Worth introducing a mandatory short module on $\LaTeX$ into a mathematics degree?

Question

On Mathematics StackExchange for a particular instance, it is highly recommended that questioners mark up their questions using $\LaTeX$. A surprising number of mathematicians and student mathematicians are perfectly happy to admit that they don't know $\LaTeX$ and are not planning on learning it.

To me, that is an attitude which is beginning to make less and less sense as time marches on. Perhaps 12 or 15 years ago $\LaTeX$ was not necessarily the best option, but now we have such convenient and (fairly) high-quality tools like MathJax, it appears to me that for a mathematician not to want to learn $\LaTeX$ is akin to an English student not prepared to learn how to use a word processor.

What is the general opinion of people in the educational and academic sphere: is it feasible, desirable and / or possible to make it a general requirement for a degree in mathematics, or physics, or even engineering, to attend a short module on developing fluency in basic $\LaTeX$ of some description?

I understand that certain word commercial processors may have extensions which allow the generation of inline mathematics, which themselves use a compiler that builds some interim $\LaTeX$ code, but these tend to be cumbersome and unwieldy, and can cause unnecessary bloat, and have considerable limitations.

Answer 1

I do think that having an introduction and primer to LaTeX in the course of a college math/computing degree is appropriate and beneficial.

For example, I work this into my discrete math course, and require that weekly submissions to the Discussions board be in that format (the Blackboard LMS supports LaTeX entry). My community-college students often have notable skill deficiencies, but even here, the students who will pass the course can usually pick it up in a day, from a one-paqe summary handout.

I agree with Ben Crowell that an entire dedicated course on the subject would be overkill.

Answer 2

It is true that $\LaTeX$ is a language, but I lived through the transition from WordPerfect and WordStar to Microsoft Word, and I lived through the transition from troff to $\LaTeX$. Already Overleaf, MathJax and other similar software developments are lowering the TeX entrance-threshold.

In the future, $\LaTeX$ will likely move "under the hood" as have other typesetting languages. This argues for supplemental instruction, rather than new courses and new requirements.

Cf. Asimov's The Feeling of Power. Could I be the last troff expert? :-)

Answer 3

(feel free to change to comment):

1. If you are going to ADD something, then you need to make an argument over what to cut. Or say camels have it easy and can carry more straw (they don't, but at least adresse the issue). I see this mental lapse all the time. Life is an optimization/choice problem. Not an "add" problem. By the way, starting to think about this would have made even your forum-y post a little more thoughtful. You would have prompted yourself to consider duration for instance, addressing the Dan Collins object.

2. I would beware of assuming that the reception you recieve on Stack Exchange (a computer-y platform) is the same as what the bulk of teachers or students feel. I've noted a huge pro LATEX bias here...even to the extent of thinking it's expected. But the rest of the world seems to get by surprisingly fine without it.

3. (related to 1) "Required" is a big stick to swing. After all, DC mentions how he does it ad hoc without a requirement.

4. Why are you sure this needs to be covered in a class. I've actually never used it (not a math major granted). If I ever needed it, I think I could get it done WHEN I NEEDED IT. I mean we routinely get the calculator/WA pushers saying we should spend time on that. But a strong pen and paper student can completely add that stuff, very easily, when he needs it. (And the weak pen and paper, spent time on gom-puter crutch, not so much in reverse.)

Answer 4

At the University where I'm teaching, there is such a course mandatory for all students (not only in math, but also physicists, chemists and computed scientists) and recommended in the first year of studies.

It's a 1 credit course (meaning the it "officially" assumes about 25 hours of work), and it's offered online - the student is supposed to read tutorials, practice, then take a practice test, then the exam; they can also ask questions from the teacher.

I would say it's a very reasonable arrangement. The students will, at the very least, be required to write a bachelor thesis, then possibly a master thesis. At which point, if you don't offer such a course, you would just assume that they are somehow supposed to have learnt to typeset math on their own. Requiring something they haven't been explicitly taught may be frustrating (although that is culture-dependent, of course). And quick learners will just easily pass the exam.

Answer 5

One clear objection to teaching Latex to all undergrads that would be raised in a faculty meeting would be: Latex indeed is pervasive in mathematical and theoretical physics publications (it is not that common in experimental physics, I don't know about engineering), but the majority of undergrads will not stay in academia, but will enter the job market once they graduate.

Now, one may argue that Latex is such a powerful tool that even outside academia one could profit from learning it. But if other people in the company do not know Latex, then its usefulness declines seriously, since you cannot share a project.

A more subtle objection would be about portability. Recall that Latex was created as a typesetting tool that would generate the same document in any computer. One chemist noted to me that, in the age of smartphones, Latex-generated documents lose their edge because although high-quality they are difficult to read in small screens.

EDIT: To clarify the portability issue, this is relevant for the majority of people who will end up outside academia and will be producing documents which are expected/advantageous to be read in small devices. For writing papers or student examinations portability is irrelevant and arguably Latex is by far the best option.

It is important to consider that Latex is not just a tool to write equations, which is the part that mathjax implements, but a full typesetting application. It is easy to learn how to produce a standard document, but extremely complicated if you want to be able to change the diagrammatics, not to mention that Latex is infamous for having one of the worst error messages of all time. I would wager that the average user of Latex is happy to send a document whose output contains errors as long as it compiles and "looks fine".

Wrapping up, while I still love latex and continue to spread its gospel, it seems unlikely that the average student would benefit much from learning Latex to justify one extra module that will consume a faculty member, specially considering that while everybody in a math department uses latex, the majority of mathematicians don't know enough of latex to teach it in a course.

Answer 6

I think it is not a good idea to oblige students to take a course on latex. Students do need to learn to write, and they can be taught to use latex as part of being taught to write, if need be, but the basics of latex are easily learned independently, and students need to learn to learn things independently. What might not be obvious to students is that it is useful to learn latex, but this is not hard to communicate.

My experience with directing undergraduate final projects written by engineering students is that some learn latex to write their projects, others get by with word/open office, and it's not necessarily a good use of time for some students to be learning latex, although it certainly makes sense for a student whose written project will incorporate a large number of mathematical expressions or equations. On the other hand, in my experience a student whose project is focused on coding (who could learn latex in a day with no difficulty) generally has little need for what latex offers.

One should also remember that while learning enough latex to write a math paper is very easy, learning enough latex to format nicely a document full of images and tables can be quite time consuming, and, there are alternatives both easier to learn and easier to use well.

There are lots of tools that students can learn - AutoCad, C++, Matlab, Rust, Innovator, Latex, Maple, Sage, R, etc. and there should be some selection because no one can learn everything nor does anyone need to learn everything.