Uh-Oh! Enforced Open Textbooks

September 30, 2009 at 9:37 pm 24 comments

I find this new bill pretty scary:Durbin Open Textbook Bill Finally Introduced!.  I don’t have anything against Open Textbooks.  The ones I’ve seen so-far haven’t been very innovative, and I do wonder if it’s a viable model. People deserve to be paid for their effort, and higher quality effort should result in greater pay. That’s the economic force that encourages better materials. Part of what this bill does is fund development of Open Textbooks, and I think it’s a great thing to develop that model.

The scary part of this bill is the enforcement that everything (“curricula and textbooks”) funded by the National Science Foundation must be open source:

In General- Notwithstanding any other provision of law, educational materials such as curricula and textbooks created through grants distributed by Federal agencies, including the National Science Foundation, for use in elementary, secondary, or postsecondary courses shall be licensed under an open license.

I suspect that most science education innovations in the United States get some part of their funding from NSF.  I am quite confident that most of those innovations in computing education are. Alice, Media Computation, DrScheme — all of them had NSF funding. Under this bill, none of those could be published commercially.

By selling books through a publisher, we (authors, but also the teaching community) get a dissemination mechanism and a support mechanism. I’ve talked to teachers (including university faculty) who won’t use a book unless it’s offered by a publisher — they want to know that it won’t go out of print while they still need it, and that it will be updated as long as there are customers.

As an author, I’m glad that Pearson has a sales staff that goes out and tells people about my books. I want those people to get paid. I don’t know if Pearson can still make money if a percentage of their market decides to simply deal with the PDF and Pearson can only sell to those who don’t trust the author’s website or who want hardcopy.

Under this bill, a commercial publisher can’t put good, well-assessed curricula in a textbook and then disseminate it via a salesperson and bookstore distribution network. How does one get high-quality, sustainable change if you can’t ever pay authors, salespeople, bookstores, and publishers?

Entry filed under: Uncategorized. Tags: , .

Operation Reboot: Georgia Tech Looks To Shift Unemployed IT Pros into Teaching Careers — THE Journal Playing Wack-a-Mole with State CS Ed Policy

24 Comments Add your own

  • 1. linda hahner  |  September 30, 2009 at 9:56 pm

    We believe agencies pay money to people who do work for projects so no one is working for free. We see nothing wrong with building something for students, as part of an agency grant, and then writing textbooks about it. This seems fair. These would be supplemental materials and textbooks just as they are now.

    However, many children in the United States don’t learn because they don’t have equal access to materials. Our organization strongly believes if the government is going to fund projects then some of these projects should yield material that can be used in the classroom. This should become part of a free NSF open source repository.

    This is not to say that the material should be designed by researchers. Children should only be presented with material that has been designed by those schooled in design and usability. To achieve optimal results, the NSF should fund teams of experts.

  • 2. Andrew Miller  |  September 30, 2009 at 10:09 pm

    Interesting! Do you think there’s room for a nonprofit model here? Might be a happy medium.

  • 3. Owen Astrachan  |  September 30, 2009 at 10:24 pm

    It seem as though you may be confusing the licensing requirements with commercialization. This is the same confusion that many opponents of open source software make. Check out htdp.org, where you’ll find the complete text of a book that’s available commercially. Check out this link http://www.cs.duke.edu/courses/cps082/fall09/books/ where all the books I use in a course are available “free” online, as well as commercially. You may not be able to alter the book, I haven’t checked on the licenses. But in this case I wanted books that were “free as in beer”, and the “free as in speech” aspect of the licensing wasn’t so important [see Stallman on free software]. Companies make money selling free software, it should be able to happen with books too.

    The price of textbooks is completely out of control. Your book is great, but it’s not $80+ great. Of course others will differ on this.

    Moreover, I doubt this bill will ever pass, so scaring up opposition with a “oh no, what about publishers” is just like Microsoft decrying open source software as either bad business or bad process. Any efforts toward lowering the price of textbooks should be lauded,not decried.

