Some time ago Tanel Põder (my technical reviewer) made the following comment in answer to the question of when he was going to write a book on Oracle internals:
“The answer is never, if talking about regular, old-fashioned, printed-on-paper books. I think the subject just changes too fast. Also, it takes at least a year of full-time work to come up with a book that would be any good, and by the time of publishing, many details would already be outdated.”
This is a good answer, and adds weight to my comments about avoiding the 1 percent and sticking to the general requirements and approximations. Tanel's response to the problem is his “living book” at
But paper is nice (even if it's electronic paper)—and ...