Whether or not piracy hurts book sales is hotly debated in some circles. For publishers, this is a no-brainer, but the evidence for this is not always clear. I have been sifting through the arguments about this for some time now and have concluded that at least in one area, college textbooks, piracy hurts sales. I am not yet persuaded that this is the case for other areas, trade books in particular, and would like to learn more about the circumstances of piracy for those segments.
So what’s this got to do with hamsters? “Hamster” is the English equivalent of the name of a file-sharing service operating in a certain non-English-speaking country. I am not going to provide any URLs here, as the last thing I want to do is to help pirate sites get more traffic. I recently spent some time in that country and spoke to several publishing people about the service. I was also provided with some hard sales data. Hamster is used for, among other things, the piracy of college textbooks. Among those other things is the sharing, whether authorized or not, of movies and music. Presumably some of the text-based files on the service are there legitimately as well. I am not a copyright expert, but I do know how to analyze sales figures. Hamster is a real problem.
So some numbers. With Hamster operating in this one market, college textbook publishers have seen sales drop by 50 percent. Two publishers have gone out of business; all the others are under strain. The number of new college textbooks is beginning to drop as publishers pull back from the marketplace. And that’s the really significant issue: investment in IP is declining because of the challenges to monetize college texts.
Hamster works as a simple file-sharing and Cloud storage service. For a small monthly fee, the equivalent of a dollar or two, subscribers get to upload files to the service, where they can be accessed and shared by others. The operators of the service do none of the file sharing themselves. For that reason, under the law in that country, the service is legal: a criminal case against Hamster failed in the courts. Presumably the individuals who share copyrighted material could be prosecuted, but chasing individuals with lawyers is not practical, as there are so many of them and the situation is really like (changing rodent metaphors) Whac-a-Mole: you get rid of one instance of infringement, and dozens of others spring up.
One publisher told me about a college text that sold about 3,000 copies. This is a translation of the English-language original, which is #1 in the field. (This publisher has a strategy to translate the #1 book for each discipline, making its list a congregation of Nobel Laureates and other superheroes of academic publishing.) The number of unauthorized copies found on Hamster for that title was several hundred thousand. Clearly, it is easy for these files to proliferate.
Now, it’s worth thinking about those numbers. That particular textbook is for a field where the number of students in that country is under 20,000 in any single year. Even including adjacent fields, the total market for that title is not more than 40,000 a year and is probably half that. So why are there hundreds of thousands of copies on Hamster? Because it is easy to do and why the heck not? Most of those pirated copies will never be read; their piracy cannot be said to take away sales. So we have a real problem with piracy (the market has dropped by half) and a mythical one (all pirated copies represent lost sales). Publishers make a mistake when they focus on that bigger number, as it just makes them look ridiculous. There is nothing to be gained from fighting criminality with stupidity.
Some publishers reading this post are probably saying, “This guy just now figured out that piracy hurts sales?” Actually, I have long thought that piracy is a problem in every segment, but I am impressed with some of the analyses that I have seen on this. There is the argument I cited above, that not all instances of piracy represent lost sales. There is the question of how to count the number of unauthorized copies. There is the problem of identifying and subtracting authorized copies (advocates of fair use, including many courts, make this point). There is the matter of not making materials available in easy-to-find-and-purchase digital editions, which prompts some people to turn to online file-sharing sites. (J.K. Rowling is, or was, the extreme illustration of this point. For years she prohibited the Harry Potter novels to be published as ebooks out of a fear of piracy. Her books went on to become the most pirated of all. Now they are all being formally and successfully published in electronic form.) And, finally, there is the promotional value of free copies, as their rapid circulation can evoke a broader conversation about a work. Publishers who have not already done so should immediately turn to Tim O’Reilly’s classic on piracy, “Piracy is Progressive Taxation.” O’Reilly, himself a publisher, has a sane view of all this, which may be a model for publishing policy just about everywhere.
But with textbooks, no, the offsetting arguments simply don’t carry enough weight. Part of this is the largely lawless character of college students everywhere, an international brigade of Mario Savios, who challenge authority wherever they stumble on it. An even bigger part is the fundamental economics of textbook publishing, where the high one-time cost of creating a work must be amortized over a tiny market. This forces up prices for texts far beyond those for trade books, and that gap tempts a would-be thief.
Publishers and publishing will not sit still while textbooks are shared without compensation. Litigation will continue; it will take new and ingenious forms and may actually stem the tide for a bit — Napster, after all, was sued out of its original incarnation. Publishers will also resort to DRM, a counterproductive practice, in my opinion, because DRM can be too easily broken; and once a single unfettered copy gets onto Hamster, anyone can access the work. Publishers will seek to make their texts more dynamic and hence resistant to simply copying. Publishers will also develop features of their work that invoke the participation of the instructor, making it such that an unauthorized copy denies a student access to the teacher’s contribution. One interesting venture I examined is a forthcoming Netflix for college texts, which provides a basket of titles for one monthly price and a handsome toolset to invite students to pay for what they would otherwise steal.
