Re: [office-metadata] Questions on the RDFa split solution..

[Bruce D'Arcus <bdarcus@gmail.com>]

On Jan 17, 2007, at 10:36 AM, Svante Schubert wrote:

> Hi Bruce,
>
> thanks for the answers so far, something is left..
>
> Bruce D'Arcus wrote:
>>> Your example:
>>> --------------------- 8< ---------------------
>>>
>>> <text:p meta:class="http://ex.net/foo";>
>>>  <text:span object:id="xyz" meta:about="http://ex.net/x"; 
>>> meta:property="ex:title">Some </text:span>
>>> </text:p>
>>> <text:p meta:class="http://ex.net/foo";>
>>>  <text:span object:id="xyz" meta:about="http://ex.net/x"; 
>>> meta:property="ex:title">Title</text:span>
>>> </text:p>
>>>
>>> ... and in the package:
>>>
>>> <rdf:Description rdf:about="http://ex.net/foo";>
>>>  <ex:status rdf:resource="http://ex.net/Important"/>
>>> </rdf:Description>
>>>
>>> There are three subjects here:
>>>
>>> 1) two blank node paragraphs
>>> 2) an external resource (another document) represent by the URI 
>>> "http://ex.net/x";
>>>
>>> --------------------- 8< ---------------------
>
>>> The parent paragraph is of type "http://ex.net/foo";, therefore the 
>>> literal "Some Title" is as well important?
>>> Does your example even tell that  "http://ex.net/foo"; as the the 
>>> property "ex:title" with the Object "Some Title" as well?
> What about the statement above? Is this valid by coincidence? The 
> Paragraph is using the content of a child, why not the properties? How 
> is this defined?

The class attribute defines the paragraph as a blank node. The about 
attribute on the child defines a different subject, so that property is 
of that subject, not the blank nodes.

>>> It would help me, if you show it in RDF XML. How would all 
>>> statements look like, if the user had exported them to RDF XML?
>>> It is obvious, that import and export of all RDF information as RDF 
>>> XML will be an important scenario for us, which we should keep in 
>>> mind.
>
>> I assume you already know the first two. The third one is an example 
>> of a blank node: a resource without a URI. In RDF/XML it would be:
>>
>> <rdf:Description>
>>   <rdf:type rdf:resource="http://ex.net/foo"/>
>> </rdf:Description>
>>
>> I am just using the RDF type there to hang different propeties off of 
>> it (without having to use a URI).
> Could you write them for me all together in a single RDF XML file? As 
> I think this is what would be exported to the user in the end.

I don't have a lot of time for this Svante (am REALLY busy with other 
work). The above is what would be exported (although it wouldn't be 
very helpful in that context, because there would be no way to 
associate it with the content node).

> Some piece is missing for me, maybe you can help me out. (We might 
> reuse this as one of the scenarios solutions to be compared on).

As I mentioned before, there are three subjects:

_:para_1_blank
_:para_2_blank
<http://ex.net/x>

You cannot reference those blank nodes, because they have no URIs. But 
this is one way to attach some kind of custom properties to the 
content.

>> Note: I'm really not sure how Elias or other RDF experts would feel 
>> about how I've done that, but it seems to be one mechanism.
> You are now our RDFa representive, I think you make a good work on 
> this. Although I am not sure, if I a should favor RDFa or a RDF XML 
> only approach, which references only to the ODF parts.

Let me be blunt: if you guys keep insisting we discuss an RDF/XML-only 
approach at the exclusion of using an RDFa-like mechanism, we will not 
have metadata in ODF 1.2 (or maybe ever).

Bruce