I've been on this site for several years and I appreciate the site a lot but I don't like to post here anymore because no matter what I write I get some kind of sarcastic comment, or put-down, or someone edits my post (removing and adding tags, etc.). I realize that you all need to monitor the site, and that I ask stupid questions sometimes (usually I am thinking of something more and I just want a reliable resource for my little question) or ask weird questions. But, I don't see what is the harm in keeping these questions up and just not saying anything unless you have something helpful to say. I posted a question about Liouville's Theorem and I just wanted to make sure that I was thinking about it in the right way. But I got people editing my post (adding and removing tags) and a wikipedia link (why is that helpful?), so I removed the question. I think that this environment is not encouraging new ideas because I can't ask the questions I want to ask and I can't say things in the way that I want to say them for fear of all this negative feedback. I'm frustrated because I would just like a little help, not to be made fun of for asking. Any ideas? The sarcasm and patronizing tone of some of the comments really bothers me.

  • 7
    $\begingroup$ While I understand the frustration, this might be better as an email to the moderators. When they think it is appropriate, they may compose a meta post for you. I do think you should talk about it, but check with the moderators first. Gerhard "Sorry This Happened To You" Paseman, 2017.03.22. $\endgroup$ Commented Mar 22, 2017 at 23:41
  • 8
    $\begingroup$ Thanks, I just want anyone to see it. And thanks for your comment on the question I posted. That and the other helpful comment that followed is all I really needed and it could have just ended there and it would have been just fine. $\endgroup$
    – user10290
    Commented Mar 23, 2017 at 0:04
  • 32
    $\begingroup$ I think this site is the right place for such enquiries -- but it might help if you provide a link to your removed question, such that people can see what you worry about. In general, editing a question or changing tags is in no way negative feedback. Negative feedback are downvotes, close votes, delete votes, low quality flags and negative comments -- which of these did you get, and how often? -- In any case, I think you should not worry in any way about posting on MO -- I see you have asked 20 well-received questions and given 2 well-received answers so far. $\endgroup$
    – Stefan Kohl Mod
    Commented Mar 23, 2017 at 0:38
  • 7
    $\begingroup$ Thanks, I do take it very personally. I think tags are subjective and should be the questioner's choice. I put real-analysis and set-theory for the question about transcendentals because to me the uncountability of the transcendentals is what began set theory. And the larger question in my mind is about set theory. But someone removed that so I had to defend it in my question. Then someone removed the real analysis tag and just put number theory. Then it starts getting downvoted probably because it is in the wrong place. I think my tags were appropriate. I don't want to come here to fight :) $\endgroup$
    – user10290
    Commented Mar 23, 2017 at 0:47
  • $\begingroup$ Plus, I did ask a deeper question. I'll try math stack exchange for this one. $\endgroup$
    – user10290
    Commented Mar 23, 2017 at 0:52
  • 14
    $\begingroup$ For reference, here is the deleted question: mathoverflow.net/questions/265294/… . It got one downvote, but I don't see anyone being sarcastic in the comments. GHfromMO is very helpful in answering many questions in number theory, at all levels. I am sure that he only meant to be helpful in editing tags, and I would concur that number theory is the suitable tag. I would also be unsure if that question was suitable for this site --- the Wikipedia article would have been sufficient for it. $\endgroup$
    – Lucia
    Commented Mar 23, 2017 at 2:00
  • 6
    $\begingroup$ By the way, in regards to your deleted question, if $x$ is obtained by changing finitely many digits of $e$, then you don't need anything at all to prove that $x$ is trancendental. Namely, the difference $e-x$ is rational; if $x$ were an algebraic number, then $e = x+(e-x)$ would also be algebraic, a contradiction. This is implicit in with Gerhard Paseman said, but I thought it might be useful to make it explicit. $\endgroup$ Commented Mar 23, 2017 at 2:17
  • 33
    $\begingroup$ I also think that one cultural thing that made your question poorly received was the large picture of handwritten text. Using TeX is culturally expected here. $\endgroup$ Commented Mar 23, 2017 at 2:19
  • 3
    $\begingroup$ Thanks everyone, I really appreciate it. Both answers were great. $\endgroup$
    – user10290
    Commented Mar 24, 2017 at 22:58
  • 24
    $\begingroup$ Tags are a bookkeeping device of the site, which are used to help organize similar material, in part so people looking for answers to things (the entire point of the site) can more easily find them. They're not there for the asker, but everyone else. So while the suitability of some tags can be debatable for some questions, by and large questions should try to use them in accordance with a site standard. Though if you have specific concerns about the function, relevance, and use of certain tags, you can bring that up. It's in the site's best interest to have useful, populated tags. $\endgroup$ Commented Mar 25, 2017 at 7:49
  • 3
    $\begingroup$ I, too, appreciate this site - and all the Exchanges - quite a bit. For certain, StackExchange helped with not only my dissertation (which is my first attempt at literature-level writing) but also in passing my Complex and Real Analysis qualifier. That being said, I am of late intimidated asking questions, especially on MO, as I know that many rules - some of which I am still unaware - could mean my post is doomed to deletion and negative reactions for reasons I sometimes overlook. The result has been a bit of a trepidation to use MO - even MO Meta. But rules are rules, and I must follow. $\endgroup$ Commented Mar 26, 2017 at 0:44
  • 16
    $\begingroup$ I also have the feeling this is a cultural question. Unlike other sites (say, Quora), MO is not obviously dominated by American users and this exposes MOers to social behaviors they may be not used to. In my professional life I've often experienced that social interactions in European academia are much less based on appreciation and praise than in the USA. This may explain why behaviors that are perceived as "normal criticism" by European users are frustrating the OP. $\endgroup$ Commented Mar 28, 2017 at 5:58
  • 3
    $\begingroup$ Just happened to me again; a troll or ninja downvoted two of my answers to questions that haven't been active for about a week. I am not against downvoting my answers or posts, but not leaving any comment isn't nice. $\endgroup$ Commented Apr 4, 2017 at 8:51

