When I was into sailing, I bought meteorology textbooks, because I reckon "What the heck, I pay $100 for a weekend on the water. For an extra $60 I might as well be able to read the skies and the synoptic charts for added pleasure." I recently bought the standard textbook on equine hoofcare, because a visit by the farrier costs $200 a pop, happens every 6 weeks, and I reckon "What the heck. For an extra $100, I want to be able to tell the difference between a good farrier and a bad farrier."
I think college textbooks provide extremely good value for money. Working though one attentively, including doing exercises and checking the answers etc is pretty much as good value as actually doing the course at university (if it's a theoretical subject; obviously you can't learn surgery that way). This is also reflected in the way that they are now routinely being up-sold from mere books into online courses. The market for online courses hasn't quite equilibriated yet. But it's not uncommon for authors of textbooks to recycle the book into a slideset, play the slides on video at the same time as reading the book into the audio track. Throw in a meaningless certificate, and, bang, your $100 textbook is now a $1000 online course. People pay it because, often times, it's actually still worth it. (Although I prefer the bang-per-buck ratio of textbooks to online courses).
Examples? What you’re describing does not jibe with my experience doing MOOCs. Even the better ones cover a lot less ground than a textbook on the same topic, for the same reason a lecture course does.
For my Newtonian Physics class, the textbook we were using was on its 14th edition and cost more than $100 new. I was fortunate that my professor asked us to use the 13th at the latest, and gave the mappings to the 12th edition problems (because the only difference between 12th and 13th was which order the problems were in). This for a subject that's been essentially the same for centuries.
Meanwhile my Computer Science classes all had $40 textbooks that were in at most the 2nd edition.
Inclusive of: (1) altering problem sets or numbering, (2) reordering chapters, (3) organizing chapters.
Where if any of the above are done, a free mapping must be available on the internet.
I presume there are actual university guidelines prohibiting using older editions (i.e. ones the official college bookstore couldn't buy in bulk), but most professors just didn't seem to care (in between everything else they were juggling).
I'd be 110% in favor of mandating open source textbooks, where available, for any course taken by a student receiving government educational grants.
Essentially, you get to have (a) high textbook costs, or (b) monopolies over categories (e.g. one single mathematics textbook publisher).
Legacy / static topics (intro linear algebra) and evolving topics (intro biology) also have very different economic models.