One ISBN, two completely different books?

August 8, 2017 8:48 PM

If you google ISBN 9781550545265, you'll find that it appears associated with two different books. What is going on here?

The google results page shows that both books are listed under that ISBN on different services:

1. Those Lake People Stories by Lynne Bowen
2. A Jeweller's Life by Toni Cavelti

This came up at work, when a customer called us to order book #2 (which appeared on our website), but our internal system showed #1 under that ISBN.

Interestingly, the Amazon produces a blended result (the title and info of #1 with the cover of #2).

As someone who has no idea how the ISBN-assigning process works, I'd like to know:
- How and where does a conflict like this originate? (eg. whose "fault" is it?)
- Is anyone (publisher, authors, etc.) ever interested in solving this type conflict? Is there a solution, after a book has been printed?
- Is there any way of knowing which book is "the real" bearer of this ISBN?

posted by ipsative to Technology (6 answers total)

ISBNs are issued worldwide by Bowker; they are supposed to be unique to each product and included in the barcode on the book itself. It looks to me like the error is in the metadata that the publisher sent out, since both Amazon listings are from Douglas & McIntyre in some form and the ISBN prefix (the part between the 978 and the last five digits, generally unique to a publisher, although a publisher may have more than one prefix) is consistent with other books in their catalog. I would bet that the error does not exist on the books themselves, although they may not have barcodes printed on them due to their publication date.

Neither the ISBN nor either title returns a result on the publisher's website, and both of these books appear to be out of print, so I don't think the publisher will be much interested in fixing the problem. You could flag it with Amazon, who also might not care since all the sellers are third-party. I think the best bet for figuring out the "real" owner of the ISBN would be to look for new title announcements in late 1995 through 1996 looking for reviews or other mentions of the books, to see what the ISBN will be listed as (although it will almost certainly be listed as the 10-digit version and not the 13; it wasn't until the mid-late 2000s that most publishers started using 13-digit consistently/exclusively in my experience).

I have worked in publishing for over a decade and I have never seen this before, so that was interesting!
posted by peanut_mcgillicuty at 9:12 PM on August 8, 2017

In my previous life as a book editor, I once came across a situation where an ISBN was assigned to one of our books, and metadata pushed to Amazon (for preorders, etc) before the book was subsequently canceled. The ISBN was then assigned to a different book, but by then the canceled book's information had spread elsewhere on the internet in association with that ISBN. It was a big mess, and never fully got sorted out.
posted by ocherdraco at 9:38 PM on August 8, 2017

ISBNs are issued worldwide by Bowker

According to https://www.isbn-international.org/agencies, they are the ISBN agency for US and Australia but not elsewhere.
posted by Gyan at 11:27 PM on August 8, 2017

I work in acquisitions for a major research library. This is so common. It's why we don't use ISBNs as a primary form of title identification. It can be from bad metadata, or from ISBNs being reused by accident, or due to cancellation of a previous title and it being reassigned to a new title, or by small ("kitchen-table") publishers ignorant of how ISBNs are supposed to work, who simply re-use the ISBN assigned to their very first published title.

The issue became worse when we transitioned from 10-digit ISBNS to 13-digit ISBNs, and became much worse with the advent of e-publishing, meaning most newly-published titles now have 2-3 ISBNs: hardback, paperback, e-book.

If we had to fix all the bad ISBNs in our catalog records, it's all we'd be doing. We have bigger issues to correct, so we don't touch ISBNs unless someone with clout raises a fuss. We do get librarians and patrons inquiring about erroneous ISBNs every so often, and we explain it: ISBNs are useful but not infallible information to help locate/identify a title.
posted by Lunaloon at 6:20 AM on August 9, 2017

Yeah, if I had a quid for every time we had a query about reused/shared ISBNs, I'd be able to behave like Scrooge mcduck and his pool.

We see it a lot because we have a third party bookjacket provider on our catalogue and people always report when cover A is on the record for book B. Most of the time it is a publisher reusing a cancelled or out of print isbn for another title - we will do some poking about and submit a case to our bookjacket system if we can find a legit source with the right book/cover image but generally there is no particular interest in the wider publishing world for fixing the issue.
posted by halcyonday at 9:01 AM on August 9, 2017

Most "legit" possible interpretation: The book contents are exactly the same; the title and pen-name changed. Bowker doesn't suggest a new ISBN if nothing in the actual book changed. Doesn't seem likely, though.

Other options - publisher cancelled the original book and re-used the ISBN; original went out of print and they decided to reuse the ISBN to save money (not allowed, but there's no ISBN police checking on this, esp in the mid-90's); someone typo'd an entry while submitting info, a small press flat-out stole the ISBN from a different book.

Publisher seems to be Douglas & McIntyre.
Toni Cavelti A Jeweller's Life hardcover at Amazon - ISBN 1550545280 | 978-1550545289.

Note: Cavelti is not the author. It seems to be a biography.

It doesn't help that Bowker's ISBN registry isn't public - you can't look up a book through them.
posted by ErisLordFreedom at 10:37 AM on August 9, 2017