4 Answers 4


While it is an ongoing concern to keep the site on-topic and to maintain its standards, it is crucial for the community how people feel when they are asking or answering questions here -- if people enjoy using MathOverflow, the site will flourish, and what happens otherwise is obvious.

Given this, I think it can only be welcomed when people say how they feel about using MathOverflow -- even more when the enquiry comes not just from a pseudonymous new user_xyz, but is made by a long-term user of the site under their real name -- and this does not depend on the merits of the particular question the enquiry is about.

Furthermore, what you write may well shed a light on the reasons for the unfortunate but obvious strong bias towards male mathematicians among the regular users of MathOverflow.

Since you are using the site basically already since its beginnings, and as you are a former PhD student of Joel David Hamkins, I assume you know the community well. In particular I don't think that there is any need to explain to you how things are on this site. Rather I think you could provide a valuable service to the community if you further elaborate (say, in a discussion post on Meta) on what you would change in order to make using the site a more pleasant experience for yourself -- and in the same time presumably for many other people as well!

  • 5
    $\begingroup$ I urge caution. I do not have the same faith in the strength of the community as you do. If this is not handled carefully, we might end up fracturing the community. It may be better to solicit community opinion in a preferably less divisive fashion. I hope we have input from the moderation team on this. Gerhard "Already Gave Up On Sci.math" Paseman, 2017.03.23. $\endgroup$ Commented Mar 23, 2017 at 18:49
  • 1
    $\begingroup$ Thanks so much. Please see the comment after the next answer - it is for this answerer as well. $\endgroup$
    – user10290
    Commented Mar 24, 2017 at 20:01

Erin, I'm sorry to hear you've found MO so negative in tone. There are a number of issues that have been brought up, both in the post and in comments, that might be useful to tease apart.

Let me start with re-tagging. I can see why you or anyone might resent others fiddling with their tags, and I do think it would be more collegial if others would (before eliminating a tag) ask first the reasons behind it. I'll qualify that slightly: if a noob comes in with a question about high-school level analytic geometry and applies the tag algebraic-geometry, then of course it means he doesn't know what that tag means to our community, and there doesn't seem to be a real need to discuss what he is thinking. But for someone with over 1000 points of reputation, and who holds a PhD in mathematics to boot, etc.: I'd say she should be asked first before removing a tag.

(Adding tags is probably less of an 'affront', although a polite comment such as "I took the liberty of adding such-and-such tag; hope you don't mind" would still be a nice collegial thing to do.)

That said, it's fine if someone who doesn't understand the point of a tag asks about it, and I don't agree tagging is (merely) subjective. Tags should be applied keeping in mind those users who filter by tag. Are those users likely to see the question as relevant and on-topic for that tag? If they don't, then they might feel their time has been wasted. (Let me put that more strongly: it is beyond dispute that many users are considerably annoyed by what they deem as improper tag choices -- why, there are two flags about precisely that in the moderator inbox, now as I speak.) So with that in mind, it should be part of one's routine to provide enough background so that hopefully all tags make sense to those concerned -- in particular, for a question ostensibly about transcendental number theory, some explanation for a tag like set-theory should be given.

