Hacker News Reader

Comment by PaulHoule | Hacker News Reader

I sampled books from my personal collection and found many didn't have an ISBN. Officially ISBN started in 1970, and I just found one on a book that was printed in 1972, I'm not sure what the adoption curve was like.

ISBN is also not guaranteed to be a primary key. It's designed to serve the needs of new book sellers. If a book goes out of print the publisher can reissue the same ISBN to a new book. It's unusual, but South End Press notably was resentful about paying for new blocks of ISBN numbers and recycled ISBNs to "stick it to the man".

Some books have an ISBN barcode on them

https://en.wikipedia.org/wiki/International_Standard_Book_Nu...

but you don't want to waste time with a "cheap" barcode reader made in China and sold on EBay. I have played around with those and find that they read barcodes when they feel like reading barcodes and it is quicker to type the codes in.

View on HN

There's a certain rare, out-of-print book that took me a couple of years to track down. It shares an ISBN with another common, inexpensive, in-print book whose only similarity to the rare one is that it came from the same speciality publisher. Every time I found a copy of the rare book on various used book websites, the common one showed up on my porch instead. Apparently there are a lot of people who treat ISBN as a primary key. A non-trivial number of sellers thought I was trying to pull a return scam on them because they could not be convinced that ISBNs are not unique. The fact that the book I tried to buy generally listed for around $300 and the book I was actually sent went for $15 used really made returns difficult.

My search finally ended when a copy showed up on eBay, complete with listing photos. I suspect the price I paid was much higher than usual for a book with similar desirability and rarity because the ISBN collision made it much harder to find.

Dumb question: If the book was harder for buyers to find, wouldn't that drive the price down, because not everyone who wanted to bid on it would be able to do so?

It was hard for buyers to find on used book marketplaces because sellers would scan a $15 book, their system would pull metadata for a $300 book because of the ISBN collision, and then you don't know which items are actually for sale until you order them and they arrive in the mail. It's worse when dropshippers get into the picture and there's effectively multiple listings per item.

On the other hand, eBay listings for books are usually made for a specific item, complete with pictures of the actual item. It was obvious that the listing was for the real thing.

Do you have a source for the non-uniqueness of ISBN? Google didn't provide anything tangible and I'd be very interested in a reputable place confirming your claim.

Actually, as far as I see [0] ISBN have to be unique. It seems like this "South End Press" is more of a case of a publisher going rogue than anything els.

[0] https://www.isbn-international.org/content/isbn-assignment (see the '“Out of print” editions' bullet)

The ISBN might be unique, but that doesn't mean the barcode on the cover is.[0]

[0]https://www.nytimes.com/2021/07/29/style/dan-brown-advice-bo...

Stephanie Marlowe's reply to

https://www.quora.com/Is-every-book-has-a-unique-isbn-If-yes...

points out the non-uniqueness of ISBNs. I used to work in the IT department of a major university library and it was a problem that the librarians were aware of.

It takes exactly one "rouge" publisher to make a problem. (Funny I used to know the people at South End Press)

rogue or communist publisher?

Fair enough, but my issue with your original comment is that it implied that _by design_, ISBN does not uniquely identify books, which is wrong, as far as I can tell. In any case, the issue of reuse of ISBN has practical consequences, so that the original intent might not be so important in the end.

Do you have any idea how many cases there are of reuse? Is it actually common?

Reply on news.ycombinator.com