This really drives home how scattershot organizing books by publisher is. Try searching for "Harry Potter and the Goblet of Fire" and clicking on each of the results in turn: they're nowhere near each other.
Or try "That Hideous Strength" by "C.S. Lewis" vs "Clive Stables Lewis", and suddenly you're arcing across a huge linear separation.
Still, given that that's what we use, this visualization is lovely. Imagine if you could open every book and read it…
Why would you expect otherwise? Titles are assigned ISBNs by publishers as they are being published. Books published simultaneously as a set might have sequential numbers, but otherwise not. Books separated by a year or more are not going to have related numbers. It's an inventory tracking mechanism, it has no other meaning.
Or try "That Hideous Strength" by "C.S. Lewis" vs "Clive Stables Lewis", and suddenly you're arcing across a huge linear separation.
Still, given that that's what we use, this visualization is lovely. Imagine if you could open every book and read it…