Techie Class Texts

I had hoped to do a techie programming class for informaticians this fall. As it turns out, I probably will not. Instead, I’m going to do a broader “Internet Research” seminar at the grad level. Nonetheless, I hope I’ll get a chance to teach the techier class at some point; maybe next year.

I had settled on How to Think Like a Computer Scientist (Green Tea Press, $25). I’m currently reviewing Text Processing in Python (Addison Wesley, $35), after reading a positive review (along with some interesting comments) over at Slashdot.

This is really putting the cart before the horse, but assigning these books raises some interesting questions. Both are cheap when compared with either the average textbook or the average tech book. On the other hand, both are available free online. The latter has been made available in text files at the author’s own site, the former is under the GNU license and can also be found online.

The question of whether I should buy the hard or soft copy is one I already wrestled with when Cory Doctorow published his Down and Out in the Magic Kingdom. I read the book and enjoyed it, but I did not necessarily enjoy it $25 worth (or even $17 worth). I thought it was worth about a matinee movie these days: somewhere in the $5-$10 range. I wrote Cory to tell him this. In retrospect, such a letter (I like your book at about 1/5 of retail) was probably mildly insulting. On the other hand, the default–since I had already read the book, was not to pay at all, to be a free rider. His suggestion was that I buy the book and donate it to a library. It’s a good solution, but given what I already spend on books, and that I have been technically bankrupted by my educational experience, I simply don’t have that much to spend on books right now. Lest the reader think that this tight-fistedness is only because I’ve already read the book for free, I have held off on reading two books–After the Quake and Pattern Recognition–until I could get reasonable copies second hand.

I’ve offered to two people–one whom I know in RL and one whom I only know virtually–that I pay $5 of their purchase price if they buy the book. Cory probably sees less than a dollar for every book sold, but this provides some encouragement for Tor to continue to support authors, books, and publishing like this. (I have a feeling that Tor is already encouraged by sales of the book, but they might not be sure whether there is a direct connection to it being released online. Moreover, they might reasonably think that while this is a novelty at this stage it may not be a sustainable way of promoting books.) All this angst over a book I read on my PDA while working out.

When it comes time to assign readings for a class, this is further magnified. Last spring, I assigned a book for my media law class that cost $80. I know the author, but I don’t think that colored my decision; it’s a very good textbook. But it was not easy for me to assign such an expensive book to the students. As a grad student, I was the first one to seek out library copies (often across town) and other ways of keeping my book bill down. I hate creating new costs for students. It seems to me unbalanced that the textbook market is far more profitable than the fiction trade.

The solution in this case seems simple enough. I order the books through the bookstore, and I tell students that there is a hard copy available as well as readings online. I know in such cases that some proportion–usually a minority–of the class will buy the text and this will help authors get rich (i.e., be able to buy canned tuna). It’s just not a very satisfactory solution. What I want is simple: I want to be able to send Cory $5 for the book. I suppose I could just send him a random check, and let him tear it up if he wants to, but that doesn’t seem right either.

At least in the form I have now, neither of my fall classes have assigned books. In the case of the theory class, this means a set of readings that they will have to track down and photocopy or print (since creating a “reader,” once cleared for royalties, would mean about $200 per student). I know they would prefer a textbook, and I recommend a few, but I don’t assign any.

The best solution might be to have students write their own textbooks. In an undergraduate class, I assigned students to write textbook chapters, and about 3 of the 20 were better than those found in most com theory textbooks. One of these (on “muted group” theories) was somewhere out on the net at one point as a pdf, but I cannot find it now. I know I have a copy of it, and asked the students at the time for permission to post it, so I will keep an eye out. This fall, I will recommend that students upload their area papers to Wikipedia. That may push us a bit in the right direction.