This ties into the general issue of motivating and providing background for questions. I think a lot of users (not speaking particularly to you, Erin, but to people out there) might be a little uptight about this: MathOverflow seems like a hard-core or austere site, and people may feel they have to make their questions very concise and constrained (sometimes very formal as well), and they wind up not sharing why the question is actually important to them. So if for example

usually I am thinking of something more and I just want a reliable resource for my little question

then please consider instead sharing a bit of that "something more", if you can. If for example you intuit a connection between a question in transcendental number theory and issues related to set theory and forcing, then maybe say something about that if you can; people might find it interesting. (As an aside, Paul Cohen once said this: as a young student he was interested in number-theoretic identities like the Rogers-Ramanujan identity, and wanted to try a logical approach to them, by making an inductive analysis on the complexity of statements. "In a remarkable twist this crude idea was to resurface later in the method of "forcing" that I invented in my proof of the independence of the continuum hypothesis" -- More Mathematical People, page 50.)

In any event, I remember (now ex-)user quid pressing the importance of laying out background and motivation for MO questions, and I agree, and think it might help create more positive experiences at MO. The Help page also mentions this.

This enters into a more general question, which is sort of how I'd like the question of the OP recast, and which all of us should anyway ask ourselves:

How can I make sure my questions are positively received?

To begin with something simple: rendering formulas in TeX, as Andy Putman mentioned, is the norm and expectation in our Community. It's not a hard-and-fast rule by any means. But to state something obvious: if I take pains over the presentation of my question, giving it a professional appearance, then I transmit the idea that this question is important to me and I want others to take it seriously. Thus I agree it's a good idea to make use of the TeX software, routinely.

Relatedly, there is some expectation here that people, in addition to taking care over the presentation of the question, also exercise due diligence in researching the question a little beforehand (using Google, Wikipedia, etc.) -- and indications this has been done are usually well-received.

Speaking of this: if someone comments by giving a bare Wikipedia link, then in the first place let's grant the benefit of the doubt that he probably is trying to be helpful (and not snarky or sarcastic, unless he makes that plain). But in the second place it might mean he wonders why the answer wasn't found in Wikipedia, or why WP wasn't enough. Actually, not a few MO commentary discussions follow a pattern like this: a WP link is given, and the OP comes back and says, "Thanks. I did see that WP article, but I don't see how it quite answers my question, because...". And that would be a good response.

(On the other hand, I think it might help too, from the standpoint of netiquette, if commenters gave a little more than a bare Wikipedia link. There is every possibility that such comes across as a little curt or dismissive, or a slam dunk, or something like that -- even if you don't mean that. Imagine a friend asking the question; you might say something simple like "Might this be relevant? [link]" Or add a few words explaining the relevance.)

Erin also asks

I don't see what is the harm in keeping these questions up and just not saying anything unless you have something helpful to say

Well, the question of what we keep up runs into the perennial question of what the community considers 'on-topic' at MO, and as you know opinions vary widely. There will never be complete agreement about that. Looking at the number theory question in particular, I do think Mathematics StackExchange is the better choice. (And therefore the question should not be left up at MO -- one site per question, please.) Certainly it got a happier reception and result at Math.SE, and I'm glad you decided to post there. Something that worries me is that MO users may think Math.SE is a lesser site and are averse to posting there. I'd like them to reconsider. In theory there is usually a best SE site to post, and it's independent of where a user happens to feel most at home.

Looking at the beginning of your post, Erin: if it's really true that you feel put down by the responses every time you post a question, then I'm not sure what to say, except that (1) I'm concerned, and (2) I think you should consider making contact with the moderator team directly about this. Normally these things should be dealt with on a specific case-by-case basis, and not allowed to fester. Please also feel free to make use of the flag system if you think someone is being rude.

My greatest concern of all was touched upon in Stefan Kohl's answer: that your experience may have something to do with cultural attitudes and reactions towards women mathematicians. I take that possibility seriously. (We discussed this once at 'tea' [thanks to Carlo Beenakker for tracking this down], and Izabella Laba also has some thoughts on this at her blog.) As for myself, I can only imagine what it's like to be 'mansplained' at on a regular basis; let me close by saying I really hope this answer doesn't come across that way.

