Re: [office-metadata] RDFa model and xml:id

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

On Dec 13, 2006, at 8:44 PM, Svante Schubert wrote:

> Let me summarize:
> You think there might be an issue for the xml:id approach as you think 
> it might be not be possible to map to RDF (without giving a example).
> Further you think it is not straight-forward (without giving a 
> example).

Svante, you're being needlessly combative in this whole process. Does 
the above REALLY help this conversation in any way?

In any case, I did provide an example; what Patrick posted is not a 
standard RDF way to reference content. Actually, it isn't even valid 
XML. But let's assume:

<meta:metadata xml:id="1">
   <meta:property rdf:resource="http://ex.net/actual_property"/>
</meta:metadata>

Creating a resource and assigning a property to reference a content 
node so as to assign it a property is neither standard nor natural from 
a modeling perspective. It's a convoluted way to solve a nonexistent 
problem.

To review, the model would say:

_:blank_node_1
	meta:property <http://ex.net/actual_property> ;
     <http://ex.net/actual_property> "foo" .

Um, ok, the object of the statement is not in fact the node identified 
with the xml:id attribute. Nope, we have to invent some new -- ODF-only 
-- processing convention to make that association.

Compare that with the much more simple and direct:

	<text:span 
meta:property="http://ex.net/actual_property";>foo</text:span>

.... which in turn models like:

_:blank_node_1 ex:actual_property "foo" .

If we need to agree to disagree here and create two separate proposals 
to solve the "associating metadata with content" requirement, then I 
say we do, but this process is not working. We have our use cases and 
requirements; let's just get one or more proposals done.

And no more snide comments, please, about this is someone how not a 
sound design process.

Bruce