If you want to sell a book that you have written, whether through a publisher or independently, then you will almost certainly need an ISBN. You’ve definitely seen one before, even if you didn’t know what you were looking at. The problem is, with digital books becoming more prevalent in the book market, you don’t just need one ISBN – you might need six. Why?
ISBN stands for International Standard Book Number, and it’s based on the creation of the Standard Book Number, by an Emeritus Professor of Statistics at Dublin’s Trinity College, George Foster. To put it simply (not really, but I’d like to quote Wikipedia here, so I will), it’s a 10-digit unique commercial book identifier – meaning it’s a code that is applied to your book that is used to collect data when a copy of your book is sold. Generally, if a book is published, an ISBN is scanned that allows booksellers and publishers to say ‘Yes, we sold 500,000 copies of that book this year, isn’t that great?’. It’s also helpful when book stores want to reorder copies and generally maintaining order in the face of literary chaos.
But you don’t just get one ISBN for a single manuscript – you also have to have ISBNs for every format in which you sell your book. So you can have a parent ISBN, a PDF ISBN, an ePub ISBN (the most common format for digital files), and a mobi ISBN (which is exclusive to Amazon, though it can be the same number as ePub files). And just when you thought it wasn’t going to get more complicated – here comes the DRM! Yes, it seems that it is now industry standard to take digital rights management into account – the International ISBN (International International?) Agency says that it should be industry standard to apply different ISBNs if significantly different settings are applied with the DRM, which might, for example, allow printing on one version but not the other. This is also the case if the DRM applied results in a specific version being tied to a particular platform. Come to think about it, you probably need two ISBNs if you are releasing both paperback and hardcover versions. Potentially, one book could amass dozens of ISBNs, if the author wants to really branch out.
It’s easy enough to ensure that all your ISBNs (and I’m sure most of you authors out there will have 18 for every book in your portfolio) can be neatly grouped together – there are identifiers available that match up textual contents of books, meaning that all versions of a book, no matter where they are located, can be found. Many sources about ISBN compare the system to fingerprints – every different version (or finger) is different and therefore has a unique fingerprint (digit). Still, it’s one heck of a confusing system to wade through, assuming you manage to get as far as locating and grouping together all your ISBNs.
This all seems… a bit insane. So many numbers for just one story? As I’m sure you’ve guessed by now, there is data to be gained from this that makes the whole endeavour worthwhile. All this segmented information can be used to determine information about what format is most popular with what author, what genre works best for PDFs, maybe even what type of grapes you like to eat while reading your ebook. Data mining is wonderful in that way.
Seriously though, while I can justify the need for separate ISBNs for reasons such as totally different formats, this whole system seems to be in place purely to gain extraneous information about readers and their purchasing habits for third-party database companies. Isn’t there a less numerically complex way of doing things? I already feel nervous for self-publishers trying to enter the field and discovering that they need approximately 480 unique identifying numbers that can only be registered through one organisation, if they want to sell their book or ebook to the public.
(If you still find the whole topic utterly confusing, I found this blog by Ron Pramschufer to be very helpful.)