    In your arguments you take both sides “many professors won’t buy a book unless it’s supported by a publisher — they want to know it won’t go out of print”. Going out of print is something the publisher decides when the book isn’t commercially viable. Any open book would presumably *always* be in print.

    You can make money selling open textbooks or open source software. I would embrace any decision that textbooks developed under federal funding would be available without the markup the publishers effectively mandate. But this bill won’t pass, and statements such as “none of them could be published commercially” don’t help. I’m not saying publishers would embrace the open/commercial hybrid, but MIT press does. So it could happen.

    Textbooks, make them cheaper, please.

    • 4. Mark Miller  |  October 1, 2009 at 8:17 am

      One of the examples of an “open book” is “Squeak by Example”. They distribute the PDF for free (and take donations), and the publisher charges for hardcopy editions.

      What I say below is about the subject of open textbooks generally, not just for scientific information or CS.

      I think where open textbooks could be a boon is in trying to make them more accurate. I’ve been hearing for years about textbooks that have mistakes in them or just plain wrong information–and these are from established publishers. It would be nice if there was an open mechanism by which information in textbooks could be challenged and corrected.

      Even when I went to college there were errors in textbooks. My professors knew all about them. My 2nd semester Calc. professor told us that the selected answers in the back of our textbook were created by graduate students, and the quality of the workmanship in those answers got worse the more sophisticated the material got. By third semester Calc. the answers in the back were practically worthless.

      In every CS textbook I got where the subject of programming was discussed, with source code examples, you could count on most of the programming examples containing spurious errors. The reason being, as it was explained to us, was that the authors just plopped the code into the text and didn’t bother testing it first.

      I saw a news feature about a month ago on textbooks in K-12 and the idea of open textbooks was profiled. One of the people trying to encourage it said that they saw it as a way for teachers to customize the material that would be presented to their students. Teachers could update the editions they got with their own information. That sounded interesting, and at the same time made me a bit wary. What if what they put into their edition was not accurate information? That doesn’t sound any better than the current system where inaccurate information can crop up in books from established publishers. Would students be given the impression that the open textbooks were somehow more trustworthy? This impression is often given of open source software, whether it’s true of a particular piece of software or not.

      The profit angle is a concern, and I think it was discussed on the show I watched. Open textbooks seems to be a consumer-driven idea of trying to improve the content that’s presented in them, since teachers have already experienced the lackluster quality of textbooks in the existing system. It seems to be a way to get around a systemic problem that lacks competition. A lot of schools standardize on what other school districts have standardized on, just following the leader. Content is not always considered. It’s just the convention of the system. If there was more discretion in school systems then open textbooks might not sound as attractive to some.

    • 5. Mark Guzdial  |  October 1, 2009 at 9:22 am

      Textbooks should absolutely be made cheaper. The current system is clearly broken. Mandating that any education development funded by NSF must be given away for free goes too far, IMNSHO.

      People who put together catalogs and salespeople who sit down with teachers to help them figure out what books meet their needs are doing a service. Can they get paid enough to make a living if they are just doing that for free books? It’s an open question. Should Congress force the issue and regulate that all science education books should be made free? I don’t think so.

  • 6. Tom Hoffman  |  September 30, 2009 at 11:26 pm

    A few points:

    * Open source licensing can’t be that scary, because two of the three examples you cite, Alice and Dr. Scheme, are open source.

    * Open source licensing does not, by definition, disallow commercial redistribution.

    * A substantial chunk of educational software written with public funding, such as Scratch, ends up in limbo where commercial distribution is prohibited (on the decision of the author or university), and there is never any income from commercialization anyhow. The only thing that is accomplished is the limiting of redistribution channels.

    * Because there are companies that make their living redistributing open source software, Red Hat, Canonical, etc.

    * If Pearson wants to pay people to create content they can license it however they want. If the people of the United States are going to pay for it, they can do the same. If you don’t like the terms, look elsewhere.

    * If this all means the government is paying researchers more up front and also funding projects to help distribute open resources, it is still a huge potential efficiency overall.

    * This is even more important for K-12 resources, where the current distribution methods for innovative technology research have failed utterly, in every facet.

