Any ISBN yields a DOI. For example, the Chicago Manual of Style, 16th ed. has ISBN 978-0-226-10420-1, which maps to doi:10.978.226/104201, allowing any book to be digitally identified in a reference list by either ISBN or DOI.

Which should be preferred? Some considerations:

  1. DOIs aren't necessarily unique. Neither ISBNs, but duplicate ISBNs are rare.
  2. DOIs come with a nice URL scheme for obtaining authoritative publisher information on the text, using the dx.doi.org site. E.g., The Cambridge Shakespeare maps to http://dx.doi.org/10.1017/CBO9780511701207, although note that only the DOI containing "CBO" maps in this way, the one derived from the ISBN, http://dx.doi.org/10.978.0511/701207 doesn't.
  3. ISBNs are shorter and, for books, more ubiquitous.
  4. DOIs are very rapidly becoming dominant in scientific publishing.
  • 1
    DOI = this? – Goodbye Stack Exchange Dec 25 '10 at 4:40
  • RE duplicate ISBNs: An ISBN is supposed to uniquely identify an edition of a book. If a registrar or publisher gave the same ISBN to two different books, that's either a serious mistake or a deliberate breaking of the system. I'm sure it happens, but it's extremely rare. Unless someone can show that an alternative system is better at preventing mistakes, I think the fact that a tiny number of mistakes exist is largely irrelevant. (I don't have any idea what the relevant rules are on duplicate DOIs. I never heard of DOIs before reading this question.) – Jay Mar 7 '16 at 0:31

As you say, DOIs are becoming dominate in scientific publishing, so it would be useful to have one in that arena.

On the other hand ISBN (And it's cousin, the ISSN) are still the main way to identify a publication. There is also a lot of useful things that the ISBN number is used for. Right now you can put an ISBN number into pretty much any bookseller (or search engine) and it will get you to the right publication. Even my library lets me search by ISBN number.

The average book fan also knows what an ISBN number is and have some idea how to use them. They are useful and ubiquitous.

So, to answer your questions, unless you are doing a scientific paper or working in a field that has a strong reliance on DOIs, I would stick with ISBNs. And, as you say, you can make a DOI from the ISBN number, that provides better coverage if you just use the ISBN number.


In the publishing world, ISBNs are the standard. They're rarely reused (though it does happen on occasion) and it's how most booksellers and the like will be able to find a book quickly. I've never heard of someone using a DOI over an ISBN. Most people wouldn't even know what a DOI is.


The jury is out on that. So far the answer is no. Just what value does DOI add when you can already search by ISBN? What does it do besides rearrange the ISBN number and add DOI to it ?

DOI are useful for individual magazine articles but would cause needless confusion if applied to books as a replacement/duplicate of the ISBN. Might make more sense to use them to point to ebooks and other things withOUT an ISBN such as those with ASINs or even nothing to identify them.

ISBNs ARE unique.
The only time they get duplicated is when somebody (usually a naive or cheapskate self publisher) reuses their old ISBN on a different book.

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.