Re: [office-metadata] question for Elias on rdf:type/odf:type

[Patrick Durusau <patrick@durusau.net>]

Elias,

Elias Torres wrote:

>I had not seen that. Where will this be used? In the manifest? If so, we
>don't need odf:type, we need to use rdf:type, much preferred.
>
>  
>
Under 1.1.1 Declaration of Metadata Files in the Metadata Manifest, File 
Types, where it reads:

***

One or more <odf:type> property elements specify the metadata type of a 
metadata file. The property may be used to relate a metadata file to an 
RDF application.

***

And under OpenDocument Elements, where it reads:

***

The <odf:Element> element can be bound to a metadata file by the 
odf:belongsTo property attribute. The odf:belongsTo property attribute 
value is the IRI of a named RDF graph.

***

The purpose as I understand it was to assist applications in associating 
an RDF application with particular metadata.

BTW, this has been discussed and you took the position that we should 
specify less rather than more and to allow implementation experience to 
decide how that should be accomplished.

The question was posed at yesterday's meeting if it we should simply 
allow rdf:type or some equivalent and simply say no more. That is an 
application could use rdf:type for that purpose if it so chooses but 
that the proposal would take no position on how implementations 
accomplish that task.

The suggestion was made by Bruce that we should ask you if rdf:type 
would be more appropriate than developing types on our own.

Taking your statement as an answer, I would suggest amending the 
proposed texts to read:

File Types:

One or more <rdf:type> property elements specify the metadata type of a 
metadata file.


OpenDocument Elements:

The <odf:Element> element can be associated with a metadata file by the 
<rdf:type> property attribute.


And yes, all of these are in the manifest.

>Why is this being introduced? I don't think we can be introducting stuff at
>every meeting, especially if we haven't agreed on it. Did everyone agree to
>it?
>
>  
>
This has been discussed for some time.

With some minor editorial nits, wording order, etc. I think this is the 
last issue of any substance.

>I really don't like the way we work, where the editors add everything they
>want to the specification and then we have to search for changes to keep
>track of them and then decide or argue for something to excluded as opposed
>for it to be included.
>
>  
>
It may not be what you are accustomed to and you are free to suggest 
changes. Personally I think seeing changes in context makes more sense 
but there are merits to what you seem to be implying as well.

Any comments on the changes in wording and to rdf:type as suggested above?

Hope you are having a great day!

Patrick

>Where are we?
>
>-Elias
>
>"Bruce D'Arcus" <bdarcus@gmail.com> wrote on 05/16/2007 11:17:48 AM:
>
>  
>
>>Elias,
>>
>>I just asked the question during the call about why we have an
>>odf:type property rather than just reuse rdf:type.
>>
>>Can you confirm whether you agree with the idea that we need to
>>define a new odf:type property, and if yes, why?
>>
>>I always thought it was a bad idea to use your own type property.
>>
>>Bruce
>>    
>>
>
>
>
>
>  
>

-- 
Patrick Durusau
Patrick@Durusau.net
Chair, V1 - Text Processing: Office and Publishing Systems Interface
Co-Editor, ISO 13250, Topic Maps -- Reference Model
Member, Text Encoding Initiative Board of Directors, 2003-2005

Topic Maps: Human, not artificial, intelligence at work!