On Wed, Mar 12, 2003 at 09:37:29AM -0500, Harry Wagner wrote:
> I guess I am confused about what the mission of the UB is? The mission
> statement at http://dublincore.org/usage/documents/mission/ leads me to
> believe that it's purpose is the evolution of the DC vocabulary. I would
> think that considering a new refinement, such as 'hasStatus' would fit
> within this mission. However, I'm not sure how defining methods of encoding
> the vocabulary, such as in RDF schemas, does?
Modeling issues -- such as whether to use rdfs:Datatype or
rdfs:Class to model encoding schemes -- are generally outside
the scope of the UB and within the scope of dc-architecture.
That said, the Usage Board does have an interest in
understanding what the schemas to which the DCMI namespace
URIs resolve are in effect "saying" -- and to intervene if
they are saying something incorrect.
But as you point out, the semantics of a term like "Status" (or
"hasStatus") are very much within the scope of UB. The issue
here is merely to understand whether it is in general better
(in the sense of "good practice") to declare our own terms
with UB-specific definitions in a UB vocabulary or to favor
the use of existing, albeit more generic terms. Is what you
refer to as "defining methods of encoding the vocabulary"?
It is not entirely clear to me whether the choice of encoding
the attribute "Decision" with "dcterms:isReferencedBy"
as opposed to "dcu:decision" is more a UB issue or a
dc-architecture issue. But note that I am now raising this
question on dc-architecture, not dc-usage...
Dr. Thomas Baker [log in to unmask]
Institutszentrum Schloss Birlinghoven mobile +49-160-9664-2129
Fraunhofer-Gesellschaft work +49-30-8109-9027
53754 Sankt Augustin, Germany fax +49-2241-144-2352