    • 7. Mark Guzdial  |  October 1, 2009 at 9:25 am

      Tom, Alice and DrScheme *software* is open source. The Alice book is published by Pearson. The HTDP book is published by MIT Press, with a free version, too, which is presumably an open-enough license to meet the requirements of this bill. The bill’s language also says “curricula.” That says to me that the Alice book has to be released under open license. How about everyone who has written Alice books? Aren’t those using Alice “curricula” (where that might include 3-D models/characters, examples, on-screen depictions)? Don’t those now have to be released under open license, too?

      • 8. Tom Hoffman  |  October 1, 2009 at 9:49 am

        Also, regarding third party texts about open source software — of course you can release proprietary documentation — heck, you wrote a commercial book about Python, and it is open source software.

        There is no precedent for licensing software in such a way that limits the licensing of documentation.

      • 9. Mark Miller  |  October 2, 2009 at 3:06 am

        To add to Tom’s point, another anecdote might help. Not everyone prefers free stuff. I used to hear from software developers who worked for defense contractors. They worked with inferior tools which were supported by the manufacturer. One was complaining about the C++ compiler he had to use, saying it was buggy. I asked if he had tried GCC. He said his boss would not allow it. I was a bit shocked and asked why. He said his boss only wanted to get software supported by a vendor. They wanted to pay for what they used, because they saw a dynamic where the customer pays for support and gets it.

        Sun Microsystems used to distribute an office suite called Star Office (which was their version of Open Office, an open source office suite) for free. They didn’t get many takers. They did some market research and found that corporate customers were uncomfortable adopting it because they figured it would get no vendor support. So they started selling it for $50 and got a lot more customers.

        So it seems from experience that in order for an “open” product to gain acceptance it needs a commercial component. It can be available for free, but it needs to have other options available where it can be sold under an agreement, implied or explicit, saying that the product will be guaranteed continued support. And of course it should be backed up with actual updates.

        The reality of this is something that I think some people in the OSS community find pointless because they see OSS as being supported by the community, where updates happen in projects with wide community support, and without the prompting of commercial support. The commercial aspect has been more of a confidence boost for OSS into corporate environments. So because of this, it is possible to make money on OSS. Likewise my guess is there will be schools who will want to buy an edition of an “open” book with the idea that you will support the book you publish. Perhaps there could be a website associated with the book that’s only available to institutions that buy your “open” editions.

        With other “open” products people can make additions to it or “forks” it of their own volition, if the license terms allow it. In that case the money flows to whoever best supports the original work in the view of the customers.

        I can imagine a different organization or company coming out with supplemental materials that support the original work.

        As with OSS the financial model is that of the “loss leader”. You forfeit profits for doing the “heavy lifting” and hope for residual income. I realize it looks like it’s taking the wind out of the sails of publishers. I’m not the best advocate for it because I’ve never seen it in action up close and personal. I’ve only heard about it from others who have tried it.

        Philip Greenspun talked about the dynamics and pitfalls of running a commercial OSS company in a podcast of an interview he did several years ago. Unfortunately I’ve lost track of it. It’s probably still online.

  • 10. David Wiley  |  October 1, 2009 at 12:30 am

    Your claim that “Under this bill, a commercial publisher can’t put good, well-assessed curricula in a textbook and then disseminate it via a salesperson and bookstore distribution network” is an understandable but naive reaction, and is completely unsubstantiated. The impossibility you describe is exactly the business model of companies like Flat World Knowledge.

    “How does one get high-quality, sustainable change if you can’t ever pay authors, salespeople, bookstores, and publishers?” Read about Flat World Knowledge’s business model to see how they do it. http://flatworldknowledge.com/

    • 11. Mark Guzdial  |  October 1, 2009 at 9:27 am

      Does Flat World Knowledge work? Is it solvent? Are its investors making a profit? I’m not opposed at all to profit — it’s what enables greater sustainability.

