Not signed in (Sign In)

Vanilla 1.1.9 is a product of Lussumo. More Information: Documentation, Community Support.


    This question has been given a bounty. Among other things, this means that one can't vote-to-close it. I consider the bounty to be wholly inappropriate; at the very least, the question ought to be CW (indeed I was pondering a vote-to-close and the bounty had actually decided me when I found that it made the ponderance irrelevant!).

    Link to question:

    • CommentAuthorHarry Gindi
    • CommentTimeJun 7th 2010 edited

    I agree. I would have voted to close it as well.

    • CommentAuthorGS
    • CommentTimeJun 7th 2010
    In at most a week the bounty will have been awarded.

    I have to admit that I don't understand the eagerness to close questions that aren't

    (a) Obviously homework,
    (b) obviously inappropriate for mathoverflow, or
    (c) offensive.

    The point has been made many times that it's hard to see that it does much harm to have questions like this hanging around.

    On the other hand, the real-world reputation of mathoverflow suffers from the perceived unfriendliness of closing and/or downvoting well-meaning questions.

    At the moment, acquiring a reputation for nitpicky, annoying hall-monitorishness seems like the greater threat to MO. The primary concern here is actually *not* the for the feelings of the OP, but rather the undetected observers: for me (and likely many others like me) the main limitation of the usefulness of MO is that it is still not populated by a sufficiently diverse array of experts. Don't keep (or drive) those people away!
    • CommentAuthorHarry Gindi
    • CommentTimeJun 7th 2010 edited

    It's not so much hall-monitorishness. We actually hate to see bad questions left up and we use "the rules" as justification. As I (and others) have noted in the past, there's plenty of wiggle room if the question is legitimately good and interesting.

    For the record, I know a number of experts who don't post on MO because they came here on a bad day when things like "math jokes" and other big lists were at the top of the page. I try to close these questions (or at least the crappier ones) when possible to prevent that from happening

    Stephen: I agree with your sentiments. But I think the prevailing opinion (or at least, concern) on meta is that "bad questions" frequently showing up near the top is also likely to keep away the more diverse array of experts that we all hope to see on MO. I don't know if we have any real information about which perceived problem (bad questions v. hall-monitorishness) is more of a problem in practice.
    • CommentAuthorVP
    • CommentTimeJun 7th 2010
    One thing worth pointing out is that this question had an erratic history: I was considering whether to post an answer when the author decided to accept one of the two answers already given. This was about 3 hours after posing it and with very low views. I was a bit annoyed, but said, "Oh well". I guess he unaccepted the answer later and put a bounty on it? All this is very weird.
    • CommentAuthorJeremy
    • CommentTimeJun 8th 2010
    @Harry: I think that's a bit of an odd assertion. For example, Gowers and Terry Tao are both certainly experts, but have made plenty of contributions to these kinds of soft questions! So I don't think it's a good argument to make that we should close them because they repel some people when it could be argued they attract others.

    A lot of the good answers in these questions can be more insightful than answers to more specific 'quantatative' questions. Even if the initial question wasn't so great, from the point of view of someone who's just here to learn interesting stuff, that can still be useful for me.

    However, if someone were to propose separating the "math" questions from the "soft" questions by, e.g., putting them on different pages through the tabs at the top, I would have no problem with that.

    A lot of the good answers in these questions can be more insightful than answers to more specific 'quantatative' questions. Even if the initial question wasn't so great, from the point of view of someone who's just here to learn interesting stuff, that can still be useful for me.

    That's not the point of MO. If it were, I doubt that MO would be nearly as active today. Soft questions are a side-show that we allow as long as they're not causing a problem. To be honest, I could do without them entirely, and I'd still visit the site.

    I was recently at Dick Gross's birthday conference, and talked with a bunch of mathematicians who said that they don't visit the site as often as they used to because of the large number of soft questions. While Gowers and Tao have answered soft questions, I think they are outliers. Both of them have public blogs that have some soft-content essays, so we can consider them to be more enthusiastic about public exposition than most. I'd like to push more toward a research-oriented focus, because there are people "on the fence" who can contribute positively, but don't want to bother wading through ill-formed questions.
    Hopefully sometime soon there will be a lower-level math stack-exchange style site and we can try to push things there. I feel like we've picked up a large number of users who aren't research mathematicians, and as a result soft questions and questions that are too easy are getting too many upvotes which encourages people to continue asking them. Having another page to send people to will make it easier to do something about this. It'd be nice if there were a tool other than voting (which is swamped by the larger number of weaker lurkers) and closing (which annoys people more than it actually accomplishes anything) to deal with questions that we'd like to discourage.
    @Scott: I really don't understand, what stops those people from opening up the site and simply ignoring all the soft questions? I don't think there was ever a point in the history of MO where more than half of questions on the front page were soft. Do they perhaps consider all questions outside their area of interest soft?

    @Noah & Scott: I don't really think it's such a good idea to send people who already visit this site away in the hopes of attracting more people (certainly until there is a viable alternative). These people already like the site for something, who knows if the people who you wish to attract will ever like it whatever you do. It's easy to scare people away, it's hard to attract people.

    As I've said before, I think there is a definite advantage to having weaker people around. They'll answer simpler questions, they'll learn and become stronger.
    • CommentAuthorHarry Gindi
    • CommentTimeJun 8th 2010 edited

    @Ilya: Soft questions add to the "irritation factor".

