A slightly modified version of the draft W3C XML Schemas for Qualified
Dublin Core is now available at
http://www.ukoln.ac.uk/metadata/dcmi/xmlschema/20021007/
This supercedes the version at:
http://www.ukoln.ac.uk/metadata/dcmi/xmlschema/
The only substantial changes in content are:
- the addition of a complexType to represent the RFC3066 encoding
scheme, which had been omitted in the previous version
- the addition of element declarations to represent the recently
approved element refinements (dateAccepted, dateCopyrighted,
dateSubmitted, educationLevel)
Some minor changes were made to the organisation/design of the schemas
- the introduction of an abstract element ("any") from which all the XML
elements representing the DCMI elements are derived (inspired by Dave
Beckett's use of that convention in the draft W3C XML Schemas for Simple
DC in RDF/XML at http://ilrt.org/discovery/2002/01/dcxml-xsd/ )
- the removal of redundant type attributes on each element declaration
- the removal of the invalid namespace declarations for the
http://www.w3.org/XML/1998/namespace namespace
Below are some comments on some of the other issues raised since the
schemas were made available to this list in July.
On the question of restricting the use of encoding schemes to specific
DCMI element/element refinements,
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0209&L=dc-architecture&T=
0&F=&S=&P=1288
we have not been able to find ways of enforcing that constraint within
the approach taken here, where encoding schemes are modelled as derived
complexTypes. If it is decided that this is a requirement, then the
schemas will require redesign.
On the extensibility of these schemas to support "application profiles",
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0209&L=dc-architecture&T=
0&F=&S=&P=5596
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0209&L=dc-architecture&T=
0&F=&S=&P=9862
our intention was to provide a balance between ease of use and
extensibility. It is possible to reuse the "base" schemas in in
association with schemas for "locally" defined containers, but we
acknowledge that they are unlikely to provide the flexibility for all of
the contexts where Qualified Dublin Core metadata is encoded in XML,
particularly in association with other vocabularies.
On the dependency on features of a specific schema language
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0210&L=dc-architecture&T=
0&O=A&P=4271
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0207&L=dc-architecture&T=
0&O=A&P=10068
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0207&L=dc-architecture&T=
0&O=A&P=4594
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0207&L=dc-architecture&T=
0&O=A&P=7661
we had not adopted this as a design/functional requirement, and we have
tried to exploit useful features of W3C XML Schema. However, we
recognise the arguments for the value of an approach which is less
dependent on a specific schema technology. If it is decided that this is
a requirement, then the schemas will require redesign.
On the question of whether the information for "dumb-down" should be
embedded within an instance document (can't find specific references,
but it has surfaced from time to time!), again we had not adopted this
as a design/functional requirement, and we took the position that it was
sufficient to obtain that information from an external source (e.g. a
schema). Again, if it is decided that this is a requirement, then the
schemas will require redesign.
The group who worked on this (the core participants were Timothy Cole
(UIUC), Thomas Habing (UIUC), Diane Hillmann (Cornell), Jane Hunter
(DSTC), Pete Johnston (UKOLN), Carl Lagoze (Cornell), and Andy Powell
(UKOLN), and we've recently had valuable contributions from Naomi Dushay
(Cornell)) feel that we have taken this effort as far as we can.
The one task I have still to do is to provide some notes on the
behaviour of some XML parsers with respect to some of the features used
in these schemas, but with the exception of that, we would like to pass
this to the DC Architecture WG to decide on its usefulness in terms of
meeting DCMI's requirements.
Cheers
Pete
-------
Pete Johnston
Interoperability Research Officer
UKOLN, University of Bath, Bath BA2 7AY, UK
tel: +44 (0)1225 383619 fax: +44 (0)1225 386838
mailto:[log in to unmask]
http://www.ukoln.ac.uk/ukoln/staff/p.johnston/
