> From [log in to unmask] Thu Feb 15 19:28 MET 2001
> X-Sender: siglun@gungner
> MIME-Version: 1.0
> Date: Thu, 15 Feb 2001 19:32:36 +0100
> From: "Sigfrid Lundberg, Lub NetLab" <[log in to unmask]>
> Subject: Re: [POLL] URI for DCES, DCQ and DCT
> To: [log in to unmask]
>
> On Thu, 15 Feb 2001, Thomas Baker wrote:
>
> ...
>
> > I'm reasonably happy with the URIs above but wonder where the Usage
> > Board could put any new DCMI elements outside the core. Something like
> > http://purl.org/dc/etc/ as a catch-call for non-core elements and
> > qualifiers would do the trick (we'd need a better name).
>
> The modularity is a virtue. On the other hand, it has its price, in
> readability and looong lists of name space URIs.
Lists of namespaces will be unavoidable. Their length will vary in application profiles.
For a five statements record you will not need a 1000 packages.
>I regard this is issue as
> a technical/architectural one. In Eric's /2000/03/13-* musings, the
> element refinements and the encoding schemes lives in one file. This may,
> or may not be the optimal solution. Again, I think this a technical
> issue. When these beasts are compiled into triples -- arcs and nodes. They
> will live on one single database anyway, and you cannot tell the
> difference.
>
> > Moreover, http://purl.org/dc/qualifiers/ is arguably mis-named already
> > for those who see Element Refinements as Elements.
>
> We talk about elements in two senses. Unfortunately we talk about *ML
> elements, like <dc:title>...</dc:title>, and we talk about the semantics
> 'title', the name given to the resource by its creator.
>
> In the DCMI jargon, we talk about elements and qualifiers. These
> elements are different things. What is the refined title, in dotty
> notation DC.title.alternative corresponds according to Eric's musings
> to <dcq:alternative>....</dcq:alternative>. dcq:alternative is a *ML
> element. But alternative is a DCMI refinement.
>
> This is unfortunate, but we can make life easier by refering to things
> like <dcq:alternative> as 'tags' (the SGML/XML communities would
> call them elements, though).
>
> Sigge
>
> ________________
> Sigfrid Lundberg, Ph.D., . [log in to unmask]
> Lund University Library, http://www.lub.lu.se/~siglun/
> Netlab, PO Box 3, S-221 00 Lund phone +46 (0)46 222 36 83
>
|