Production's Perfect Storm
Surely you have bought a book on occasion that had some binding error, such as having a chunk of a different book bound into it.
Today I got an email that indicated a version of this error taken to the max: The text of one book bound into the case and jacket of a different one. Here's how it happens:
For mass market paperbacks in the US, the binding is all in-line with the gathering. The signatures (sections) of the book are gathered up in order and the stack continues right on the conveyor belts to get a cover slapped on.
For most other books, though, this is not the case. The signatures are gathered, then the large stacks of them are collected and brought to the binding machine. Usually it's pretty easy to tell if the inside and the outside aren't for the same book, but sometimes you get the perfect storm, which happened on the book I was emailed about. The conditions were these:
1. Both books from the same publisher, binding in the same week, possibly the same day.
2. The authors names, which are the primary identifier of materials, were both four letters long, and had the first two letters the same. For example, MARX and MATH.
3. One book was 312 pages long, and one book was 272 pages long. The shorter book, however, was printing on a slightly thicker paper. Both books had the same bulk--they were the same thickness.
It's that third point that's really important, here. If the bulk had been different, the mistake would have been noticed when the first book hit the binding line. The cover would have been too loose or too tight, and if it didn't blow the equipment into an emergency halt, it would still have been noticed by the manual inspection a human does at the end. But because the bulk was the same, the book looked perfectly normal to the bindery operator. It fit snug in the case and the jacket wrapped properly.
The vexing thing, of course, is that one of the books involved was rather high-profile for us. I'm hoping that just a couple dozen were affected (a pretty good bet). But I can imagine that if any one of the above conditions had not occurred, the error would have been avoided, too.