Re: [office-metadata] Multiple content nodes representing on RDF subject

[Bruce D'Arcus <bruce.darcus@OpenDocument.us>]

On Dec 22, 2006, at 4:08 PM, Bruce D'Arcus wrote:

> The literal "important" is a property of the resource identified by 
> the URI <http://ex.net/Important>, which is in turn a property of the 
> description of the content node. There is built-in indirection of 
> sorts in the RDF model.

E.g. the way you do it is to *use RDF*:

<text:span xml:id="_foo1">This is an important and </text:span>
<text:span xml:id="_foo2">divvikult</text:span>
<text:span xml:id="_foo3"> text!</text:span>

... then:

   <rdf:Description rdf:about="content.xml#_foo1">
     <ex:status rdf:resource="http://ex.net/Important"/>
   </rdf:Description>

   <rdf:Description rdf:about="content.xml#_foo2">
     <ex:status rdf:resource="http://ex.net/Important"/>
   </rdf:Description>

   <rdf:Description rdf:about="content.xml#_foo3">
     <ex:status rdf:resource="http://ex.net/Important"/>
   </rdf:Description>

Three triples, about three separate resources. You really don't need 
any complicated indirection beyond that (at least as I understand the 
problem you presented).

This is why I and others (like Florian) have consistently been asking 
in these discussions: what is the subject you wish to describe? In this 
case, it is three separate spans, each with their own (local) URI. That 
they each have the same properties is irrelevant; they are three 
different statements.

Or I suppose (though I'd need to think on it some more) you could also 
define a style and include in the style definition:

	meta:class="http://ex.net/ImportantParagraph";

Bruce