On 20.07.04, Paul Stainthorp wrote:
> > I'm trying to figure out how to identify a relation wherein a
> > document contains another document (as an attachment) in its
> > entirety. HasPart does not seem to really fit, as it indicates an
> > excerpt.
> >
> > In addition, I rarely have the attachment as a document in its own
> > right. It is often an item issued by another organization. However,
> > References does not truly apply either.
> (1) Requires "The described resource requires the referenced resource
> to support its function, delivery, or coherence of content".
I would think of this as a needed external source -- if it is included,
the referenced (original) document will no longer be necessary for
functionality, delivery, ...
> (2) hasPart "The described resource includes the referenced resource
> either physically or logically".
This seems to be the case for the above case. My basic example for
hasPart would be chapters of a book or papers in a collection a
conference proceedings volume. As an appendix can be thought of as a
special chapter, hasPart seems to be a good choice.
> (3) References "The described resource references, cites, or otherwise
> points to the referenced resource".
This could overlap with the "logical inclusion" in hasPart, so a link in
a webpage could be both. If the attached document is also
published/available separately, it could be given in both hasPart and
References (e.g. with just the name in hasPart and a fully qualified
citation in References)
> I'd be very interested in finding out how other people have coped with
> describing more "fuzzy" relationships.
I use DC qualifiers for a software library and there are some problems
with "fuzziness", as the terms are made with mainly text (in a natural
language) in mind. The problem is solved by establishing some
conventions for the use of the DC terms in the meta-meta documentation.
Günter Milde
--
G.Milde at web.de
|