Re: [office-metadata] Content duplication and ODF related RDF vocabulary

[Elias Torres <eliast@us.ibm.com>]

Below are my responses. BTW, I think people might be thinking that it's
either RDFa or xml:id and I want to reiterate that it's not the case. RDFa
is a superset of xml:id. RDFa *does* use xml:id to identify resources
withing the content.xml. Supporting RDFa doesn't meant abandoning xml:id,
it simply means allowing us to avoid content duplication and more
in-context metadata.

Svante.Schubert@Sun.COM wrote on 12/18/2006 07:54:41 PM:

> Hi Bruce,
>
> thanks for the example, Bruce.
> I am not a fluent RDFa speaker, yet. I am curious how would you specify
> in RDFa three further semantics aside of the exiting semantic 'foo1'.
>
> 'foo2' is about 'fooA' and 'fooB'
> 'foo3' is about 'fooB' and 'fooC'
> 'foo4' is about 'fooC' and 'fooD'
>
> <text:span style:name="s1"
meta:about="http://ex.net/foo1";>fooA</text:span>
> <text:span style:name="s2"
meta:about="http://ex.net/foo1";>fooB</text:span>
> <text:span style:name="s3"
meta:about="http://ex.net/foo1";>fooC</text:span>
> <text:span style:name="s4"
meta:about="http://ex.net/foo1";>fooD</text:span>

I'd try to answer, but I don't understand what do you mean by "about".
meta:about is to establish the subject, but there's no triple-generating
attribute on the snippets below.

<text:span meta:about="http://ex.net/foo1";
meta:property="rdf:label">fooA</text:span>

This examples does actually generate a triple:

<http://ex.net/foo1> rdf:label "fooA" .

The relationship is rdf:label.

Could you try to re-state your scenario so I can try to show how RDFa would
do it?

>
> For xml:id the content.xml would look like:
>
> <text:span style:name="s1" xml:id="id1">fooA</text:span>
> <text:span style:name="s2" xml:id="id2">fooB</text:span>
> <text:span style:name="s3" xml:id="id3">fooC</text:span>
> <text:span style:name="s4" xml:id="id4">fooD</text:span>

Maybe it's because I'm confused with the overall hypothetical example, but
could you explain what did xml:id accomplished here? Not sure I see any of
the mix and match that you stated earlier between foo2 with(fooA and fooB)
and so on.

>
> I don't want to be unfair, I just want to stress the examples a little
> bit. Although this might be a common case, I want to test more extreme
> situations.

No problem at all. BTW, I'm also confused about Bruce's example. If you use
meta:about alone it doesn't mean anything. meta:about is not establishing
the span as id "http://ex.net/foo";, so I'm not really sure what he really
meant.

>
> Best regards,
> Svante
>
>
> Bruce D'Arcus wrote:
> >
> > On Dec 18, 2006, at 7:50 AM, Svante Schubert wrote:
> >
> >> If for example a quotation is done 100 times, it would be written 100
> >> times in the content and 1 time in the metadata.
> >> This brings the idea of writing it once in the model and referencing
> >> it from the content, similar as XForms is using bindings from the
> >> content to the model.
> >
> > <text:span meta:about="http://ex.net/foo";>foo</text:span>
> >
> > <text:span meta:about="http://ex.net/foo";>foo</text:span>
> >
> > <text:span meta:about="http://ex.net/foo";>foo</text:span>
> >
> > ... then:
> >
> > <rdf:Description rdf:about="http://ex.net/foo";>
> >   <dc:description xml:lang="en">blah, blah</dc:description>
> > </rdf:Description>
> >
> > With the attribute-based approach, it's that simple (though correct my
> > if I'm missing some nuance Elias).
> >
> > Bruce
> >
>