I'm tutoring a high school student who is taking calculus for the first time. I've been trying to encourage her to become more interested in math, and so I thought it could be a good idea to teach her how to ask a question on math stack exchange. As a teacher, it seems like I failed at this, because the question I helped her write was downvoted and then closed for not meeting the math stack exchange guidelines. The question is here: Implicit derivative $y=\sin(xy) $

I know that it's a short question, but in my mind we gave the post a good title "Implicit derivative $y=\sin(xy)$," formatted the question properly, tagged it correctly, provided context "I'm taking calculus" and outlined two features of a possible solution "I have to use chain rule and then solve for $y'$." Working on her own, she wouldn't have known this type of problem was called an implicit derivative, or that chain rule and solving for $y'$ are required to solve the problem. To be honest, I'm not sure how much more there is that an asker could know about this problem without already being able to solve it.

I took a look at the guidelines on how to ask a good question. We could have included a source/motivation for the question, but I'm not sure how much good it would do the community to know that the problem came from a review sheet for her test. I'm trying to think about this from the perspective of someone who doesn't know how to solve this problem, and maybe also doesn't know precisely what aspects of it she doesn't understand. What else can such a user say in their question to avoid rejection from the site?

  • 2
    $\begingroup$ I think math.meta.stackexchange.com/questions/30038/… may be relevant. $\endgroup$
    – subrosar
    Jan 13, 2022 at 16:55
  • 15
    $\begingroup$ You've asked a number of Questions yourself in the past year, so I'd expect you have a handle on meeting the requirement for context. Something more than "Can you help me with this?" is necessary. But that Question can be edited, to explain the student's background, what research was done before posting, or simply explaining something about the problem to show the problem statement was properly digested, e.g.. how implicit differentiation is supposed to work (perhaps by connecting it to a previously worked exercise). The student can do that themselves. $\endgroup$
    – hardmath
    Jan 13, 2022 at 17:04
  • 5
    $\begingroup$ Someone is trying to help in the comment. I wonder why the OP just ignore it. $\endgroup$ Jan 13, 2022 at 17:05
  • 2
    $\begingroup$ @hardmath As I said in the question, the student does not even know what implicit differentiation is, so it would be difficult to explain how it is supposed to work. $\endgroup$
    – subrosar
    Jan 13, 2022 at 17:16
  • 18
    $\begingroup$ If that is the case, the student needs to back up and ask a more threshold Question. It is important to the learning mission that the person who posts a Question states it in terms they understand well enough to recognize a correct Answer for themselves. In the absence of that we would degenerate into "proof by authority". $\endgroup$
    – hardmath
    Jan 13, 2022 at 17:49
  • 9
    $\begingroup$ The question seems to have been downvoted and closed with little to no feedback. How is someone supposed to know why their question was closed/downvoted and improve their questions in the future if they barely had any feedback as to why it happened? (And I also am not a fan of how questions have to be asked the “correct” way in this community but that’s a different topic) $\endgroup$ Jan 13, 2022 at 18:14
  • 15
    $\begingroup$ There is actually a message available to the OP, which says "Please provide additional context, which ideally explains why the question is relevant to you and our community. Some forms of context include: background and motivation, relevant definitions, source, possible strategies, your current progress, why the question is interesting or important, etc." @blakedylanmusic $\endgroup$ Jan 13, 2022 at 19:28
  • 1
    $\begingroup$ I agree with @hardmath that it is simply ridiculous to ask questions about implicit differentiation if one does not even know the fundamental prerequisites. It's the same as those people asking how to prove Riemann's hypothesis without knowing what an analytic function is. $\endgroup$
    – user21820
    Jan 16, 2022 at 6:00
  • 4
    $\begingroup$ @user21820 In an ideal world, I agree with you. In the real world, students have to do homework and study for tests. Is it ridiculous to want to pass your classes? $\endgroup$
    – subrosar
    Jan 16, 2022 at 7:17
  • 5
    $\begingroup$ @subrosar: Of course, but that's not what the Math SE main site is for. Students who want to get an interactive learning experience should talk to their teachers first and use SE chat if their teachers are incapable of teaching them. There is just absolutely no reason to dump bad questions on Math SE that show zero effort. $\endgroup$
    – user21820
    Jan 16, 2022 at 8:24
  • 3
    $\begingroup$ We had a very similar discussion just a short time ago: math.meta.stackexchange.com/questions/34409/… . The general consensus was that when the question is closed it doesn't mean it is dead. It is in sort-of "on hold" state where the OP is encouraged to provide feedback before anyone is allowed to answer. This is to prevent people from answering while it is not clear what the question should be. The OP can keep engaging with the community via comments, and add more information to the question, then request it to be reopened. $\endgroup$
    – user700480
    Jan 17, 2022 at 12:56
  • 3
    $\begingroup$ @user21820 The chat is not accessible for new accounts, you need 20 points. The only action available to a new user is asking/answering questions. Personally I've had many beginner questions that I considered "not serious enough" for the main site and would have liked posting them on the chat but did not have access to it. meta.stackexchange.com/help/privileges $\endgroup$ Jan 19, 2022 at 9:43
  • 2
    $\begingroup$ @MehdiSlimani You're right, access to chatrooms is unavailable for users below that rep threshold : alternately, it takes just three up votes to cross it, so one well-received question is enough. I would suggest this chatroom to those who have over 20 rep, specifically for question improvement. I want to know what you mean by "not serious enough" : which specific criteria do you think could be used to close such a question? $\endgroup$ Jan 19, 2022 at 14:06
  • 2
    $\begingroup$ @SarveshRavichandranIyer I had in mind: 1. Questions that save time: the documentation or theory contains the answer to the question but the person who asks has trouble navigating it and is looking for a shortcut from a person with more experience. 2. Questions to check understanding of a subject: The person who asks thinks the theory states something but is not sure. The person summarizes what he understood and asks whether they are correct. $\endgroup$ Jan 19, 2022 at 15:25
  • 2
    $\begingroup$ @StinkingBishop: In fact, are you aware that when the author edits a closed question, it automatically triggers a review? So the system is actually biased towards them, if only they put in the effort. $\endgroup$
    – user21820
    Jan 19, 2022 at 18:46

