[openbiblio-dev] future development and names of things

[Jim Pitman]

Jonathan Gray <jonathan.gray at okfn.org> wrote:

> For what its worth I'd really like to use the Bibliographica name and
> domain for an open source web service that enables you to make, share,
> import and export lists of publications, as per my original blog post
> on this at: http://jonathangray.org/2010/01/22/bibliographica/
> .....
> I really wanted to use this name [bibliographica] for a user facing service that
> enabled people to use bibliographic metadata to help them do useful
> things like maintain a list of references for a book, create a reading
> list for undergraduate students, maintain a comprehensive bibliography
> of works on topic X or author Y, etc. I'd happily spend some time
> being a demanding user for this, and would gladly give input on things
> like a simple, intuitive user interface.

I am very supportive of creating such a service which is superior to existing competitors in this space, and fine with bibliographica as a name for it. 
The data model for ownership of collections would have to be similar in principle to what underlies BibSonomy and Delicious, and allows these services to provide tagged collections
of bibitems and web links.  Another similar service is LibraryThing, but this is quite closed to machines, with no API for the biblio metadata.
Questions are

--  how bibiographica would distinguish itself from these services, (not to mention CiteUlike, Mendeley, Connotea)?
(I suggest by providing superior faceted features with BibServer. Already, we could do a big improvement on BibSonomy by providing a BibServer outer shell over BibSonomy data.)
-- whether it might not be profitable to join forces with one of them?
I might be able to negotiate that with delicious. And I think BibSonomy should be approached too. The others are unlikely to want to cooperate.
At the very least, easy import/export of data between bibliographica and such other services will be essential for rapid data acquisition, 
and demonstrating that bibliographica is a truly open biblio service.

I envision bibsoup as a cruder, lower level, more chaotic biblio environment, where large data providers can tip in open data of varying
quality, and various agents can be encouraged to perform some automated processing/indexing/stirring/mashing/...  to assist with deduplication and entity extraction.
The output of such processing, perhaps using tools like Google Refine, could provide input to a more civilized bibliographica service as envisioned above.

Perhaps this does not answer Mark's immediate question, but I think both names are worth cultivating, and worth developing along the above lines.

--Jim