  • 25
    $\begingroup$ Hey everyone, I really appreciate all the comments and answers. I did sort of figure out what I did wrong early in the comments and decided to re-ask my question. It was actually very helpful for me to have this meta site to express my feelings. It helped a lot to feel supported. If you saw all my deleted questions you might see what I am referring to. I think I'm getting a thicker skin about it anyway, and learning that you all are open to communication. And that maybe everyone experiences the same scrutiny. $\endgroup$
    – user10290
    Commented Mar 24, 2017 at 19:57
  • 7
    $\begingroup$ I'm glad you posted at meta too, Erin. And, it's true that one can find a lot of negativity at MO. Most of it I think has to do with underdeveloped netiquette skills. Tricky business. $\endgroup$
    – Todd Trimble Mod
    Commented Mar 24, 2017 at 20:14
  • $\begingroup$ For reference, I guess this is the tea discussion you mentioned. $\endgroup$ Commented Mar 24, 2017 at 21:58
  • $\begingroup$ @CarloBeenakker Ah, great; thanks! I'll edit my post to reflect the information I found there. $\endgroup$
    – Todd Trimble Mod
    Commented Mar 24, 2017 at 22:05
  • 10
    $\begingroup$ @Erin Carmody : I personally answer questions under my own name, but mostly ask them under fake ones. I have thick skin, but I prefer not to have to deal with the occasional snarky comments I get when I ask questions about topics that are outside my comfort zone. $\endgroup$ Commented Mar 24, 2017 at 22:22
  • 4
    $\begingroup$ Thanks, I'm not upset anymore. It is very helpful to talk about. I'm sure a lot of it has to do with people not really knowing each other, so not knowing their level of understanding or misinterpreting words, without any other cues, over the internet. $\endgroup$
    – user10290
    Commented Mar 24, 2017 at 22:40
  • 9
    $\begingroup$ re: "How can I make sure my questions are positively received?" I'll take this opportunity to complain again that I think the awesome advice in this old post meta.mathoverflow.net/a/883 deserves to be placed in a much more prominent place. Compare the relatively generic info here mathoverflow.net/help/how-to-ask . $\endgroup$
    – j.c.
    Commented Mar 25, 2017 at 21:35
  • $\begingroup$ @j.c. Thanks for the reminder. If my memory is correct, that's what used to be under Help before the merger with SE. $\endgroup$
    – Todd Trimble Mod
    Commented Mar 25, 2017 at 22:39
  • 6
    $\begingroup$ Thank you Todd for a very thoughtfully composed answer, for acknowledging that gender bias probably plays an undesired role on this site, and for emphasizing that you believe the OP's description of her experience. Related note: one choice I have made is to use gender-neutral pronouns, almost all the time really, but certainly when I'm not referring to a specific person whose gender is known. I observe that in your answer, you twice use male pronouns to refer to hypothetical people. I'm not saying you must automatically make the same choices I do, but I wanted to at least suggest it to you. $\endgroup$ Commented Mar 26, 2017 at 20:10
  • 4
    $\begingroup$ @GregMartin Thanks. I was quite aware of my pronoun choices, and indeed they were deliberately made that way to mirror the actual circumstances of the MO post that was referred to in the discussion. (Please note the I also used 'she' in the second paragraph, again as part of the choice-design. The 'he' in the same paragraph is strictly speaking not part of that design, but refers to a 'noob' where I didn't think it would be seen as biased against women.) Hope that makes sense! $\endgroup$
    – Todd Trimble Mod
    Commented Mar 26, 2017 at 20:21

It looks like many of the specific issues raised by the OP have been empathetically and constructively addressed, which is good. But I wanted to add that some of the OP's concerns may be related to structural weaknesses in the stack exchange platform itself which emerge as a site scales up. There are three factors in play:

  1. Positive and negative feedback are asymmetric. There are many ways to provide feedback which can at least be perceived as critical: down vote, critical comments, modify the question, vote to close, and flag for moderators all come to mind. But the only way to provide positive feedback is to up vote (or leave "Nice question!" as a comment, which not too many people do).
  2. Privileges do not scale with the age or size of the community. More and more people get 2k rep, and all of them can edit questions and answers.
  3. Questions and answers with a high number of up votes are emphasized by the system and thus attract a lot of attention.

You can probably see how these three observations interact to make the experiences that the OP describes more common and more severe. Good questions attract more attention, more attention increases the expected number of people who see the question and don't like something about it, and those people are more likely than they used to be to have powerful ways to express their criticism.

Stack Overflow today embodies these dynamics taken to their extreme. A great many of the highest voted (and frankly most useful) questions on that site are closed as off topic, and the reason is that when thousands of highly empowered users encounter any conceivable question it is almost inevitable that five of them will eventually vote to close, and there is no offsetting "vote to keep open" option. This question on Stack Overflow meta could be a cautionary tale for Math Overflow, though I think we can all agree that things aren't anywhere near as bad here as it was even two years ago there. (They are not helped by the fact that people have professional incentives to gain reputation on Stack Overflow.)