1 Answer 1


First of all, thank you very much for attempting to educate your student about the guidelines and how to ask a question here. I sincerely hope we can help you and them grow as contributors.

On question gaps and OP

Roughly, when a question is asked by an OP (OP = Original poster, the asker of the question), there is a certain "gap" between the point where the OP is, and the actual answer. On occasions where it's clear that the gap can be filled with the information provided by the author, in a manner that the author understands it, someone posts an answer.(The guidelines capture these key pieces of information). Otherwise, people comment, in an effort to bridge the gap and make it a collaborative effort with the poster if necessary. If OP does nothing to address these gaps, then two problems occur :

  • Someone posts an answer, and I , as a visitor, don't know whether it's worked for the OP or not : I could say it's useful to me and vote it up, but I wouldn't know OP's response. It could perhaps stunt the organization of the site as well. For example, if your student had accepted the second answer in that question, then we'd have probably closed the question as a duplicate, given that duplicates using implicit differentiation are available.

  • There is no subsequent activity by commenters on the post because the OP hasn't help bridge the gap. Then closure/deletion etc. follow because the gap is likely to remain as such without OP participation.

Either way, the most important reason I'd imagine you asked your student to come to this site is because they would benefit from it. It is likely that neither of these situations is beneficial to them, unless an answer is perfectly on point : in this question missing that key piece of context regarding implicit differentiation, both answers use it!

What your student needs to do for sure

That's why I really think there's one very important thing that your student , and many posters when they don't know if they've provided enough context, need to do very diligently : it's to be receptive and reply to every comment/answer, and admit the fact that some things are going over their head if and when they are (which can be difficult to do particularly if the student fears being seen as dumb or feels embarrassed at not knowing certain facts/results). Therefore, your student should know that they can comment and refer to users using the @ functionality. Really, I genuinely believe this is the most beneficial action for a user that isn't sure on what context to add.

There will be two types of comments (some of both types). One will tell them what to add to their post. The other, will attempt to resolve their mathematical question. Both will be important, one to find the context in the question and the other to place that context in the original post. There are commenters that are willing to converse, improve the question and eventually help the user out : particularly if the question is of an elementary level and hence attracts a larger audience.

In this question e.g., a conversation may look like so :

  • Commenter : Do you know implicit differentiation? Here(link to some post) is a post that solves the same question using implicit differentiation.

  • Poster : No, I don't know implicit differentiation.

  • Commenter : Thanks. If I take out implicit differentiation, then it's likely that I might require a different tool. Are you aware of the arcsine , or inverse sine function?

