The British Library has made BNB data freely available for
downloading. Just over 3 m records covering 1950 - 2010 are here:
http://www.archive.org/details/BritishLibraryRdf
More about it:
http://www.bl.uk/bibliographic/datafree.html
The format is not MARC21 but a (newly invented?) XML schema.
We have tried to convert it into a database format we can
handle, and made a trial version available for searching:
http://www.allegro-c.de/db/a30/bl.htm
The interface is an Adobe Flash based "Rich Internet Application",
named "a30", already in use for a number of projects, among which our
database of RDA test records. There's more potential than becomes
apparent in this trial. One might say a30 is Ajax without JavaScript
and XML, but that's more or less what a RIA is all about: using the
HTTP protocol inside any browser but without much help from the
browser, not page-based but "asynchronously", and defining its
own higher level protocol. The backend is in PHP in this case, but
it might be any scripting language. The database is our allegro-C,
but it might be anything, Solr for instance. The front-end receives
a Unicode data stream and interprets it according to protocol.
Not much time has gone into this, so don't expect perfection. There
are, just for demonstration of connectivity, links to WorldCat and GBV.
We could add an e-mail function for results.
For the BL database, we have no information yet as to whether
it has open data URLs and if yes, what they look like. Does anyone
know?
Keyword search is by the input slot at the top left, click "Find" after
input.
Note the browsable indexes in the right hand panel; use the drop-down
box at the top to switch between indexes.
B.Eversberg
