I am wondering if the following situation would be handled best by using
DCMI-DCSV (Dublin Core Structured Value) as recommended in Simon Cox's and
Renato Iannella's document [1].
I am working on a project that involves creating Dublin Core metadata
records for web-accessible library pathfinders. We are using DC Publisher to
record the name of the library where the pathfinder was created. However, we
want to record some additional information to identify the library where the
pathfinder was created. We have decided to do this by adding the following
"value components" to DC.Publisher:(1) name, (2) url, (3) location of the
library (for geographic reference to the library's location),and (4)
category (to classify the library type as academic, public, or special).
FIRST, Am I correct in viewing these as value components, as opposed to
locally developed qualifiers? (In my view, they do not dumb down as
qualifiers should, except for " name"). Furthermore, this sounds very much
like an "agent" issue.
SECOND, Is using the DCSV proposed "syntax" a good way to handle this
information, while remaining interoperable with qualified Dublin Core
structure? (These metadata records will sit in a database alongside Dublin
Core records for other resources that do include these Publisher value
components).
FINALLY, can anyone comment on the best source to consult for the current
consensus on how to express Qualified DC in XML? For example, would it be
the draft proposal by Miller, Miller and Brickley, "Guidance on expressing
DC within the RDF [2]?
[1] http://dublincore.org/documents/2000/07/28/dcmi-dcsv/
[2] http://www.ukoln.ac.uk/metadata/resources/dc/datamodel/WD-dc-rdf/
Thank you in advance for your comments.
Carolyn Guinchard
Metadata Librarian, The Alberta Library
Mail: 4-30 Cameron Library
University of Alberta
Edmonton Alberta T6G 2J8
Phone: 780-492-4645
Fax: 780-492-9243
Email: [log in to unmask]
www.thealbertalibrary.ab.ca
|