Both monetary cost and freedom matter, independently of each other. Biella already pointed out why price matters.

Freedom matters because quoting isn’t the only thing one might want to do with a text: one might also want to make derivative works based on it. For example, translations! This is not a merely theoretical point: niche books and texts that are released under free licenses really *do* get translated by volunteers, see <a href="http://producingoss.com/translations.html">here</a> and <a href="http://www.questioncopyright.org/translations">here</a>, just to pick two off the top of my head.

I offer translation as the first example only because it is most familiar. There are many other things one might not think of right away, if one is used to regarding books as static objects. What if I like someone’s book, but don’t like their writing style? Why shouldn’t I just release a re-edited version of it? What if I see that it’s missing obvious references — why shouldn’t I just add them? If an argument is ill-made or incomplete, why should a reader not take matters into her own hands?

(Not referring to Chopra and Dexter's book specifically, of course.  I haven't read it; I'd like to, but probably won't because it costs $90.)

And just to be clear: this isn't about attribution, but about accurately-attributed copying and derivation. Of course accurate attribution should be required, by law if necessary, though it's worth noting that attribution problems has so far not been an issue in the free / open source world, and that's true for the documentation as well as the software. You can have complete freedom to modify others' works without that implying the right to steal credit.

David, when you say the freedoms to quote and critique are not infringed, that is a very impoverished notion of "freedom" I think. In the age of the Internet, of a world-wide copying and editing network, we can do much, much better.

Adam, when you say that a publisher offers "distribution, publicity, endorsement", what you're essentially saying is that it is appropriate for the *distribution* function to be unified with the *endorsement* function. Isn't that a bit odd? After all, newspapers prohibit, say, their music critics from receiving payments from the performers they review. Why do we consider it a good system, that the same people who recommend the books (that's what publicity and endorsement are) are the ones who stand to make more money if the books sell well. If anything, it would make more sense to argue for those to be separate functions. Marketing is an arms race: I understand how it might be to Chopra and Dexter's advantage to purchase some more ammunition (at the expense of their readers' freedoms), but the overall arms race dynamic is not conducive to objective evaluation nor creative reuse. What does a publisher offer the public that a word-of-mouth network or reviews site couldn't offer as well or better?

Peer review? Well, why not do peer review the way we do it in the free software world: *post*-publication.

The same freedoms that are important for software are important for all creative works. The "freedom to fork" is the ultimate peer review mechanism, and should apply to all bit patterns, not just the ones interpretable as running code by machines. Where we must regulate, we should do so only to protect truly scarce resources (i.e., attribution), not artificially scarce resources (the works themselves).

I'm not saying Chopra and Dexter did something morally objectionable here, by the way. I do worry that they may have fatally compromised their book: it will probably find fewer readers this way, and have a shorter active life (in part because readers are not free to keep it up-to-date, and Chopra and Dexter will not have complete freedom to do so either, since the publisher controls release dates now). At $90, I certainly wouldn't assign it in a class.

FWIW, I released a book about free software under a free license, and I know of at least three places where it's being used as a classroom text now. The fact that the instructors can just point the students to a URL may have had something to do with this. Also, the text is searchable; that might help. Also, quoting from it is easy, because the text is electronic and online as well as in print and in bookstores. Also, I can keep updating it incrementally, without worrying about when someone might decide to publish a particular snapshot (anyone is free to do so whenever they want). I have volunteer translators translating it into (at last count) eleven languages. I have seen people take the book and transform it into other non-book forms that I don't even quite understand myself (a "process description" website, in one case).

Do you see what a richer world this is than static content stuck between two covers, bound to a publishing schedule based on batching large quantities of dead tree pulp for expensive, centralized print runs?

I worry that lack of these dynamics will be harmful to Chopra's and Dexter's book, and therefore to them. Even if I could afford the book, it would start out with a strike against it from my point of view: the authors are writing (apparently with approval) about free software, but don't think highly, or perhaps broadly, enough of it to apply the same principles to their own work. So why should a reader take them seriously?

That's not a rhetorical question. I know why *I* should take them seriously: because they come recommended by Biella, whose judgment I trust. (There's an endorsing authority unconnected to distribution, by the way.) But for other readers, whose principle knowledge of the work is its own summary, the license terms should set off an alarm.