Thanks, Matt. The RDF here uses BIBO and DC, and is therefore
definitely lossy. I'm not saying that's a bad thing -- loss from MARC
may well be the only way to save library metadata. What I would be
interested in learning is how one decides WHAT to lose. I"m also
curious to know if any folks have started out with a minimum set of
elements from MARC and then later pulled in other dat elements that
were needed.
This brings up another point that I haven't fully grokked yet: the use
of MARC kept library data "consistent" across the many thousands of
libraries that had MARC-based systems. What happens if we move to RDF
without a standard? Can we rely on linking to provide interoperability
without that rigid consistency of data models?
kc
Quoting Matt Machell <[email protected]>:
Owen mentioned the Talis (now Capita Libraries) model. If you'd like
more info on that, our tech lead put his slides from the Linked Data
in Libraries event online at:
http://www.slideshare.net/philjohn/linked-library-data-in-the-wild-8593328
They cover some of the work we've done, approaches taken and some of
the challenges (in both released and as yet unreleased versions of the
model).
For some context, the Prism data model is used on some 70 or so
University and local authority catalogues in the UK and Ireland. Any
item in those catalogues can be accessed as linked data by appending
the appropriate file type (.nt, .rdf or .json) to the item uris (or
.rss to search uris), for example:
http://catalogue.library.manchester.ac.uk/items/3013197.rdf
Hope that's helpful.
Matt Machell
Senior Developer, Prism 3 - Capita LIbraries
Me: http://eclecticdreams.com
Work: http://blogs.talis.com/prism
--
Karen Coyle
[email protected] http://kcoyle.net
ph: 1-510-540-7596
m: 1-510-435-8234
skype: kcoylenet
