On Thu, 18 Apr 2002, Thomas Baker wrote:
> I am assuming -- please correct me -- that by defining a
> DC-Simple our intention is to promote a generic template for
> metadata records. My mental image, to use a paper-based
> metaphor, is that of a rubber stamp with a simple,
> fill-in-the-blanks metadata form. Rubber stamps are useful
> because you can mass-produce them and they ensure that the
> metadata are presented in a uniform manner. But rubber stamps
> cannot be just "extended" -- to do so, you would have to use
> molten plastic and create, in effect, a new rubber stamp.
> (The rubber stamp, of course, is the XML schema.)
>
I think you are addressing the nub of what I was trying to get at when I
suggested that we should distinguish
- the role of DCMI as a 'standards making body' with a role to advise
implementors how to encode terms
- the role of implementors in expressing the set of terms they wanted to
use for a particular application (which might be as wide and pervasive as
the OAI 'DC Simple' schema)
(Of course the same people may be involved in both activities)
I think it might be helpful to view the 'documents', recommendations
and 'published schemas' that come from both these perspectives as subject
to a different approval and recommendation process.
I am not sure (at this stage) we have the level of feedback to ensure that
DCMI recommendations re usage of 'conforming terms' justifies them
becoming 'recommended'. I think this might happen when we get more
application profiles published in schema registries, or registries that
specialise in user input of suggested terms... but we have not got there
yet.
> I do not have a strong opinion about whether Audience
> should be on that rubber stamp, but the question about
> which criteria to use for that stamp -- all terms or just
> cross-domain? just the DCMES namespace or all elements
> that are not subPropertyOf? -- has implications for how
> we see the rubber stamp.
I think this question could be re-phrased as 'what element
set(s) should DCMI be 'marketing' in its documentation?' it would be
useful to get feedback from wider audience, including those with
experience of training such as Erik Jul??
Perhaps more fundamentally, does DCMI *need* to define these sub-sets?
The issue seems to when one draws up an XML schema of DC terms, in effect
(in my limited understanding) one is creating a 'logical instance' of the
abstract semantics (as expressed by RDF schema) and one starts to have to
make decisions about which elements are included and about data typing.
One has to start assuming requirements of a particular application.... In
this case would it be preferable for the DCMI recommendation to be
explicit about this and create an example encoding of the OAI DC schema,
the DC Library Application profile etc
Rachel
> Tom
>
> --
> Dr. Thomas Baker [log in to unmask]
> Institutszentrum Schloss Birlinghoven mobile +49-171-408-5784
> Fraunhofer-Gesellschaft work +49-30-8109-9027
> 53754 Sankt Augustin, Germany fax +49-2241-14-2619
>
---------------------------------------------------------------------------
Rachel Heery
UKOLN
University of Bath tel: +44 (0)1225 826724
Bath, BA2 7AY, UK fax: +44 (0)1225 826838
http://www.ukoln.ac.uk/
|