To gain our seal of approval an open source mathematics textbook must be able to serve as the primary text in a mainstream mathematics course at the undergraduate level in U.S. colleges and universities. That means that we are not evaluating instructional modules, Java applets, supplementary lecture notes, or other materials that are designed for limited use within a course. Since the minimum length of a traditional course is 10 weeks with 30 hours of class instruction, the books we evaluate must have enough material for that, and most will have more since most college courses are 14 or 15 weeks in length with at least 40 hours of class time.
We are looking for books that we can recommend for consideration to any of our colleagues, books that are suitable for use in traditional university courses. Books must be mathematically sound and written in standard English with evidence that they have been proofread and edited. They need not have color, flashy graphics, or ancillary materials. They must have exercises, and though it would be desirable to have short answers to many of the problems, it is not necessary for approval. It would also be desirable to have more complete solutions available—at least to an instructor.
Furthermore, we expect books of the quality we seek to be class-tested. They should have been used (and be in current use) by faculty other than the author. We are certainly interested in fostering the writing of open source books, and so we may encourage people to try out a book that looks promising, but we would not label such a book as having our full recommendation.
We do not require any particular form of license, although the Gnu Free Documentation License and the Creative Common licenses (in its many varieties) are what we recommend. These licenses allow the book user the greatest flexibility in making use of the material. We will approve a book with copyright restrictions as long as the book is available in some format at no cost. It is not necessary that printed copies be available from a print on demand service like Lulu, but many students and faculty are willing to spend $25 to get a paperbound or spiral bound copy of a 500 page text.
We are impressed with books that have a support system, signs of which are a website that is maintained, a community of users, a means for submitting support material from faculty such as test questions and project ideas, and a mechanism for improving the book by correcting errors, publishing errata lists, and displaying user comments.
For books that use mathematical software, we consider it desirable to use Sage, which is open source. It makes little sense to have a free textbook that requires expensive proprietary software.
The evaluation process is ongoing. We began several years ago with texts for linear algebra and abstract algebra and now we can recommend books in 19 different course categories. To recommend a book for consideration, please send a message to textbooks@aimath.org.
I have a book on the list. I'm lucky enough that at this moment I'm at an institution that will cover the download traffic. But my prior school actually withdrew that support at some point, and in any event I will retire someday. I don't think I can ask my wife for us to pay for that. (I won't mention the possibility of a deadly bus encounter.)
In thinking about where the work might live in order that I could hope it is available, say, for the next decade, it is reasonable to look at people who can use the traffic to keep their lights on.
I used to use Sourceforge. They became unacceptable. I switched to Google code. They turned off, as I understand it because people were uploading pdf's (pitated ones, but big tech seems not to want to make a distinction like that). Then I used GitHub and in the run up to getting bought they made some changes that I thought looked ominous. So I am now at GitLab. Can you blame me for wondering if it is too much to hope that a site will freely host me for, say, a decade even if I go away?
By the way, I put a stable version of my latest on GitLab.
To gain our seal of approval an open source mathematics textbook must be able to serve as the primary text in a mainstream mathematics course at the undergraduate level in U.S. colleges and universities. That means that we are not evaluating instructional modules, Java applets, supplementary lecture notes, or other materials that are designed for limited use within a course. Since the minimum length of a traditional course is 10 weeks with 30 hours of class instruction, the books we evaluate must have enough material for that, and most will have more since most college courses are 14 or 15 weeks in length with at least 40 hours of class time.
We are looking for books that we can recommend for consideration to any of our colleagues, books that are suitable for use in traditional university courses. Books must be mathematically sound and written in standard English with evidence that they have been proofread and edited. They need not have color, flashy graphics, or ancillary materials. They must have exercises, and though it would be desirable to have short answers to many of the problems, it is not necessary for approval. It would also be desirable to have more complete solutions available—at least to an instructor.
Furthermore, we expect books of the quality we seek to be class-tested. They should have been used (and be in current use) by faculty other than the author. We are certainly interested in fostering the writing of open source books, and so we may encourage people to try out a book that looks promising, but we would not label such a book as having our full recommendation.
We do not require any particular form of license, although the Gnu Free Documentation License and the Creative Common licenses (in its many varieties) are what we recommend. These licenses allow the book user the greatest flexibility in making use of the material. We will approve a book with copyright restrictions as long as the book is available in some format at no cost. It is not necessary that printed copies be available from a print on demand service like Lulu, but many students and faculty are willing to spend $25 to get a paperbound or spiral bound copy of a 500 page text.
We are impressed with books that have a support system, signs of which are a website that is maintained, a community of users, a means for submitting support material from faculty such as test questions and project ideas, and a mechanism for improving the book by correcting errors, publishing errata lists, and displaying user comments.
For books that use mathematical software, we consider it desirable to use Sage, which is open source. It makes little sense to have a free textbook that requires expensive proprietary software.
The evaluation process is ongoing. We began several years ago with texts for linear algebra and abstract algebra and now we can recommend books in 19 different course categories. To recommend a book for consideration, please send a message to textbooks@aimath.org.