The concerns that Bernhard describes are very real, and they have not
escaped attention on the part of the DC community, and the RDF crowd are
very much aware of this.
One way to look at these problems is that details of grouping, whether
elements are optional, and how to encode structure may best be dealt
with in the context of applications, not in the specifications
themselves. THere is a tendency to speak of (and criticise) the DC as
if it were a mature, turn-key metadata system, and of course its not.
I think we've made a lot of progress in defining the semantics (which is
what we set out to do in the beginning). Our success has become the
source of our major problem: people want to implement, and it will be
much better if we can implement in interoperable ways. We've got a long
way to go in this direction. Nice problem to have.
stu
> -----Original Message-----
> From: Bernhard Eversberg [SMTP:[log in to unmask]]
> Sent: Saturday, February 07, 1998 6:16 AM
> To: [log in to unmask]
> Subject: Re: Qualified DC...
>
>
> Andy Powell wrote:
>
> >
> > 2) In terms of embedded HTML META tag syntax, the proposal to have a
> Type
> > sub-element for Relation as in:
> >
> > DC.Relation.Type: isPartOf
> > DC.Relation.Identifier: http://www.ukoln.ac.uk/
> >
> > is not consistent with the proposed usage of other sub-elements.
> >
> Not only that, but it is a very bad idea as long as there is no rule
> that would bind elements to each other that belong together. Since the
> order of elements is not prescribed and there's no no formal
> indication that these two are in fact a pair. From a programmer's
> view,
> especially, repeatable pairs (or triples, ...) of elements become a
> nightmare when there is no such formal indication.
> In short, I call the above a serious design flaw.
> The other suggestion:
>
> > DC.Relation.isPartOf: http://www.ukoln.ac.uk/
> >
> is very much to be preferred. (It is also shorter, but this concerns
> nobody
> these days, I know. Forget it.)
>
> B.E.
>
>
>
>
> Bernhard Eversberg
> Universitaetsbibliothek, Postf. 3329,
> D-38023 Braunschweig, Germany
> Tel. +49 531 391-5026 , -5011 , FAX -5836
> e-mail [log in to unmask]
|