And now I can't stop thinking about the economics of the textbook market. There's an interesting NBER paper recently on the demand side of the market that finds students are quite rational and forward-looking when it comes to purchasing decisions. They'll make educated guesses about publishers' new editions and buy only those textbooks that are likely to hold much of their value. This certainly fits my own perceptions of the extremely economical students at Queens College. They even dealt in second-hand course readers!
An interesting issue in the textbook market --- on the supply side --- is the incentive for creating a quality product. I think these parameters are debatable, but one could argue that key qualities in a textbook are:
(The last is probably the most discipline-specific. At teaching colleges in economics, online homework grading such as provided for specific textbooks by Aplia is becoming an absolute necessity due to class sizes and the lack of graduate student course assistants.)
The problem is that writing or updating a textbook is very costly to an academic in terms of lost time. But by and large, the academic marketplace does not reward professors for writing textbooks in the form of career advancement or prestige; instead, they're rewarded for publishing scientific articles (or books). This is where textbook publishers come in. They're intermediaries that polish up the product, mark up its price, and compensate professors for writing and updating textbooks.
It is surely no coincidence that two of the professors committed to open-source textbooks in the Times article are already full professors at prestigious universities. Although that's arguably exactly who you want to write textbooks --- established members of the field --- my point is that they don't have to worry as much about the opportunity cost of writing and updating textbooks that bring scant returns in other senses.
Another issue laid clear in that article is the large difference between fast-moving sciences in technology and relatively slow-moving sciences like history and economics, where textbook obsolescence is less of a huge issue. In the case of the former, forces seem to favor open-source publishing, maybe because it can respond so much quicker. But maybe also because there's more money in the industry!