It is well known that Allen Hatcher has created a free textbook for algebraic topology that is high enough quality to be used in a large number of graduate courses in the united states, saving students a large amount of money.

Calculus is one of the most-taught courses in the united states; as such, many high-quality, expensive textbooks have been developed to teach calculus, often costing over $100.

What are the major obstacles to a group of mathematics educators collaborating on an open source calculus textbook with a high enough quality to replace a commercial calculus textbook?


4 Answers 4


To directly address your question, I would venture the major obstruction to universities developing their own textbooks is the publishing industry lobbying. You can read more about this in Educause Review. They state the major roadblocks are: discoverability, quality control, bridging the last mile, and acquisition. But here is an alternative that some universities are piloting to help with those.

Developed by Michigan State University, LON-CAPA or Learning Online Network with Computer-Assisted Personalized Approach is a course management and assessment much like Blackboard and WebAssign systems. The system is open-sourced under GPL, so has no direct licensing fees to use. At my institution, such a system is used to avoid the massive fees on the students that WebAssign currently uses as support. I just want to mention, that as a content manager, it is not meant as a complete replacement of Blackboard systems. In fact, it can work in sync with Blackboard, synchronizing grades between the systems. Here is an flyer about the LON-CAPA system.

For the content on such a system, it seems that some publishers do provide resources for the LON-CAPA system, but in more of a bulk pricing instead of price per student. A resource that I found is Course Weaver. Their pricing system seems much more reasonable than the large fees with ones like WebAssign.

There are other resources for free open source textbook as well. Check out College Open Textbooks. One feature of LON-CAPA is a cross-institutional repository of content for courses, so if you pick a textbook, you can use problems from other similar courses. You can check out the Step-by-Step guide for an idea for how it works.

Additionally, it seems that there is some research into the effectiveness of such a system: Experiences using the open-source learning content management and assessment system LON-CAPA in introductory physics courses - ResearchGate. [accessed Mar 20, 2015].

I must say, this is just what I've looked into. I have not interacted with this system, so I don't have any personal perspective on it. It seems I could be using it as early as this summer or fall, I can update then with a perspective if I use it. I'm also not in a position to comment on the future course content that we use just yet.

Another additional alternative is the Open Courseware from MIT. I have not looked too much into what they provide, but it seems comparable.

  • 1
    $\begingroup$ I just want to add as a comment. If you seek an open source textbook and you want some things to be online as well, you have to think about such systems. $\endgroup$
    – Chris C
    Mar 20, 2015 at 19:39

I am the author of a number of Free texts, for instance one on Linear Algebra that has some reasonable exposure. There are a fair number of such works and more are appearing. Check out the the Open Mathbook blog for some start at a community for authors. I see that the current entry references an open Calc text.

Such efforts would benefit from more support from the mathematics community and from administrations as scholarship, but there are some of us lucky enough to have convinced our colleagues that this is a useful contribution.

However your question is a bit unclear. If you really mean crowdsourcing, that would, in my opinion, be much harder. A classroom textbook whose style changes from section to section, or even from paragraph to paragraph, would be hard to use, it would seem to me.


I use Active Calculus by Matt Boelkins et al, and like it a lot. It is very good and totally open source (the author provided me the LaTeX files, which are licensed under Creative Commons, so that I could modify at will!) I also use and like (though they are less traditionally structured calculus texts, but in a way this is one of their strong points) the Calculus in Context series. For online homework I use Webwork from MAA, which is free and open source - you can even write your own problems and contribute them to a national database. More generally, I tend to use AIM's Open Textbook Initiative to find open source texts. I've had very good experiences with the texts I have found.

  • $\begingroup$ Thanks, edited to include links! :-) By the way, the MAA can host Webwork for you (for a fee), or if your helpdesk (or you) can run an Apache server, you can host it yourself. We are hosting it ourselves and it's going just fine - upkeep seems to be pretty minimal, and once you get the hang of setting up the courses in the first place, that's really the only work, which takes maybe a portion of an afternoon. $\endgroup$
    – Idempotent
    Mar 23, 2015 at 0:48
  • $\begingroup$ This is on my list of things to do. It would help if I had more colleagues on board... $\endgroup$ Mar 23, 2015 at 0:51
  • $\begingroup$ I did it and talked it up, and then a whole boatload of people got on board the next year. $\endgroup$
    – Idempotent
    Mar 23, 2015 at 14:08
  • $\begingroup$ PS. Also relevant to this thread: mathwiki.ucdavis.edu From the site: "The MathWiki is a collaborative approach toward mathematics education where an Open Access textbook environment is constantly being written and re-written by students and faculty members resulting in a free mathematics textbook to supplant conventional paper-based books. The development of the MathWiki is currently directed by UC Davis Professor Delmar Larsen." $\endgroup$
    – Idempotent
    Apr 3, 2015 at 2:25
  1. I agree very much with Hefferon but his Linear Algebra is not a very good example. Rather, it is too good an example. To write a book like this requires an immense amount of work.
  2. Re crowdsourcing. It seems to me that this would be extremely difficult because between the moment you start writing and the moment when you start to have adoptions, and you need to reach a critical mass, quite a long time will have elapsed. See for instance what is left of the NSF' "Calculus Initiative" of circa 1990. And it could be argued that without the millions the NSF invested, even Harvard's calculus text would not have made it. And even it did not last that long.
  3. Instead of crowdsourcing, then, I would try to go for collaboration with like-minded potential authors. But that too is not easy if only because of the problem of "How to write a book that is different" with "co-authors who might put different meaning on different".
  4. Finally, I also agree with Hefferon that a big problem is to "convince our colleagues that this is a useful contribution."

Here again, anyone interested in what experience I am basing these opinions on might want to look up freemathtexts



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.