ames- I actually heard this from a colleague of mine while I was a teacher and I think he was very serious. Some of the authors list the names of the students who helped review the textbooks.
As far as the software is concerned- I agree that it’s very interesting.
Pretty often when I was creating an exam I started with an answer and went backwards to a question. (For example: X = -3 and 5. Now the question would be “Solve for X: X^2 – 2X = 15.)
The software would ask for ranges for the answers, say from -10 to 10, and design the question accordingly. An advanced option would create a different exam for each student (same questions but different numbers and the correct solutions were randomly placed from A to E for each student). Each test would have a bar code that had to be scanned before the answer sheet was submitted to the Scantron. Of course I couldn’t justify asking the Yeshiva to buy the necessary hardware to combat cheating when it was nonexistent.
I can’t begin to imagine how textbooks were written before PCs. Forget about coming up with problems, just think about the formatting and mathematical notations!