The best class I've ever taken was an intro class in statistics, in which the professor recommended... no book. He had an entire set of notes, practice problems (with extraordinarily detailed answers), and problem sets. Every year, he creates new exam problems so the old exams with answers become practice material. Honestly, his lecture notes (and all the additional material) is so vast that a student would hardly have enough time to study anything outside the notes.
My point is, if a lecturer is really dedicated to providing the best possible experience to his students, he doesn't need a textbook. I'm not saying that it's easy, or that it doesn't require a lot of time and effort from his part. It does. Some lecturers chose to accept the load, while others go the easy way and just point to references and exercises in the book.
That answer is essentially, "each professor should write their own textbook", since an extensive series of notes intended to be used in a course of study, along with practice problems, is pretty much a draft textbook.
That could be an interesting solution, but is mostly practical if you already have tenure, since it is a quick way to not get tenure otherwise. If you spend your junior professorship time writing a textbook (in a field that already has many) rather than spending that time on more research and publications, tenure is unlikely. If you're tenured and have been teaching the same course for a while, on the other hand, you both have more leeway to spend your time doing that, and may have already sort of "naturally" accumulated years worth of notes from previous iterations of the course. Then it can be a good idea, and many senior professors do prefer this approach, because it lets them teach the course in a way precisely in line with how they feel the material should be taught, rather than having to buy into someone else's approach.
Universities could influence that outcome at all levels if they wished, though. Universities formally budget preparation time per course. Currently it's typical to budget quite little, because the assumption (and sometimes requirement) is that you'll be teaching using a standard text in the field. If they instead want extensive preparation to be done, along the lines of what you propose, they can: 1) budget sufficient time for it, e.g. specify that each course requires, say, 100 (or 200, or whatever) hours of preparation prior to the beginning of the semester; and 2) expect that the preparation time be used towards the budgeted end (i.e. expect extensive independent course materials to be developed for each course, in line with the amount of time budgeted).
> "each professor should write their own textbook", since an extensive series of notes intended to be used in a course of study, along with practice problems, is pretty much a draft textbook.
This isn't really true. A majority of my math text books at UW were written in note form by the professor teaching the class. They were just very efficient...and hey, its math where a glossy text book really gives you nothing.
> Currently it's typical to budget quite little, because the assumption (and sometimes requirement) is that you'll be teaching using a standard text in the field.
This is definitely not true in math and barely true in computer science. I think it is true in physics and chemistry, but only for entry-level courses. It also depends on the university.
Open sourcing material between professors could go a long way; each could simply collect what they want for the course and then easily publish it in electronic (free) or paper form (students pay printing costs). I think this is how it mostly works already in upper division courses.
Sometimes there is a very nice up to date book; but just as often, say in computer science, everything is out of date anyways (computer architecture, compilers, ...).
> This is definitely not true in math and barely true in computer science.
I haven't surveyed everywhere in CS, but I know the operations of quite a number of schools, and I don't know any that budget significant preparation time for "normal" classes. Some do allocate extraordinary preparation time if you're developing a completely new course that's never been taught at the school before. But in the common case, where it's a longstanding standard course, you're expected to use reasonably standard material, whether it's whatever was used in the previous professor's version of the course, or some moral equivalent that you're expected to be able to adopt within <10 hrs. You definitely aren't supposed to (officially) spend more than a minor amount of time preparing your own curriculum and pedagogical materials for the course.
Furthermore, the first edition of most[1] textbooks suck- draft books much more so. I think that this would be a recipe for disaster.
[1] Note to the pedantic: I am taking license with the fact that I've only read a few first edition textbooks- and they sucked. "Most" in this case should be read as a rhetorical vehicle.
That answer is essentially, "each professor should write their own textbook", since an extensive series of notes intended to be used in a course of study, along with practice problems, is pretty much a draft textbook.
Is the purpose of the textbook for problem sets, or as reference? In either case, previous editions are often much less expensive than the most modern edition (sometimes the professor already owns the previous edition too). But if the purpose of the textbook isn't for the problem sets, then there are likely inexpensive or free (by author or by copyright expiration) books that are fine for the other purpose. (I have a liking for Schaum's Outlines that cover a nice spread of topics.)
I'd add that if cheating is a concern (whether from using books whose solution set is easily available, or using in-house solution sets that don't update with each year), then assign watching a previously recorded lecture (by the professor, or one of the excellent already available ones online) as homework, and use the in-class time to have quizzes, tests, or professor-assisted homework sessions. In addition to weeding out cheaters there's the benefit, if you believe Salman Khan, that such methods are probably better ways to learn the material.
Finally, it's not too time-consuming for a professor to scan the problem sets from the book and distribute them to the students. So long as the problems aren't highly specific to the text, the student ought to be able to solve them without the text. For many subjects there are lots of free resources out there to learn from.
Hopefully some folks have heard of Boundless now, we make free e-textbooks out of open-source / public domain content and are now used by students at half of all universities.
We have a program to help educators quickly get up to speed using our content with minimal effort
Unless you're teaching at the very cutting edge where new research is emerging regularly, (in which case I imagine you'd be teaching from your own notes and journal articles anyway), then how about just intentionally using an older text and write your own exam questions? There are usually plenty of copies of "recently out of date" editions of most texts on the second-hand market, and the fundamentals of the material probably haven't changed that much.
Or find one of the many "released free for use" maths texts[1][2] that are "out there" and see if one is suitable for your classes?
>how about just intentionally using an older text ...
Tried it. Those god-damned greedy SOB's buy up all the previous editions and shred them. It's hard to find copies even on the internet after a couple semesters. I still list previous editions of texts on my syllabi though, but students who are on fin-aid are herded toward the campus bookstore.
