> From [log in to unmask] Tue Jun 3 22:15 MET 2003
> X-RAL-MFrom: <[log in to unmask]>
> X-RAL-Connect: <sun8.loc.gov [140.147.249.48]>
> X-Authentication-Warning: sun8.loc.gov: rgue owned process doing -bs
> MIME-Version: 1.0
> Date: Tue, 3 Jun 2003 16:15:17 -0400
> From: "Rebecca S. Guenther" <[log in to unmask]>
> Subject: Re: Encoding scheme URIs
> To: [log in to unmask]
Rebecca wrote:
>
> I think this will get even more fuzzy in the future. There is an effort to
> get metadata scheme maintainers to provide URIs for metadata elements for
> the purpose of interoperability (some DC folks are involved in this). That
> can mean elements or values on a controlled list or the name of the list
> itself. I recently prepared a statement about what the Library of Congress
> might do for the standards we maintain, and if and when it gets done,
> there would be a URI for LCSH, since it is a value on one of our source
> lists. That URI would be maintained at LC, since the encoding scheme
> itself is maintained here, and thus would be different from that
> registered in DCMI terms. This is an issue that will need to be explored.
Dear Rebecca,
i agree with your analysis with the exception of "fuzzy".
The behaviour you describe is not fuzzy - it's perfectly natural.
The issue is to make relations - as those you describe for instance - explicit.
In my view it's a requirement for metadata architectures to be able
to cope with such situations.
PartOf the role of schema languages is to declare relations and not just
to enable the invention of new objects.
rs
>
> Rebecca
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> ^^ Rebecca S. Guenther ^^
> ^^ Senior Networking and Standards Specialist ^^
> ^^ Network Development and MARC Standards Office ^^
> ^^ 1st and Independence Ave. SE ^^
> ^^ Library of Congress ^^
> ^^ Washington, DC 20540-4402 ^^
> ^^ (202) 707-5092 (voice) (202) 707-0115 (FAX) ^^
> ^^ [log in to unmask] ^^
> ^^ ^^
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> >
|