  • 12. Alex Rudnick  |  October 1, 2009 at 1:23 am

    I hadn’t considered the case of textbooks yet, but there’s been a fair amount of noise about open-access publishing for research. Do you buy the argument that if the taxpayers fund the work, then everybody should have access to the resulting papers?

    Adding to the list of open-access books: Natural Language Processing with Python and Real World Haskell (both from O’Reilly) are fantastic.

  • 13. Tom Hoffman  |  October 1, 2009 at 9:42 am

    Is Pearson’s business model dependent on government subsidy? Other publishers pay authors to write books without government grants. Can’t educational publishing companies?

    • 14. Mark Guzdial  |  October 1, 2009 at 9:45 am

      It’s a great question, Tom. I don’t know. I do know that, within computing education, most innovative curricula receive NSF funding at some point in their development. I’ll bet that it’s similar for other science education efforts. Thus, the current model is dependent on government subsidies. Could it survive without those subsidies? Maybe. (Perhaps we should pass this bill and cancel farm subsidies at the same time, and see what happens to both industries in parallel? 🙂

  • 15. Mark Guzdial  |  October 1, 2009 at 10:20 am

    Tom, I do see your point about documentation on open-source software. However, this bill is about “educational materials such as curricula and textbooks.” I don’t think that software precedents apply here. Documentation about software is clearly not covered by the software license. Other curricula based on the funded curricula? Other books that contain elements of NSF-funded curricula and textbooks? This feels like new, murky ground to me — but I’m not a lawyer, nor do I teach a cool course on this topic, like Owen. So maybe my concern is ill-placed. I’d be interested in reading something well-informed that speaks to this issue of what this bill means by “open license” for “educational materials.”

  • 16. Eric Frank  |  October 1, 2009 at 10:21 am

    I’m a founder of Flat World Knowledge, mentioned above. I’m responding to Mark’s question – does Flat World Knowledge work. We are building a robust business around open-source textbooks. We raised $8m in private venture capital in Feb 09 and published our first 11 open source textbooks. We went from 30 universities and 1,000 student users this past spring semester to 490 universities and over 40,000 student users this spring. We are signing high quality authors, many of whom have successful textbooks at the majors. Paying them, and our sales and marketing organization (and everyone else involved) is a cornerstone of our model – we think incentives matter. It’s early days, Mark, but things are on a great growth trajectory.

    • 17. Tracy Wells  |  October 1, 2009 at 10:49 am

      I am employed in the traditional K-12 publishing industry, and I am really interested in finding out how businesses can make money while providing their content to customers for free. Anyone who has been in the business knows that development costs are huge, and profits are slim at the best of times (at least in my slice of the publishing market). How do you generate revenue to pay back your investors and to reinvest in your business? Are the programs ad-supported? Do you provide online content for free and charge for print supplementals? Or do you have another model?


      • 18. Owen Astrachan  |  October 1, 2009 at 12:07 pm

        Redhat uses a subscription model. You basically pony up for a year, then you get their services, their updates, their fixes, etc. This is for free/open source software.

        This could work for textbooks as well. As is mentioned above, errors and bug fixes could be a communal effort. Suppose paying adopters, e.g., those with a subscription, could contribute their materials that would then be made visible and incorporated into the next “edition” — paying homage and citations rather than money. There are all kinds of hybrid models.

        As someone else noted — if the licensing in getting a grant is too restrictive, don’t get the grant. Perhaps the publishers are having their cake and eating it too — someone else pays for the development, we take the profit. Despite what Pearson tells you, they don’t need to sell your book for $80. Except there is a used book market, except they have sales staff, except blah blah blah. Just call my a cranky old man

      • 19. Tom Hoffman  |  October 1, 2009 at 1:29 pm


        The key words in your comment are “their content.” Structuring any kind of business around producing stuff (books, software, widgets, whatever) and giving it away almost never works. You can build a business around open content or software, because you are providing services around content that other people mostly paid to create and, for one reason or another, were motivated to make open.

        What you have to keep in mind is that if this bill passes, going forward, your employer will have free and equal access to all the content and curriculum created with NSF funding. Your company will be able to add these resources to their commercial offerings at no additional cost (well, for most definitions of “open license,” I’ll have to check the language in the bill).

