My jaw is a little agape about the whole thing. online passes died in games after awhile, but it almost seems like the college textbook racket could probably get away with doing it forever.
The biggest difference between this application of it and the way it worked in games seems to be that it's designed to force your hand a little more. With games, you just got screwed out of the online features. With this, it sounds like you can't even complete the course without the online pass.
As an undergraduate at a small private college, I can only recall two professors who required their own books. One was a paperback physics reader for a very large class, available from the department for $20 (or from upperclassmen for less). The other was for a small math class, was not unreasonably expensive, but also was not very good, but when the professor left halfway through the semester with health problems, the department bought us all copies of the new book. My professors were always accommodating to anyone who wanted to buy an older edition, and in one case explicitly told us not to buy the expensive new edition.
In my current department at a large public university, I know of only one professor who requires his own book. It's about $40 and has only had one edition since it was published 15 years ago. In most lower division classes we use an in-house online homework system, although we don't have a good database of problems for differential equations, so we've typically required the latest edition of Boyce and DiPrima so students can use the publisher's homework system. But we've started work on our own problem sets, and we are at the point where we can use our own in-house homework system for it, if the professor is willing to put in a bit more effort organizing and debugging homework.
Anyway, this sort of exploitation is not typical in my experience.
Writing your own book is fine, but manipulating these books EA style (where you buy a game that comes with a one time use multiplayer key, neutering your ability to sell it afterwards) on the thinly veiled "improvements" is just downright shady. I commented about this on the other math book to pdf submission, but I had a professor did just this. He basically shuffled chapters around and renumbered problems, making the previous year's book worthless because the problems he assigned are numbered out of the new book, and he refused to provide any mapping of new problem sets to old problem sets. I had mentioned this to the department heads and none of them really cared at all about this practice.