On Tue, 27 Jan 2004, Pete Johnston wrote:
> On Tue, 27 Jan 2004 16:18:04 +0000, Bill Oldroyd <[log in to unmask]> wrote:
>
> >Would it be correct to interpret the Abstract Model as allowing a structure
> >along the following lines (abusing Relax NG Compact to show an example):
> >
> >element metadata {
> > element dc:title { text="journal article" },
> > element dcterms:isPartOf {
> > element dc:title { text = "journal title" }
> > }
> > }
> >
>
> I think the answer is that the Abstract Model is completely silent on
> whether that XML tree is "allowed" or not.
To go a little further...
If you are asking "does the abstract model allow a syntax like the one
above?" then the answer is "that is a syntax issue and the abstract model
doesn't concern itself with syntax". Which is basically what Pete is
saying above.
If you are asking "does the syntax above make sense in the context of the
abstract model?" then the answer is "yes, I can interpret that syntax as
being meaningful and compatable with the abstract model".
I would intepret that syntax as follows:
- there is one record
- the record contains two 'descriptions' about two 'resources'
- neither 'resource' has a 'resource URI'
- each 'resource' has a dc:title 'property'
- the dc:title 'value string' for the first 'resource' is "journal
article"
- the dc:title 'value string' for the second 'resource' is "journal title"
- the two 'resources' are related by a dcterms:isPartOf 'property'
but that's just me guessing! The point is that it is up to you to define
what your made up syntax means in terms of the abstract model. In
particular, it is up to you (and/or Relax NG) to define what nesting
means (sorry I'm not familiar with Relax NG).
Of DCMI's currently defined syntaxes, only RDF/XML explicitly allows the
kind of nesting shown above.
Andy
--
Distributed Systems, UKOLN, University of Bath, Bath, BA2 7AY, UK
http://www.ukoln.ac.uk/ukoln/staff/a.powell +44 1225 383933
Resource Discovery Network http://www.rdn.ac.uk/
|