  • Poster : Yes , but how can I use it here?

  • Commenter : Can you write $x$ in terms of $y$ using the arcsine function?

  • Poster : I think so , $y = \sin(xy)$ so $\arcsin y = xy$ so $\frac{\arcsin y}{y} = x$.

  • Commenter : Thanks. I believe you can find $\frac{dx}{dy}$ from here?

  • Poster : I don't know what is $\frac {dx}{dy}$ / Oh, then I use $\frac{dy}{dx} = \frac 1{\frac{dx}{dy}}$? / I don't know how to differentiate the arcsine function?

  • Commenter : Thanks: by the way ,consider adding ... to your post (although with newbies, perhaps context in comments may be good enough) , do you know the differentiation rule for inverse functions? etc.

and so on. If the question is closed then conversation can carry on, with emphasis on what can be added to the question for it to be reopened.

If I were a commenter, I'd probably ask for details on the textbook , questions such as : "did this appear as an exercise? If so, what was the last thing taught to you before you reached this question?" That might be enough context and is part of the "source" of the guidelines.

Benefits and issues

Providing the OP sticks it out in the comments, I can see one of two outcomes :

  • No commenter is able to completely help the OP, despite everybody's honest and entire efforts. In that case, one of two things happen : either someone writes a lengthy exemplary answer that addresses the OP's concern with its sheer breadth, or the OP recognizes the gap and attempts to bridge it with further questions/reading elsewhere. In short, it's a productive experience for the user.

  • The OP reaches a point where they can self-answer the question, or someone else can clearly see a way for the OP to reach the answer from the question using tools available to them that have come out in the comments, and writes an answer.

That shows clearly that open participation in the comments is the best way for users that may not know what context to provide to benefit from the question they've asked. The big problem is that a lot of things need to go the right way for the best experience : answerers need to be careful they don't confuse the OP, down-close votes should do their job and prompt improvement, and everybody including the OP needs to be available and open for conversation. Seven-eighths of the trouble is that one of these break and the question dies, with someone else to pick up the pieces perhaps.

Of course, another problem is that people are split over whether the site should be accommodative of questions with many back-and-forth comments , given that SE itself isn't very welcoming of discussion : but if someone doesn't have context, then they need to discuss this with users, and I think visitors and potential answerers are likely to be helpful in figuring out what context is needed when a question is posted, or while it is closed. That's my personal take on this issue.

What posters should definitely know as well

I would ideally love it if any new user read a MathJax tutorial, the guidelines page, and How to search on this site prior to posting questions. I still believe the most important thing is to be receptive to comments on one's question and respond quickly. MSE is not Quora,reddit etc. so the format and system may be off-putting. In any case, we need to help new users integrate, and likewise new users need to make sure they want to be helped. I consider what I've said to be the primary way to do it smoothly.

  • 4
    $\begingroup$ Thank you for this thoughtful response. Before reading this, I hadn't understood the importance of context for a post: it's just as important for someone answering a question to understand what OP doesn't know as it is to understand what OP does know. The suggestion to engage more with commenters is a good idea too. $\endgroup$
    – subrosar
    Jan 14, 2022 at 6:38
  • 3
    $\begingroup$ @subrosar You are welcome! In short : If I don't know what to add to my question even after reading the guidelines, I post it and keep adding to it, what people think will aid the question , until someone answers it or I've learned what the gap is and what I need to do to bridge it. I would encourage your student to actually try this with her existing post : it's only closed for answers, not conversation. $\endgroup$ Jan 14, 2022 at 7:55
  • $\begingroup$ @subrosar Hey, I just happened to remember this question and answer pair (I came here after voting up your answer on the automorphism group of a bipartite graph). I sincerely hope these tips helped you mentor other students in contributing here. If you wish to integrate any students that you think will benefit from this platform, let me know and I can try to format their questions well. If I may : how is the student doing whose question was discussed in this post? $\endgroup$ Jul 4, 2022 at 11:00
  • $\begingroup$ I really appreciate your help. That student graduated from high school and is moving on to college. I'm starting a PhD program in the fall, so I should have the opportunity to send more students to you soon. Thanks again! $\endgroup$
    – subrosar
    Jul 5, 2022 at 3:27
  • $\begingroup$ @subrosar Really nice to know. I hope the website can help them grow as mathematicians. $\endgroup$ Jul 5, 2022 at 5:54

You must log in to answer this question.

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