I've been wondering about MathJax and the site's policy towards it recently. I am talking about questions which are otherwise good and up to site standards, but are formatted either without MathJax or with a limited amount of it.

I understand that new users usually do not know how to use MathJax, meaning that when a user posts a good, but poorly formatted question, the correct response is (1) edit his question and (2) mention the MathJax tutorial in the comments.

But what about experienced users which asked a lot of questions (say, more than $20$)? Should the community still have the same response?

I feel that users like that show an amount of laziness that damages the site, but at the moment, I completely understand why they do not bother with MathJax: there is no motive in doing so. I can ask a poorly formatted question and it will either get edited or even answered in a while.

So why do we even bother with warning people about using MathJax if we have no way to actually force them to use it?

  • 18
    $\begingroup$ Poorly formatted questions are more likely to be downvoted. So there's at least that incentive. $\endgroup$ – user147263 Jan 6 '15 at 8:02
  • 2
    $\begingroup$ A question that is not as well formatted as it could be is "damage"? Really? $\endgroup$ – Rahul Jan 6 '15 at 8:44
  • 1
    $\begingroup$ There was also this suggestion: Three strike rule to using MathJax. However, it looks pretty extreme. (And it got lost of downvotes, quite deservedly I'd say.) $\endgroup$ – Martin Sleziak Jan 6 '15 at 8:48
  • $\begingroup$ @MartinSleziak Yes, I agree with the downvotes there, it was an extreme suggestion I no longer agree with. $\endgroup$ – 5xum Jan 6 '15 at 8:49
  • 6
    $\begingroup$ @Rahul Damage may be a strong term, but yes, it does. If the precarious balance of people willing to edit and lazy people unwilling to post good questions shuold ever break, the site will devolve and no longer be what it is. $\endgroup$ – 5xum Jan 6 '15 at 8:50
  • 8
    $\begingroup$ What a dystopian future! Or, what might happen is that if we run out of people willing to improve poorly formatted questions, those questions will remain poorly formatted and hard to read, and will either be downvoted or ignored. People will spend their time answering the other well-formatted questions instead. $\endgroup$ – Rahul Jan 6 '15 at 8:58
  • $\begingroup$ See this question and in particular this answer for some reasons why using MathJax is a good thing even for simple mathematics. Of course, quid's example of incorrectly coded sin(x) should be corrected for this to be most useful. $\endgroup$ – Davide Cervone Jan 6 '15 at 12:31
  • $\begingroup$ I prefer users not using MathJax, over users using it in strange or wrong ways. For a simple example sin(x) is preferable to $sin(x)$; and this is not even an extreme example. Say, "the determinant of A" is fine but $determinant(A)$ is terrible and $det(A)$ not really good either. If we pressure people into using MathJax we will not get well-formatted posts but mainly terrible MathJax. Added: I do agree that the standard of the site should be to use MathJax for mathematics as explained by @DavideCervone. (Sorry, I reposted that comment, I wanted to modify something; it should before D.C.'s) $\endgroup$ – quid Jan 6 '15 at 12:37
  • 2
    $\begingroup$ @quid But for an experienced user, we would expect them to first write sin(x), then move to $sin(x)$, and then be experienced enough to write $\sin x$. So at least the guy writing $sin(x)$ is trying to write like we expect, and he eventually will get the hang of it... $\endgroup$ – 5xum Jan 6 '15 at 12:40
  • $\begingroup$ @quid, I agree, $sin(x)$ always bothers me as well, and will be problematic for those needing assistive technology. I think the solution is to fix the badly coded $\rm\LaTeX$ when we see it, but expect people to use MathJax for mathematics. Since this site is about learning new things, helping people to learn better $\rm\LaTeX$ practices seems to fit in with that. $\endgroup$ – Davide Cervone Jan 6 '15 at 14:23
  • 5
    $\begingroup$ @Rahul If the present is any guide, those question will still receive answers with meticulously formatted formulas including color highlights. No edits will be made to the questions themselves, however -- they will remain a blurry cellphone picture posted sideways, under the title "Can anyone help me?" $\endgroup$ – user147263 Jan 6 '15 at 16:30
  • 3
    $\begingroup$ I learned mathematical formatting in a modern sense through this site and by being guided by others and doing the tutorial. Very important to give people a decent chance, since this is advertised as a site for mathematics at every level. I was just out of date. My pet hate is the spacing which comes with "mod" e.g. $56 \mod 6$, where the second space is too big, which I tend to avoid by workarounds. $\endgroup$ – Mark Bennet Jan 11 '15 at 19:05
  • 3
    $\begingroup$ @MarkBennet Use \bmod: $56 \bmod 6$ gets you $56 \bmod 6$ with reasonable spacing. (I suppose you're familiar with \pmod for $56 \equiv 2 \pmod{6}$, but just in case I mention it nevertheless.) $\endgroup$ – Daniel Fischer Jan 12 '15 at 0:31
  • $\begingroup$ What you're really talking about here is TeX. Some people might know TeX but might not know that MathJax is the rendering engine on this site. And some people might not know what either of these things are, which would certainly be understandable for a newcomer and less so for someone who's asked more than 20 questions and gotten answers on at least half of those. $\endgroup$ – Robert Soupe Jan 12 '15 at 2:48
  • 2
    $\begingroup$ There is also the disincentive of users, while not down voting, simply ignoring questions which are not properly formatted. $\endgroup$ – Tim Raczkowski Jan 16 '15 at 1:48

