[openbiblio-dev] bnb/Bibliographica namespace again
ww at styx.org
Mon Apr 18 13:41:16 UTC 2011
* [2011-04-18 12:35:08 +0200] Tatiana de la O <tatiana.delao at okfn.org> écrit:
] * We need to have our own data realm, so we can make relationships
] between all this read only datasets (i.e. this record in the BNL is
] about a translation of this author from the Wikipedia dataset)
] * Every external graph we add, be it
] http://bnb.bibliographica.org/entry/GBA0V5805 or
] http://dbpedia.org/page/War_and_Peace, should generate our own
] bibiliographica object, or link to an already existing object. This
] objects are the ones we should see in the bibliographica site and edit,
] dedup, tag.
] What do you think about this?
I agree. This means that we need to duplicate the current
infrastructure and make a read-only instance of openbiblio for the
British Library because there's no feasible way to create this kind of
partition with the virtuoso back end, and the database server is not
powerful enough to run two instances with this quantity of data
side-by-side. There is some expense involved in doing this so is
really a question for the project management.
In related developments, the work I've done recently with
http://semantic.ckan.net/ though in a completely different realm, may
prove useful from a user-interface standpoint. It is done all in
resources and is pretty neutral as far as the back-end is
concerned. Perhaps worth a look:
JSON representation of a record:
There are also some interesting proposals for enhancing SPARQL with a
CONSTRUCT JSON operation which could be incredibly useful and elegant,
but none of the stores support this yet.
Also related, semantic.ckan.net doesn't yet really model curated
groups of datasets yet, but I have a feeling that this should be done
with our familiar bibo:Collection...
William Waites <mailto:ww at styx.org>
http://river.styx.org/ww/ <sip:ww at styx.org>
F4B3 39BF E775 CF42 0BAB 3DF0 BE40 A6DF B06F FD45
More information about the openbiblio-dev