Ah, interesting. I've seen older texts listed on Amazon.com many times and had always assumed that getting them was pretty easy. OK, well, nix that idea then. :-(
That might work at some universities, but there are a (not insignificant) number of students who simply refuse to buy used textbooks. After all, they aren't paying $20k per semester to learn old calculus!
The sad reality of the situation is that a professor's student evaluations (which do matter for things like tenure) will be negatively affected by the decision to use an old edition. Yes, sometimes it makes sense to tell an unprofitable/annoying customer to go away, but professors don't have that luxury in the university environment.
Edit: Curious why this is being down-voted. The truth hurts?
Write up lecture notes, and design your own assignments. This is how a lot of my upper division math classes were setup, and the arrangement worked quite well. Textbooks are usually full of fluff not relevant to the class. Lecture notes, being designed by the person teaching the class, only contained material directly related to what was being lectured about / tested on. Of course, creating comprehensive lecture notes is not an easy task, but it's well worthwhile - write up your lecture notes once, and you can reuse them every year, without your students paying $100 for a textbook they'll never look at again.
MIT CS class (6.047) I just took had a mix between course notes and a textbook. It started as a compilation of course notes and then each year, was improved by the class. During each lecture, one or more people were assigned to update the relevant chapter of the book to reflect this year's lecture/changes in the field. Some chapters that were the core of the class or didn't change from year to year were really good and some were only a basic framework. He also recorded all his lectures, so that helped with any places where the book might not be ideally up to date.
Tell your students not to purchase whatever books the department head has been bribed into using in the coming semester. Instead, spend a few hours curating online resources that may serve as notes for students; better yet, just photocopy the handwritten notes that every mathematics teacher I've ever had seems to have for each lecture they give :-) Otherwise, Paul's Online Math Notes tackles Calculus and ODEs, and IIRC gives an intro to PDEs as well. The students collectively save thousands of dollars in exchange for a few hours of your time spent photocopying notes or otherwise googling for a resource that you approve of.
> Instead, spend a few hours curating online resources that may serve as notes for students; better yet, just photocopy the handwritten notes that every mathematics teacher I've ever had seems to have for each lecture they give :-) Otherwise, Paul's Online Math Notes tackles Calculus and ODEs, and IIRC gives an intro to PDEs as well.
I recently taught a course on ODEs, and it took far more than "a few hours" to curate enough online resources to cover the curriculum. The sad truth is once you get out of calculus, the number of quality online resources for advanced mathematics decays rapidly, and the coverage gets spotty as well.
This is not to mention that those resources rarely agree on notation, which makes them more confusing to students (and therefore less helpful). There's something to be said for requiring a single edition of a single textbook, just for consistency's sake, even though that's caused all this publisher drama in the first place.
Our CS department head was co-author on 5 of the text books for the degree program. Several other professors were also co-authors. I don't think he needed to be bribed, I think he was whole-heartedly for it.
In most fields that would be called a "conflict of interest". When you're lucky enough to be in such a position, you're right, you generally don't need to bribe yourself.
Except that most reputable universities don't let you do that. You can use your own textbook for the course, if you want, you just can't accept any royalties for it.
> No USF employee may receive any money in exchange for requiring a textbook. However, if the professor is the author, royalty payments may be received.
It references and quotes Florida Statutes 1004.085. "An employee may receive: ... Royalties or other compensation from sales of textbooks that include the instructor’s own writing or work."
> Harvard has no policy on whether professors can profit on the sale of texts to students in their classes. The University leaves that decision to the individual instructor, according to faculty officials.
> Since N. Gregory Mankiw returned to Harvard to teach the College’s introductory economics class, 2,278 students have filled his weekly lectures, many picking up the former Bush advisor’s best-selling textbook, “Principle of Economics” along the way. ... So, what has professor of economics Mankiw done with those profits? ... “I don’t talk about personal finances,” Mankiw said, adding that he has never considered giving the proceeds to charity.
Given that Florida law doesn't prohibit collecting royalties and that a Florida school and a Massachusetts school both allow collecting royalties from assigning one's own text books for class, I think the burden of proof is on you to show that "most reputable universities don't let you do that."
(In any case, I went to university about 20 years ago, and the laws and rules were different than. But it doesn't look like it's changed all that much.)
In addition to the wonderful suggestions below, don't forget that most publicly available resources will (most likely) gladly accept revisions and expansions on topics that you don't feel are "up to snuff" for your class.
It's, admittedly, more work, but what good thing isn't?
The most practical thing to do would probably be to start a book bank; or, just have the department library stock up on textbooks which can then be loaned to every class that uses it. Those that want to keep the textbook can buy their copy, but most, who just want to read the material for the course, will happily use it
Most university libraries will keep 1 or 2 copies of a book on loan so you can study/photocopy them while in the library. The last few semesters of school I wound up doing this instead of buying the books.
First, unless you are in a very specialized field, there are surely several books that cover the topic you are lecturing. In the interest of politics have an official book, but in the interest of time offer alternatives. Your students, or the good ones, at least, will be able to use whatever book they can fit in their budget, including previous editions. Especially if you're teaching calculus.
Second, if you are grading homework then produce your own problem sets. You already have to do this for exams, which means you know what you expect your students to learn, so you may as well produce the problem sets as well. At the very least indicate what types of problems a student should practice on so they can follow along in the book they do have.
Third, thank you for lecturing. I know that lecturers hate the raw deal with textbooks as well.
Having to support many different textbooks across different students is a recipe for chaos. Especially when the topic is something like calculus which can have such radically different treatments.
Would you please recommend what I might do, subject to severe limitations of (1) time and (2) political clout within my department?