RE: subject indicators ... ?

Good technical remarks from Lars Marius.

>   1) xtm:subjectIndicatorRef is not an RDF property, but an XML
>      element type. I'm not sure it's good practice to treat XML
>      element types are RDF properties, but maybe it is.

Agreed. In fact I was thinking about xtm:subjectIndicatorRef as a property of a xtm:topic,
but this is only so far in my personal RDF interpretation of the TM data model and syntax.
But might be at some point, in some standard TM-RDF mapping, this will be true ... this is
another story.

>   2) subjectIndicatorRef is a somewhat strange name for an RDF
>      property, since the property does not really constitute a
>      reference. Instead, it's saying that the object is the subject
>      indicator of the subject (subject of the RDF triple, that is).

Right! ... "subject" is indeed a bit overloaded and confusing in this context :))

> One solution to this may be to define skos:subjectIndicator and then
> just say in prose that it is semantically equivalent to the XTM (and
> ISO 13250:200X, where X > 4) concept of a subject indicator.

Logical conclusion. +1

Bernard

Bernard Vatant
Senior Consultant
Knowledge Engineering
Mondeca - www.mondeca.com
bernard.vatant@mondeca.com

Received on Thursday, 14 October 2004 18:05:17 UTC