14

I would like to know if a repository of all known ISBN exists; either a worldwide database, or per country.

If there is no such downloadable database, among all the available API's for querying books by ISBN codes (think Google Books, Amazon, etc.), what is objectively the most complete resource that would allow me to query books in English AND French languages ?

12

A quick Google search turns up several APIs.

You can limit your queries to these resources by knowing that French ISBNs start with a 2 and English ISBN's start with a 0 or 1.

Edit

As far as getting ISBNs for older books

Reprints of older books will be assigned ISBNs by the publisher. That is why books in the public domain (like most 'classics': Shakespeare, Chaucer, etc.) are issued by different publishers and have different ISBNs. - Source

I would assume this goes for any books that existed before ISBN was in wide use. ISBN became a standard in 1970 but in France, ISBN codes have been mandatory only since the 3rd December of 1981 (link is in French).

  • The problem I have with these is that more often than not for older French books (say pre 1980) I can't find anything... These API are very English oriented – Jalayn May 9 '13 at 1:44
  • 1
    Well, I think I found the best database/API in Worldcat (check out my answer), at least for the app I want to build. Search is fast, and results are excellent. – Jalayn May 9 '13 at 13:15
  • @Jalayn, +1 nice find. – John May 11 '13 at 17:39
  • 1
    @Jalayn : ISBNs have only existed since the mid/late 1960s ... it's possible that there wasn't strong French adoption 'til after 1980. – Joe May 14 '13 at 15:55
  • 1
    @Joe I found out when it has been adopted, but I didn't post the link here, so here it is (it's in french): legifrance.gouv.fr/… It basically says that an ISBN is mandatory starting from the 3rd December of 1981. – Jalayn May 14 '13 at 16:00
7

After doing some research and testing its search features today, I found the Worldcat database to be the best candidate for multi-language ISBN search. It has more in that it also has other media, such as DVD's and audio CD's, but for books it gets correct results every time, even for 1980 books not printed anymore. All media included, they say its database contains more than 2 billion items.

It has also books in many languages including English, French, German, Japanese, Chinese, Arabic, etc.

Worldcat seems to be owned by the OCLC, an organism I didn't know existed, and which is according to Wikipedia:

"a nonprofit, membership, computer library service and research organization dedicated to the public purposes of furthering access to the world’s information and reducing information costs".

It is also indicated that they (emphasis is mine):

produce and maintain WorldCat, the largest online public access catalog (OPAC) in the world.

... which by the way falls I think into the Open Data SE website.

4

Take a look at this tool https://github.com/xlcnd/isbntools. From the command line you can enter commands like this isbn_meta 978032153496 and get metadata about books from WorldCatalogue, Google Books, isbndb.com, OpenLibrary, ... and it is very easy to build more providers with a few lines of python!

You can access ISBN ranges (isbn_mask ISBN) and editions isbn_editions ISBN and with this info query more specific databases.

  • An anonymous user tried adding the following, which I felt was more appropriate as a comment : Now you can download a standalone version (doesn't need python installed) for Windows here. – Joe Apr 1 '14 at 16:08
  • isbntools features access to worldcat.org, Google books service and OpenLibrary.org without requiring their API keys. If you bother to get a API key from isbndb.com you can used this service additionally to the other ones. Also, being written in python, it's very easy to integrate in your own python project. – MadMike Apr 17 '16 at 20:05
3

You can query ISBN considering the Ranges.

There are many querible ISBN dbs, like the ones exposed by johnthexiii.

3

ISBN issuance is country-specific. Each country has their own registrar that issues ISBNs. It makes compiling such data virtually impossible.

Looking at registrars for US and France there is clear pattern that this data is not provided by registrars in bulk.

Since ISBN data is part of bibliography data collected by libraries and available via search/catalogs for free. That seems a way most of the ISBN databases are compiled. So it means that realistically there is a chance that such databases are not complete (but probably close to 99%)

For specific APIs see johnthexiii answer.

  • Interesting, are you suggesting I should dynamically build my database by aggressively querying multiple sources such as the ones named by johnthexii? – Jalayn May 9 '13 at 1:52
  • Not really, just commenting on the way how those APIs obtained their bulk databases of ISBNs. It's probably license/terms of use violation to build such database based on APIs. But if you go directly to library catalogs, it will be more legit (since libraries should have less restrictive terms of use) – Dmitry Kachaev May 9 '13 at 1:54
3

ISBN was established as a bibliographic resource not for libraries but for booksellers. And an ISBN is assigned to a particular edition of a book by its publisher. So it is not nearly the universal identifier for books that some think it is.

Of course, it depends on what you want to do -- library cataloguing information may serve your purposes better. If that's the case, I'd begin with the Library of Congress site, loc.gov. Its Z39.50 gateway (information here) allows for programmatic access to the electronic catalog and other electronic resources.

OCLC has asserted ownership of the cataloguing work performed by its member libraries; free access to WorldCat is limited to qualifying libraries. I'm not so sure how open its data truly is [but see my downgrade of this caveat below].

  • nice, did not know about loc.gov. And you are right about OCLC' catalog, the search API is free, but the API for retrieving content is for libraries only... I think I'll have to do some scrapping... Thanks for loc.gov though – Jalayn May 11 '13 at 22:32
  • 1
    Actually, from [this question[(opendata.stackexchange.com/questions/261/…) I learned that OCLC now recommends its member libraries release their data under ODC-BY (Open Data Commons Attribution License). But of individual libraries, probably LOC has the most data anyway. – Roger_S May 14 '13 at 0:12
  • WorldCat's services and data remain restricted to non-commercial use. They've backed off their attempt to assert ownership of their members' work, but that hasn't affected the union catalog. – Tom Morris May 24 '13 at 22:54
3

The British National Bibliography contains data on books published in the UK.

http://www.bl.uk/bibliographic/datafree.html

The data can be downloaded in various formats and is also available through online APIs.

-1

For a reasonable fee, you get this database of over 18 million unique ISBN as MySQL dump or CSV: https://www.usabledatabases.com/database/books-isbn-covers/

For disclosure, We are DataSN behind this database and the next update will eventually come out with over 25 million unique ISBN after integrating all the books APIs we publish here: https://datasn.io/p/221

Some very famous ISBN services and top books stores use this database.

  • This seems like a very strange thing to be advertising on OpenData when the database is not open and is for sale at enormous cost! – user02814 Mar 29 at 9:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.