Month: July 2008

Gene Spafford (right) is no Luddite (though he worries, below, he might sound like one with this rant he mailed in to Dave Farber's list) but he is a smart guy and a digital pioneer who worries about the implications of generations of students hooked up to Bittorrent never having to pay for a textbook.:

As noted, the whole mechanism of textbooks (and books in general) is changing.

It used to be that authors toiled over texts to present distilled information about their expertise — often hard-won, and usually with careful research. The resulting books were valuable for self-study, for teaching, and especially for reference. Books with valuable content and organization were treasured not only for classes, but for later reference. Years later it is possible to go to a particular book to find an algorithm, scientific constant, or quote that is needed.

Over time, we have seen trends that have eroded the model — and more quickly than most have realized.

First, computerized type-setting and faster printing allowed a lower barrier to entry for printing. Small publishers could get a profit from a smaller press run, and enticed many new authors to write books on topics where they perceived demand. Some also turned to cheaper materials — high-acid, low density paper, cheap glue binding, paper covers — that result in books that wear faster and don't hold up to repeated use or storage as a reference. I look at my reference library (about 800 books) and see many recent publications of very limited utility and likely short life-span. Nonetheless, we see this flood continue because there is a profit to be made and few of the audience read enough of the books to distinguish good from bad, so there continues to be a market. Plus, books with errors or are incomplete aren't a big deal anymore — put the errata on line, or wait for the next edition (sound like the software problem?). With Google, Yahoo and Wikipedia, many people don't feel the need for physical references on their shelves any more.

(Aside — our new buildings with (small) faculty offices are being constructed with limited bookshelf space. Faculty are told to either take books home or donate them to the library. The image of a learned professor surrounded by books is also becoming passé.)

As a (former) author, the question is why would I write a book in this environment? Well, it certainly isn't for the money. As Mary Shaw noted, there isn't a lot of return. I co-wrote a couple, plus many book chapters, and although they sold well, I can't say I made a lot of money. It certainly didn't cover the time away from family, and the permanent damage to my hands (which has limited my ability to write much of anything over the last decade). Many current academic colleagues — particularly the ones who don't write books — don't judge them as too significant. Furthermore, using some of the poorer books out there as metrics, they don't value the scholarly effort some of us put into our writing, either.

The textbook publishers are in business to make money. So, the ones producing the better textbooks need more incentive to offer authors, plus a bigger profit margin to cover fixed expenses with sales of fewer books. Not all their books are hits, either, so they have this balance between bringing out new titles and sustaining the long-term balance. The result is that costs creep up, even if they are trying to contain them (and I doubt they are as rapacious as the media publishers).

So, as an instructor, what do I do? I can certainly assign essays and work off the WWW, but how do I find the best ones in an area where I may not be a top expert? The input of an editor and/or co-ordinating author who expertise I can judge would be a help, but I don't get that from Wikipedia or Google.

Do I want to be teaching fundamental principles from “The Big Dummies 1-2-3 Guide to C++”, 19th edition, knowing that my students are going on to program critical infrastructure and national defense applications? I would rather include sound pedagogy, reinforcement of material on critical algorithms and data structures, issues of ethics and law, and more that is in some of the more carefully- designed textbooks.

But then I have students who balk at the $100 differential and they don't get it when I explain they are paying for quality: they're focused on getting through school as quickly and cheaply as possible to get a job. Unfortunately, many of them carry that over as a work ethic — do it is as quickly and cheaply as possible to get it out the door. 🙁

If we are trying to advance any scholarly field, we should all be working from common terminology and well-documented experiments and facts. How can we trust something we find online that has no author or reviewers listed, or else they are pseudonyms, or people we have never heard of? Is that a stable foundation on which to build future science with confidence?

And what happens 20 years from now when researchers try to go back to underlying principles and results, and cannot find canonical versions of texts to verify that they have been cited appropriately because there are dozens of versions stored electronically….and which may differ in both subtle and significant ways?

The same problems have been happening with journals and conference proceedings. People don't understand that the money they pay goes towards making a fixed archival copy, and to help ensure that there is some quality control in what is published.

I'm sure I sound like a crusty old Luddite to a few people reading this. I know all the arguments about the cyber revolution making knowledge quickly available, at how we can avoid cabals and politics by publishing new results quickly, about how scarce funds can be spent on items other than books, and how even 3rd world scholars can have instant access. I've also heard the arguments about “many eyes” fixing problems in publications and code, and it has been proved specious, and is part of the reason we have a “we'll fix it in the next release” attitude. I'm certainly both a vendor and a customer in the vast marketplace of ideas enabled by all our innovation.

Yet, as a scholar and educator, i worry how to ensure that all our students get the best, most correct materials, that our researchers use correct and commonly-available results, and that we document our progress in correct and archival formats for generations to come. I don't see a cost-effective, workable model yet. What I do see is the same problem I see in many other enterprises, and especially in software systems development — the whole rush to cheap and fast because people don't understand the lasting impact of quality.

I think those problems are part of the whole discussion, and textbook costs are only part of the issue.

Technorati Tags: copyright