.w3.org/2002/07/owl#Thing"/>
    <rdf:type rdf:resource="http://www.w3.org/2002/07/owl#ObjectProperty"/>
  </owl:TransitiveProperty>
            Topic1		--> whole-part	 --> Topic2
                                  *
              Topic2		--> isPartOf  	 -->  Topic1


Cheers,
Christo


============================================
Hi Christo,

Thanks for letting me know about this. The semantic relation extensions such as broaderPartitive have always been part of a "SKOS Extensions" vocabulary, which is where we put stuff we weren't sure about at the time of publishing the current drafts, see:

[1] <http://www.w3.org/2004/02/skos/extensions/spec/>

We haven't yet talked about whether these will become part of the normative specification of SKOS when we develop the W3C recommendation, or whether these will be published as an informative "Note", or whether they'll be published at all. So there's no problem with you using them, but there's no guarantee they'll be part of the final SKOS recommendation.

To give you an idea of where we're at, the Semantic Web Deployment WG has just published the first draft of the SKOS Use Cases and Requirements [2], and is starting to address some of the main issues [3] (such as interoperability with OWL etc.).

If it's important that SKOS support specific semantic relation extensions, then it would be really helpful if you could send an email to public-swd-wg@w3.org with some basic information about your requirements and your use case - this would give us a basis for raising an issue, which would put it on our agenda.

Cheers,

Alistair.


[2] <http://www.w3.org/TR/2007/WD-skos-ucr-20070516/>
[3] <http://www.w3.org/2006/07/SWD/track/issues>

--
Alistair Miles
Research Associate
Science and Technology Facilities Council Rutherford Appleton Laboratory Harwell Science and Innovation Campus Didcot Oxfordshire OX11 0QX United Kingdom
Web: http://purl.org/net/aliman
Email: a.j.miles@rl.ac.uk
Tel: +44 (0)1235 445440  

> -----Original Message-----
> From: Dichev, Christo [mailto:dichevc@wssu.edu]
> Sent: 23 June 2007 00:09
> To: Miles, AJ (Alistair)
> Subject: SKOS partitive relations
> 
> Hi Alistair,
> 
> We are in a process of implementing Topic Maps to RDF mapping this 
> summer as a plug-in to our TM4L editor 
> (http://compsci.wssu.edu/iis/nsdl/download.html
> <http://compsci.wssu.edu/iis/nsdl/download.html> ). We intended to use 
> the SKOS vocabulary for mapping the TM4L built-in association such as 
> Superclass-Subclass, Related-to and Whole-Part. From the SKOS  Core  
> Guide that I printed a couple of years ago I remember that 
> BroaderPartitive / NarrowerPartitive were included in the proposal. 
> But now it seems that this taxonomic relationship is missing in the 
> last version.
> 
> I would appreciate your comment on this issue since it affects our 
> mapping decisions.
> 
> Christo
> 
> 
Received on Friday, 29 June 2007 04:34:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:17:29 GMT