On Thu, 24 Aug 2000, Jane Hunter wrote:
> Sigge, I'm afraid that any discussion on application profiles, which is
> going to satisfy the requirements which you've listed, will become
> technical.
You are, off course, right. The reason I complained about the
technicalities in this discussion is that I'm walking around with
uneasy feelings that, in spite of the fact the DCMI concentrate on
semantics of metadata elements, qualifiers etc, we do take somehow the
XML and RDF machineries for granted.
I do think that this has been mostly good for us, but it does not mean
that we still we need to concentrate much of our work on to a higher
level of abstraction than SCHEMA definitions.
> I also don't want to keep harping on XML Schema but it allows you to
> do more than "extension elements and specialized qualifiers and
> encoding schemes". I think that it could probably provide a solution
> to most of the significant things which people want to do with
> "application profiles" including those you listed:
> - narrowing semantics through derivation by restriction;
> - providing readable labels and descriptions through appinfo and
> documentation tags;
> - enabling deletion, merging and creation of new elements.
And adding to that XML schema provides means of validation. I still
work with DTDs, but XML schema definitions _are_ very elegant :)
> But there are some problems. In order to import elements and types
> from the dc namespace, it requires an XML Schema for the DCES to be
> defined in the dc namespace and this doesn't currently exist. It is
> also quite complex for new users but example schemas can overcome
> this to some extent. It is also not yet a stable W3C Recommendation
> but the XML Schema WG are hoping it will go to Candidate
> Recommendation in September.
>
> If the requirements for 'application profiles' could be clearly
> specified with the help of examples, then I'd be happy to write up a
> document or work with others to show how you can or cannot satisfy
> these requirements using XML Schema.
The foundation for such work did exist once; they were defined in the
"so-called" Dublin Core Reference model [1] (to which DSTC made core
contributions). It is now obsoleted by new and generally more sound
principles of qualification [2].
If your not already a member of the dc-datamodel wg, please join.
> regards,
> jane
>
> +----------------------------------+---------------------------------------+
> | Dr Jane Hunter | Senior Research Scientist |
> | DSTC Pty Ltd | Distributed Systems Technology CRC |
> | Level 7, General Purpose South | Tel : +61 7 3365 4310 |
> | The University of Queensland | Fax : +61 7 3365 4311 |
> | Queensland 4072, Australia | Email : [log in to unmask] |
> +----------------------------------+---------------------------------------+
Sigge
[1] http://www.mailbase.ac.uk/lists/dc-datamodel/files/wd-dcrm-10.htm
[2] http://www.mailbase.ac.uk/lists/dc-usage/2000-02/0129.html
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|