On Aug 10, 2005, at 9:15 AM, Matthew Cockerill wrote:
> This in fact ties back to my original question, when joining the list,
> about the best ontology(ies) to use for bibliographic data.
>
> Right now, BMC is planning to go the route of RSS+dublin core+PRISM.
I seem to have missed most of this thread for some reason. Given that
I've thought a lot about this question from a practical end-user
perspective (e.g. I am a publishing scholar; working at the border of
the social sciences and humanities, I'm frustrated with most solutions
because they're designed by and for hard scientists) and from now a
software developer's perspective (see citeproc*), here's my take:
The new FRBR schema is going to be really important in enabling smart
meta views of bibliographic data. It will enable answers to queries
like "give me all German language translations of Hamlet, but only
those available as eBook."
More grounded vocabularies like PRISM will work well when linked up
with a more FRBR-view, though some ontological glue (that says, for
example, that a Book is a published monograph, that it is a subclass of
an frbr:Manifestation, etc.) is going to be needed.
It's worth noting that Ingenta is moving to an RDF backend (with 200
million triples!), and they are using a mix of PRISM, DC and FOAF,
which is sensible to me (notwithstanding previous discussion about
limitations of name representation in all of them; authors really ought
to be ordered resource lists).
BTW, on Stefano's point about "it doesn't matter": I'm not exactly sure
what I think about the argument. However, I will say that I fear it
can easily be used as an excuse to create/use bad ontologies poorly.
Most bibliographic software sucks, and a lot of it is because its
designers never think deeply about the metadata.
Bruce
*
http://xbiblio.sourceforge.net/citeproc.html
Received on Wed Aug 10 2005 - 21:45:35 EDT