According to JSTOR "130 Public Libraries in 32 countries participate in JSTOR." According to the American Library Association there are about 16,000 public libraries in the US. So the actual number of public libraries that provide access to JSTOR is very, very low. SOME people who live close to PUBLIC UNIVERSITIES may be able to access JSTOR, but most of the US public (much less the rest of the world) is SOL.

It isn't hard to get the Amazon data out of their database. What is needed, and what Amazon has in its ASIN, is an identifier for books that can be used link up copies of the same book in different libraries or locations. The ISBN only came into existence in 1968, so there is no ID for older books. That's why Amazon needed the ASIN.

Because it has such a large database, the OCLC record number has become a de facto identifier for books and other resources. The OCLC number is in every one of those (now restricted) records in thousands of library systems across the globe. But if we want to get free of OCLC, we obviously can't use their identifier.

The difficulty is getting an identifier into the millions and millions (or 1 sagan) of records in library databases. The options seem to be

1) develop a good, solid, computable identifier from the bibliographic data itself (nearly impossible)
2) create a switching system that will take bibliographic information as input and switch to a common identifier (like ASIN) (maybe more plausible?)