In the meantime, publishers will invest less, publish less. Until the problem of a revenue model is worked out, the textbooks available in this particular market will drop in number. It cannot be otherwise, as no company will make an investment without a promise of a return.
Discussion
22 Thoughts on "Do Hamsters Pirate Books?"
Joe, you’re right that DRM won’t slow down the pirates. I also agree that piracy is a huge problem in the textbook space; students saving what little money they have for beer are probably even more tempted to pirate textbooks than the typical consumer is tempted to steal a novel, for example.
What would happen if textbook publishers only made their content available via a streaming HTML format? No print and no self-contained ebook. So there’s no EPUB/mobi file with DRM to crack. The content is delivered to the student in chunks as they browse/search through the work. I’m sure someone could write an app to grab each chunk of the book and stitch it together to create something that could be shared illegally, of course. Would that be enough to deter or at least slow this piracy down or is that just one small bump in the road that pirates would quickly work around?
I think this is a suggestion worth exploring. For my part, I am most interested in solutions that make piracy undesirable or complicated without interfering with the user’s experience. I am most keen on dynamic texts, which, as they continue to change, are inherently hard to copy. But I don’t pretend that I have an answer to this problem.
It’s a good idea. I think that two things might complicate the adoption of a streaming HTML content delivery for college texts:
1) Device, power, and internet connectivity required. Not only would you be beholden to access to a functioning electronic device, with sufficient power, but also in range of a useful Internet signal anytime you wanted to refer to your text.
2) Annotation. How to highlight, dog-ear, underline, etc. important passages?
I think Mr. Esposito hit it on the head: it’s a breakdown of the economic model of producing and monetizing college texts. While I certainly agree that publishers have every right to monetize the IP that they developed, the current model of how to do this is no longer sustainable. Trying to sell a product for relatively high prices to a small demographic with a high need but limited purchasing power — it’s a guaranteed recipe for development of a black market. The current model only worked until now because college students had no other choice; they were a captive purchasing audience.
Subscriptions, rentals, display ads, institutional purchase/subscriptions (as a pass through to the student in form of fees or higher tuition) — these are all different economic models that could lead to more acceptable student text pricing while still providing for monetization of the IP. But the publishers are the ones that will need to make the changes.
I think I’m going to have a t-shirt made that says “There is nothing to be gained from fighting criminality with stupidity.”
While not condoning textbook piracy (and not trying to blame the victim), I would argue that the temptation is more than monetary. We have all seen new editions of expensive college texts which are “new” in only the most minor details, but which render older, used, editions immediately obsolete. I would suggest that some textbook piracy is driven not by a choice between beer or books or by some generalized anti-authoritarian impulse, but by genuine moral outrage over the perceived exploitation of a captive market.
Perhaps a semantic question, but if the textbook is new “only in the most minor details” then how does that render older used editions obsolete?
I do, of course realize that some publishers frequently update textbooks as a business strategy, but the more scrupulous publishers do this to improve the textbook and keep it current, rather than as a means of squashing the used market. What is a fair life cycle for a textbook? Is it okay to use a 4 or 5 year old text in a fast-moving field if lots of its contents are no longer correct?
My experience in college with an “updated” text book was that sections and chapters were slightly rearranged and the entire text was set in a different typeface to force the text to reflow, both things that didn’t change the scholarship in any appreciable way but seemed designed soley to make following professor instructions like “read chapter 3,” “read pp. 38-54,” etc., impossible w/o the “correct” edition of the book. And this book was intro to theatre history, hardly a fast-moving field.
Advanced courses in fast-moving fields shouldn’t be using textbooks anyway, at least not for the state of the art in the field. They should be using primary literature or recent review articles. The basic courses that use textbooks cover basic, foundational topics, which are much less prone to change.
Has there been so much advancement in, say, biology or calculus in the last ten years that a 10-year-old Biology 101 or Calculus 101 textbook is out of date? I doubt it. And yet biology and calculus textbooks have probably had three new editions in that time.
Yes, I agree that publishers who merely rearrange the deck chairs don’t actually make the Titantic voyage a better trip. But there are plenty of publishers with higher standards who don’t practice this sort of behavior.
To answer your question, yes, Biology 101 should be considered a fast moving field. I wrote an introductory textbook called “DNA Science” that came out in 2003 (http://bit.ly/UE4xaj makes an excellent holiday gift). That book was written just as the Human Genome sequence was being announced. Much has changed since then, and in many ways, the book was obsolete within a few months of publication. Can you teach a molecular biology beginning course without any information about what we’ve learned from the Human Genome? To their credit, the publisher (Cold Spring Harbor Laboratory Press) has never tried to rearrange a few pages to make an extra buck, but instead are tearing the whole thing down and replacing it with a new, modern textbook, “Genome Science”. I suspect that one will be obsolete within a few years as well.
Okay, maybe basic biology is moving faster than I thought. Having not taking a biology course since 10th grade, maybe it would have been wise of me not to use that as an example.
However, your giving an example of when a textbook does legitimately need to be updated does not argue against the prevalence of specious new editions of textbooks.