For short/trivial to understand and edit posts, I can see how one might say that there is no incentive to learn MathJax, as indeed someone will likely fix it. However, in that case, the amount of effort an editor has to make to A: Understand the original question, and B: Fix it, are both minimal and therefore not really damaging to the site.

Once a question increases in complexity, the difficulty to parse the intent and fix it both go up in a more than linear fashion. At that point, the problem should be self correcting, in that people will see more negative votes and less useful answers with comments to the sort of "If you want more useful answers, learn MathJax so we can understand what you're asking!"


Here is an incentive: I will not read your post if it is not written in MathJax.

Why? Life is too short.

Well, what if I miss some wonderful thought? Not my concern.

There will be a subpopulation of users for whom life is not short and will contort themselves to read all sorts of horror so long as it resembles a math posting. But I want to have a conversation with someone who cares that I understand them as clearly as possible. If that makes me an old fogey, then so be it.

  • 1
    $\begingroup$ "I want to have a conversation with someone who cares that I understand them as clearly as possible." I feel the same way. There is a point where I stop caring what someone says if they make no effort in communicating their point in a coherent manner. $\endgroup$ – john Oct 10 '17 at 2:43

At the present, there is not much of an incentive. The average user who isn't committed to the site probably doesn't care about reputation or site etiquette. As long as the can keep asking questions, they will be happy. Yes, they could get downvoted but once the question is edited many of their post will need to maintain negative scores for a question ban to kick in.

If the community would be willing to work together, we could have an incentive promoting the use of mathjax to seasoned users (not considering brand new users since their experience with the site norms would be zero) that refuse to comply with etiquette. First, let's consider an example. Suppose you have a neighbor who borrows your tools but never cleans them when he is done. One day a new guy moves down the block so you warn him that Jim over there will borrow tools and never return them clean no matter how many times you have asked. Now, this new guy is abreast of the situation and may not let him borrow any tools. How can we apply this concept to the site?

If we know of a user who has refuse to help the community by learning a little mathjax, we can start warning other users via comment on their post--nothing rude or demeaning just the truth. If then the regular users ban together to stop editing and answering those post, this non-complaint user will have three options $(1)$ learn to conform to the site norms (path of least resistance), $(2)$ use another site, or $(3)$ quit using online help all together, but since many users use online help as a last minute aid for assignments, this probably will not be a realistic option for them. In order for this to work, we would all need to work together on this task since not all of us will be familiar with all users. If I know a user who refuse to work within site norms, I would need to make the comment letting the rest of the community know about this or someone else who isn't familiar will just edit and answer. Eventually, the regular active users will be aware of the major offenders and the site will cease to function how they would like. However, they can regain this functionality by learning mathjax.

Out of the three potential possibilities, $(1)$ and $(2)$ are probably the most likely and $(2)$ requires more work than $(1)$ since not all online math sites have a huge user base so finding one that gives them seemingly instant help may not be an easy task. Additionally, most people dislike change so learning mathjax is the path of least resistance since they wont have to go searching for a new community.


When you need attention and time of experts to a specific question, some of it should not be wasted this way. Do you read text-books that are not well printed or formatted? With Tex/MathJax nowadays you are able to almost reach that standard. You will also value it for following a uniform procedure and presentability. No pains no gains.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .