On Fri, 18 Aug 2000, Carl Lagoze wrote:
<snip>
> The idea, as I understand
> it, is to allow communities to create "profiles" that use selected
> elements with contraints that correspond to the communities needs. As
> Rachel points out in her paper, and as others have noted, a good number
> of communities will require a record format that includes semantics
> outside the bounds of DCES sementics. Thus, the notion of the
> application profile abstraction, which accepts the fact that in a number
> of cases a community may want elements that do exactly fit into those
> defined within DCES but also accepts the fact that the "records" needed
> by these communities will require elements from other namespaces.
Yes :-) The reality of an implementation's requirement to mix two or more
element sets is a fundamental of what we are trying to capture with the
idea of application profiles.
>
> In this context, however, the problem I raised earlier exists. We want
> more expressive "creator semantics" including affiliation which doesn't
> match the semantics of the DC "creator" element. So, do we have a DC
> creator element that co-exists with a more expressive OAI creator
> element?
>
I am proposing that, as a solution to your problem, you excise DC:Creator
from your application profile... you as an implementor have a requirement
for a view of your metadata that includes the richer OAI creator element
alongside simpler DC elements.
So you define the 'Cornell application profile' as (say) five elements
from DCMES : title, subject, format, description, date; along with another
element - OAICreator- an expressive element that is already defined in the
OAI namespace. These elements all make up the Cornell application profile.
You will have already published an OAI schema to define OAICreator and
other OAI elements.
Meanwhile for a particular target audience you also define a 'view'
of your metadata that consists of OAI creator, OAI corporate affiliation,
OAI metadata owner, OAI identifier, IEEE LOM audience level, and MARC $400
notes field (?!). You publish an OAIRich application profile for this.
So, yes there is 'co-existence' of rich and simple creator description
but only so far as that views of one or other might be derived from the
same data. These 'views' will be served to different audiences and as I
see it the requirement is that we facilitate defining these views by
publishing 'application profiles' which declare the particular schemas.
This will allow for re-use, reference by tools, sharing definitions etc
Maybe I am being too simplistic?
Rachel
---------------------------------------------------------------------------
Rachel Heery
UKOLN (UK Office for Library and Information Networking)
University of Bath tel: +44 (0)1225 826724
Bath, BA2 7AY, UK fax: +44 (0)1225 826838
http://www.ukoln.ac.uk/
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|