I've learned to build all of this into my prior expectations when using the stack exchange network in general - when I post a question which gets edited or attracts some critical feedback I take it in part as a positive sign that the question was interesting enough to attract attention. And actually this is often in the spirit that the edits or critical remarks were intended: while I don't edit questions very often, when I do so it is usually because I think the question is interesting and I want to maximize the chances it will receive good answers. This works for me, but I'm not sure it is the basis for a good long term strategy to grow and maintain the community.

  • 2
    $\begingroup$ "people have professional incentives to gain reputation on Stack Overflow". Really? Can you tell me more about this? Point me to a link somewhere? $\endgroup$ Commented Apr 4, 2017 at 7:24
  • 2
    $\begingroup$ @GerryMyerson I don't know if this actually came to pass but the original plan for monetising StackOverflow was to use it as a recruitment platform for programmers. $\endgroup$ Commented Apr 4, 2017 at 19:29
  • 2
    $\begingroup$ @GerryMyerson Some links: itworld.com/article/2707360/it-management/…, softwareengineering.stackexchange.com/questions/20407/… $\endgroup$ Commented Apr 4, 2017 at 22:14
  • $\begingroup$ The tl;dr is that there is no formal process which transforms Stack Overflow reputation into a job or a higher salary, but some employers do consider it as a factor. To the best of my knowledge your Math Overflow reputation doesn't have much of a career impact for mathematicians, but I suppose I don't know for sure. $\endgroup$ Commented Apr 4, 2017 at 22:17
  • 2
    $\begingroup$ Successive Math Department chairs have made it clear to me that they consider MO participation to be a negative and not to be indulged in during business hours. $\endgroup$ Commented Apr 4, 2017 at 22:36
  • 13
    $\begingroup$ That is not at all surprising, though it doesn't really make sense to me. Asking and answering interesting questions with other mathematicians is a key part of doing research (one could argue that it is equivalent to doing research), and it is not clear to me why doing it in your colleague's office is "business" while doing it on a public internet forum is a waste of time. I can understand assigning little or no value to MO rep - it does not obviously correlate with anything - but assigning a negative value feels a bit antiquated. $\endgroup$ Commented Apr 5, 2017 at 6:53
  • 2
    $\begingroup$ I think that all the negative feedback mechanisms listed in this answer are good and useful, and MO would be worse off if any of them was missing (e.g., comments and edits help keep answers correct and readable, downvotes help remove useless answers out of sight, etc.). But I think it wouldn't hurt to have more positive feedback mechanisms! We could try uncoupling the "thanks, this looks readable and useful" meaning of an upvote from the "this answer has been checked and all details found correct" meaning (maybe by creating an extra kind of non-anon votes for the latter?). $\endgroup$ Commented Apr 6, 2017 at 17:00
  • $\begingroup$ @GerryMyerson Back in 2017, I agreed with you that such a department chair was wrong. But, lately, I'm starting to think the department chair might have been right. One could rationally argue that MO is contributing to the problem of underrepresentation in math, and supporting the site is de facto adding to the problem. That said, I remember a lot about issues with Macquarie department chairs from when I used to visit there, and generally wouldn't take the side of the dept chair from that era. $\endgroup$ Commented Jan 31 at 1:33
  • $\begingroup$ @David, believe me, the chairs' opposition to my efforts on MO had nothing to do with their having any concerns about the problem of underrepresentation in math. $\endgroup$ Commented Jan 31 at 3:28

I have found that the mathoverflow community has become slightly more strict with time, downvoting and closing posts more aggressively if they are not in accordance to the general rules.

I guess that's because over time the number of messages have increased significantly (1 every 5 mns) and around half of them are not really about "research level" mathematics. As a result, people are quick to downvote messages that are not perfectly phrased or which are not 100% clear, in order to spend more time on the interesting questions. This of course favors questions from well-established users over casual/beginner users.

So you can say that the community has become more "collé-monté", and there are a lot of early questions a bit vague, broad or of a philosophical nature, some of them gathering a lot of reputation, that would be discarded nowadays. I think that if you adhere strictly to the rules, you should see the negativity go away.

I didn't downvote any of your messages, and I hope you won't feel umconfortable if I make a comment on one of your questions entitled "Is there a Hotel California of set-theoretic geology?". The California Hotel reference can only cater to a specific geographic and demographic audience on mathoverflow. Keep in mind that the majority of the users here do not live in the US and unfortunately humor is a very cultural trait. Also I can't understand the body of the question, you don't give enough details. That's fine for me, don't modify anything, but that may explain the downvotes.

So yes, I agree, the community has become less friendly than in the early days.


You must log in to answer this question.