The timing on this for me is really interesting, as last week I got an ISBN issues for a book I'm working on (9781633438002 if anyone is curious!).
This will be the first book I'm the author of, but the second book I've worked on (the first I was the technical editor for). Neither of these books are out yet (I start writing tomorrow) but they both have ISBNs issued. Even if I never publish the book that ISBN is locked in.
I imagine there's a lot of books that started out but never got finished. That said it looks like ISBNdb doesn't grab directly from the source, but instead crawls the internet looking for ISBN data to put into its database. I'll be interested to see at which stage my ISBN shows up in the database.
It's a good unique key to use for tracking the book, even internally. You might change the title of the book at a late stage. You could use your own ID scheme, but what if your publisher merges with another while the book production is in process?
Before writing might be a bit early, but before finishing and producing is useful as it can be listed early in catalogs for preorder. Getting the ISBN earlier probably is cheap and allows the publisher to use the ISBN as identifier for the whole project.
I have no idea why they assigned it this early, but that seems as likely a reason I can think of.
It was only assigned the day we finalized the contract, and there was a lot of work before that working the proposal through the system and getting reviews from the target audience and people familiar with the topic. It's only now that I'm expecting to hand content over on a schedule that they assigned the number.
This will be the first book I'm the author of, but the second book I've worked on (the first I was the technical editor for). Neither of these books are out yet (I start writing tomorrow) but they both have ISBNs issued. Even if I never publish the book that ISBN is locked in.
I imagine there's a lot of books that started out but never got finished. That said it looks like ISBNdb doesn't grab directly from the source, but instead crawls the internet looking for ISBN data to put into its database. I'll be interested to see at which stage my ISBN shows up in the database.