W3C home > Mailing lists > Public > public-esw-thes@w3.org > September 2004

Re: [Proposal][SKOS-Core] Some extensions around dc:subject

From: Charles McCathieNevile <charles@w3.org>
Date: Wed, 22 Sep 2004 11:34:54 -0400 (EDT)
To: Morten Frederiksen <mof-rdf@mfd-consult.dk>
Cc: "'public-esw-thes@w3.org'" <public-esw-thes@w3.org>
Message-ID: <Pine.LNX.4.55.0409221119330.16559@homer.w3.org>

On Tue, 21 Sep 2004, Morten Frederiksen wrote:

>
>Hi,
>
>On Tuesday 21 September 2004 19:40, Miles, AJ (Alistair) wrote:
>> skos:isSubjectOf		a	rdf:Property;
>> 	owl:inverseOf	dc:subject;
>> 	rdfs:range		rdf:Resource;
>> 	vs:term_status	'unstable'.
>This looks good to me.

This would look a lot better if you added the properties

   s:label "Is the subject of"@en
   s:comment "For example, you can describe a particular concept as being the
subject of its definition"@en

or something similar. (I don't think much of vocabularies that don't manage
to give at least a couple of words as a legible name, and a line or so of
explanation, so this suggestion holds for all the properties :-)

>> skos:primarySubject	a	rdf:Property;
>> 	rdfs:subPropertyOf	dc:subject;
>> 	vs:term_status	'unstable'.

>The FOAF equivalent foaf:primaryTopic is an owl:FunctionalProperty, but I'm
>not sure about the consequences re Full/DL if this were to be declared as
>such.

I'm not so concerned about that (because I am not concerned in my life about
the distinctions, although I understand why some people are), but I think it
is a difficult property to use well since it implies that if one thing has
two primarySubjects that the are the same, or that one of the statements is
false. This dichotomy doesn't match the real world very much, IMHO.

>> skos:isPrimarySubjectOf	a	rdf:Property;
>> 	rdfs:subPropertyOf	skos:isSubjectOf;
>> 	owl:inverseOf	skos:primarySubject;
>> 	vs:term_status	'unstable'.
>Likewise, this could be declared as an owl:InverseFunctionalProperty.

Ditto

>I think it would be a good idea to define the range and domain, respectively,
>of the last two, as skos:Concept, to keep people like me from misusing them
>in other contexts. :-)
>At least if the expectation will be that the object/subject is in fact a
>Concept.

I am not sure if I think so or not. I did initially, then wondered about it.

>/me thinks in passing about using a skos:Concept as the object of
>foaf:depicts...

You can do the inverse already, using skos:symbol or something with a similar
name. I would have to check, but I suspect you can also use skos:example to
do the inverse (this might be a more clearly inverse relation).

But I think that foaf:depicts was recently changed to allow for this kind of
usage, no? Anyway, it seems like a useful thing to me.

cheers

Chaals
Received on Wednesday, 22 September 2004 15:34:55 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:38:52 GMT