Maybe physics is a better example. My college intro to physics course used a textbook that had a new edition every couple of years, but it was entirely Newtonian physics. It wasn’t even up to date to the twentieth century let alone the twenty first. But in that field, that’s how it should be, because a firm understanding of Newtonian physics is required before moving to more advanced topics, like relativity.
I think it varies quite a bit, from field to field, and from publisher to publisher. But those doing the updating for no reason other than to rearrange the order of pages are likely poisoning the well for the more scrupulous publishers.
Well, it unfortunately seems that some of the biggest educational publishers are doing this. The theatre history book I mention above is published by one of the largest textbook publishing companies in the world. (I won’t mention any names, but they have a skyscraper with their name on it in New York City.)
Primary literature often isn’t synthetic (although review articles can be) and not pedagogical. I’d suggest that, at that level, often, monographs do a better job. And monographs with something of a pedagogical bent that include exercises probably work best.
Scott
(I work at Springer, opinions are my own.)
When a college bookstore carries only the newest edition of a textbook and professors insist upon that edition, the older editions are, effectively, obsolete. I doubt that calculus, for instance, is a “fast-moving field” that requires new texts every four years. I looked up some examples, but will refrain from shaming the publishers here.
Do students still buy their textbooks from the campus bookstore?
And not all new versions of textbooks are made solely to make a quick buck (see above). Some certainly are, but it’s up to the professor teaching the class to make that determination, to find the best textbook possible for the class, and to avoid books from less scrupulous publishers placing unnecessary burdens on students.
It is untenable for a professor to try to keep using an old edition of a textbook if the publisher no longer sells it because the professor likely may not want to bet on a large enough stock of used copies being easily available.
Hi Joe, don’t know if you saw this, but there was a recent meta-analysis of studies of the impact of piracy on media sales (short version: it harms sales):
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2132153
Though the decision is still under appeal, the judge in the Georgia State case suggested that if a written work wasn’t readily available in digital form from the publisher, then making a digitized version freely available was fair use. I wonder how that precedent, should it stand, would affect something like the standing of the J.K. Rowling books you mention in your post.
Flat World Knowledge deals with this problem by making its textbooks free for online viewing but charging for print versions. This puts a pirate like Hamster in an untenable position because it gains no advantage by posting the textbook online and is not set up to be a supplier of print editions, which it could not do anyway without directly infringing copyright. So long as that mixed OA/POD approach can be viable economically for the company, it may be the way to take care of this problem forever. (P.S. The same holds true for an OA monograph series like the one we published at Penn State Press in Romance Studies.)
Education is a public good. Ebooks have a zero marginal cost. From those two facts comes an obvious solution to the problem of pirated ebooks. Educational institutions should underwrite the costs of producing textbooks by paying those Nobel Laureates and translators. There is nomreason to charge students for textbooks. If the costs for textbooks need to be spread out, the textbooks can be licensed to other institutions. The way we pay for e-textbooks is suboptimal. That, not piracy, is the problemm.
To this comment I would remark that plurality should not be posited without necessity.
I am presently teaching several sections of a gen-ed, introductory geology lab course (for non-majors) at our college. This course has been taught for the past several semesters by adjuncts and another professor; this is my first time at it. I discovered that the previous teachers have had a difficult time finding a suitable lab manual for the course, in part because basically all such manuals are intended for geology majors and teach far more than any of us feel is necessary for non-majors, and partly because all the available manuals are incredibly expensive. The one we are using now–which none of us particularly like for various reasons–costs $118. And students will end up using only a fraction of it. Our solution? Write our own manual, one targeted to our non-major students, and one that we can print ourselves and make available (in print for at least the first edition; possibly as a PDF for future versions) to our students at a fraction of the cost. Our school’s biology department has long been doing something similar with their labs: the students in those classes pay a $40 lab fee as part of the registration for the class, and they receive the manual for free–the printing costs come from the collective lab fees for the course. We may do something similar for the geology labs, though they collectively have lower enrollment than the biology courses (as one might expect). I have been wondering to what degree such a solution might be preferable/useful for any and all colleges/universities…? Of course, at many institutions, there may well not be professors willing or able to devote the time necessary to create such a thing, and mass-produced texts are a quick & easy solution to the problem, though I strongly suspect that no matter what or where such texts are used, teachers end up modifying the contents or presentations in ways to reflect up-to-date information anyway. We have no illusions–or real interests–in making a profit from our manual…not that I’m under any illusion that should we try to funnel our manual through a big publisher in the future for mass consumption that we would make any real profit anyway. But this solution substantially eases the financial burden on the students, and they get something customized to their particular needs for this one course. Other than the time and effort it takes to produce something like this, I don’t see how this isn’t a win-win situation.
I am stunned that the notion of OERs (Open Educational Resources) did not even enter the blog post and was largely absent from the comments. How about questioning the necessity of university textbooks in the first place? Why do we need to worry about the business models of people who are benefiting from a captive audience. There are so many tools out there now that can facilitate an alternative http://techczech.net/2012/09/22/turning-wikipedia-into-textbooks-tools-ideals-and-accessibility/ so why should we worry about the profits of entities producing what is essentially of like the alternative of vinyl in the music industry.