      • 20. Eric Frank  |  October 1, 2009 at 3:21 pm

        Hi Tracy – here is a link to a recent interview about our business model done for “Spark”, the national weekly technology show broadcast on CBC (public radio in Canada). Hope it helps answer some of your questions.


  • 21. Mark Guzdial  |  October 1, 2009 at 12:35 pm

    I suspect you were cranky when you were younger, too, Owen, so the age part doesn’t really matter 🙂

    I think these hybrid models are interesting and well-worth exploring. I don’t know how well publishers are doing. Since I hear about mergers and don’t see new publishers springing up (e.g., fewer publishers show up at SIGCSE each year, not more), I suspect that it’s not a booming business with lots of profit.

    I’ve been thinking about this bill and what it’s impact on me would be. The original MediaComp CCLI grant paid for doing the pilot course and first assessment. All of those materials are still available at http://www.mediacomputation.org — the Powerpoint slides, all the software, and all the research papers. *In addition*, I wrote a book. The book was not part of the deliverable for the grant. Of course, it wouldn’t have happened *without* the grant. Would the book be covered by this bill? Would it have to be open licensed, or could Pearson sell it?

  • 22. Tom Hoffman  |  October 1, 2009 at 1:40 pm


    You need to try to keep clear in your head that even if something is openly licensed it can be sold. In fact, my ongoing beef with people doing research and development in educational technology is that they have a strong tendency to distribute their software at no cost but a non-commerical (and thus not “open source” by definition) — thus preventing Pearson from selling the work. Open source advocates expressly require the option of commercial redistribution, by anyone.

    Beyond that, I can think of no reason or precedent that would require your book, or anything not directly funded by NSF, to carry an open license (well, it might make Richard Stallman sad, but pretty much anything you do hypothetically makes somebody sad).

  • 23. linda hahner  |  October 1, 2009 at 3:23 pm

    Mark, Thanks! I have been waiting ten years to have this discussion. I sat on the BOD of the Web3D Consortium when we moved two browsers into open source. Owen is obviously correct. There are differences when we discuss textbooks and software. In the future, we will discuss information. We won’t be so interested in the form it is delivered in. What I don’t like about the comparison to Red Hat or other Open Source software movements is if you develop a brilliant education program that is proven to work, you don’t want other folks to pick it up and water it down.This is new territory. Venture Capitalists were not interesting this area until the CTO of the USA praised CK12.0rg in Silicon Valley. Suddenly a bill. Very interesting developments.

  • 24. Christian  |  October 7, 2009 at 2:15 pm

    The only argument I hear against open source text books is one of profit.

    If the government paid for the work through the NSF, then it was funded with public money and thus belongs to the public (or logically should).

    If truly an open source book model is a non-starter for profitability, then the current publishing industry has nothing to fear as the model will simply starve out. If on the other hand there is potential then the current publishing industry has everything to fear.

    Given the success and quality of open source software, and the fact the exact same arguments have been made against it by incumbents in the field (Steve Balmer: “Linux is a cancer”), I think it’s safe to draw parallels between the software industry and the publishing business.

    Really this is a case of publishers worrying that they have to compete with ‘free’. In fact though, Linux is a viable competitor to Microsoft in the operating system space. Ironically, Microsoft and other have lobbied specifically to BAN open source software in the government. As a matter of brand, many government employees prefer the Microsoft brand. MS has lost market share to Linux and thus money but Microsoft is far from being out of business.

    In summary: Yeah, the publishing industry will probably see strong new competition. That competition is welcome if it drives the cost’s out of a textbook (Seriously? $200 for a intro to psychology book?). So, good for consumers, bad for profits.

    Guttenberg’s invention was a gift to the world — locking it up in laws is an insult to our collective humanity


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Trackback this post  |  Subscribe to the comments via RSS Feed

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 9,021 other followers


Recent Posts

Blog Stats

  • 1,977,499 hits
September 2009

CS Teaching Tips

%d bloggers like this: