> They are available for
> review and comment at the following URLs:
> - http://wip.dublincore.org/test/dces
> - http://wip.dublincore.org/test/dcq
> - http://wip.dublincore.org/test/dcmitype
> It is important that we publish the new schemas, and import
> them into the registry, in a timely manner. To that end I
> would like to invite discussion regarding them, on this
> mailing list, for a period ending 3-15-2003.
Ooops, missed the deadline here (I blame the non-W3CDTF date format used
above... ;-)) but I haven't had time to look at these much till
1. On the basis of the argument in my last message about
hasFormat/hasVersion, I'm wondering whether the use of
dcterms:hasVersion in all these descriptions correct?
My understanding of the semantics of dcterms:hasVersion is that the
object of a statement is another resource which is a
version/edition/adaptation of the current resource, and (at least on the
basis of the description of the inverse property dcterms:isVersionOf),
that means that there is a _difference_ between subject and object which
implies "substantive changes in content rather than differences in
I'm not sure that is what is intended here at all: I think the intended
statements in these schemas are "the _current_ version of this term is
If that is so, then the intention is not to _differentiate_ version of
subject and version of object in the way dcterms:hasVersion does.
So I'm not sure dcterms:hasVersion is quite right? Do we need
someprefix:version? (Now I understand why Harry wanted to use
2. I notice that in
the schema says
http://purl.org/dc/terms/ http://purl.org/dc/elements/1.1/title "Dublin
Core Qualifiers" .
Is that an appropriate title given that that vocabulary includes
dcterms:audience, which is not a "qualifier"?
the schema says
"The Dublin Core Element Set v1.1 namespace providing access to it's
content by means of an RDF Schema" .
I think I commented a while back that this title seemed a bit oblique
because I wasn't sure whether it was describing the vocabulary or the
representation (the RDF/XML document), but if I learn to live with that
ambiguity, I think it would be good if the title literals followed
similar conventions in each schema? The dcterms title is the "odd one
out" at the moment.
And if we do keep this second convention, there's no need for the
apostrophe in "it's"... ;-)
4. In both
the value of the dcterms:modified property for the vocabulary (or the
schema representing the vocabulary) is earlier than the value of the
dcterms:modified property for some of the terms described.
http://purl.org/dc/terms/ http://purl.org/dc/terms/modified "2002-05-24"
Maybe that is OK, if a modification made to a term doesn't constitute a
change to the vocabulary? Just thought I'd check as it looks a little
5. If (as I think a number of people suggested), we use rdf:type instead
of dc:type to describe the relation between the terms and the resources
identified by the URIs
then should we be providing RDF descriptions of these resources as
rdfs:Class-es (since the rdfs:range of rdf:type is rdfs:Class)?
I'm not sure it's strictly necessary but it might be "good practice"?
(And it probably applies even if we stick with dc:type)