Re: [office-metadata] Revision and stable ids (was: Re:[office-metadata] Rought notes 14 March)

[Svante Schubert <Svante.Schubert@Sun.COM>]

Hi Bruce,

Bruce D'Arcus wrote:
>> I would then infer that in-context metadata would only be used for 
>> non-revision specific information.
>
> This gets us back to the fields. I would call the field information 
> (subject URI and parameters) to be "non-revision specific." For that 
> reason, I would actually say we should exclude the xml:id approach on 
> fields and require subject (or object?) URIs on them.
I guess xml:id is not only intended to be used by metadata, somebody 
else might want to identify (e.g. point at) such an element, therefore I 
think xml:id should not be removed here.
>
>> The other approach is the Named Graph approach.
>
> ... I prefer this option myself. It's simple and elegant.
I like it, as a RDF plug-in using Named graph might express that it 
works on it's own data and is only responsible for this.
One possible implementation for plug-ins would be that a plug-in only 
gives guarantees for it's own data, it won't change data from other 
plug-ins.

Svante.