    I also find that people who are at a "lower level" as you say tend not to have their hearts in it as much (not to say this is true of everyone at a lower level).

    @Ilya: I did not interrogate these people in detail about their browsing habits, because I think it is beside the point. They have skill sets that enable them to answer questions that few in the world can tackle, and for that reason, I view them as more valuable to the MO community than the people who may be driven away by an increased focus on research-level questions. If lots of professional mathematicians feel their time is being wasted by the presence of soft and easy questions, then I think we should cut down on soft and easy questions.

    @Scott C: +1 again!

    • CommentAuthorVP
    • CommentTimeJun 8th 2010
    @Harry: The irritation factor from discussions like this one and several more in the past few days, with lots of extreme views expressed, is x10 more irritating than anything I can possibly see on the front page of MO. Frankly, there are many poorly worded or thought through questions that nobody cares about. So my overall feeling is that a small group of people is trying to impose their idealistic idea of what MO should be. That's a big concern especially since the system for voting/closing questions is so non-transparent and undemocratic.
    • CommentAuthorHarry Gindi
    • CommentTimeJun 8th 2010 edited

    The irritation factor from discussions like this one and several more in the past few days, with lots of extreme views expressed, is x10 more irritating than anything I can possibly see on the front page of MO. Frankly, there are many poorly worded or thought through questions that nobody cares about.

    And Scott C. has given evidence to the contrary.

    This discussion is taking place on meta out of view from the front page. Your argument doesn't hold water.

    That's a big concern especially since the system for voting/closing questions is so non-transparent and undemocratic.

    I somewhat agree with you that votes down should at least have some sort of comment left, but votes to close are transparent and democratic. It takes five votes to close, and the people who successfully close a topic have their names listed.

    • CommentAuthorVP
    • CommentTimeJun 8th 2010
    @Harry: ALL WRONG

    1. Sorry, that's not evidence, that's anecdotal evidence - just the kind that you object to in answers to some questions. FWIW, I have evidence that some people don't visit MO because MO'ers can't answer their research level questions, but I wouldn't make any claim about the prevalence of the phenomenon.

    2. To be precise: Scott wasn't comparing irritation factors from two things I mentioned, was he?

    3. Big part of the discussion is taking place right where the questions are, with links posted to meta that anyone can follow (and many people do).

    4. I didn't make an argument: I expressed my opinion and concern.

    5. I didn't say anything about downvotes on questions. Sorry if I wasn't clear: I meant that there some voting taking place before a question is closed.

    6. If you think that seeing 5 names of people who made a decision, without any clue of how and why it was made, after the fact, is transparent, you don't understand the meaning of transparency. And I am still laughing my head off about "democratic"...

    7. You have just added to the irritation factor by your dismissive reply.
    • CommentAuthorHarry Gindi
    • CommentTimeJun 9th 2010 edited


    1. Anecdotal evidence is evidence. What matters is how much we weight it. Sorry, try again. As for your anecdotal evidence, it would mean a lot more coming from someone who identified him/herself. I'm sorry, but based on Scott C's track record, I place quite a bit more weight in his opinions.

    2. I don't answer rhetorical questions.

    3. These questions are clearly in violation of MO policy.

    4. Fine, you offered your "perspective".

    5. You said voting/closing, not voting to close.

    6. Unlike you, the people who usually vote to close have their real names available and are easy to contact by e-mail if the person was sufficiently interested.

    7. Irritation factor and your personal level of irritation are quite different things....

    I think that it is up to people with high rep what ought and oughtn't to be closed. My e-mai is available through my userpage, and if anyone has questions about why I voted to close a certain topic, it's very easy to write me by e-mail.

    • CommentAuthorGS
    • CommentTimeJun 9th 2010 edited
    Hi Scott,

    Your explanation notwithstanding, I agree with Ilya and VP that it seems a bit silly for someone to avoid MO because of the presence of questions they feel are beneath them. There are plenty of easy ways for someone who wants only to consider technical questions to avoid most of the "soft" ones. Example: bookmark the unanswered page and don't visit the rest of the site. Or: never click through to a soft question (I guess you may be fooled occasionally). People for whom this type of sifting is impossible won't be able to participate in *any* online forum.

    On the other hand, not participating because of worry over seeming stupid in public, or because of worry that your vague but serious question will not be well received seems completely reasonable to me. And there remains a troubling (to me at least) empirical fact: as compared to the percentage of female mathematicians, female participation in MO is shamefully low---for instance, on the 1st five pages of users sorted by rep (as of right now, the best objective measure of time+effort invested in MO), I'm not sure that there is a single woman. Do you believe this is because of the presence of soft questions?


    PS- I mention the lack of women on MO not to stir controversy, but because it's the only hard data point I can think of right now. If anyone can think of another non-anecdotal piece of data, that would be great!

    I'm sure you may realize that female participation in the overall mathematical community is also quite low, although I'm not sure of the statistics. If you were hinting that women will tend to ask softer questions, I think that's a rather condescending thing to say. The majority of women I've met in the mathematical community have been extremely serious about research, and perhaps the reason there are so few active women at MO is due to the level of off-topic and soft-question types of discussion. If that's not what you were implying, then I apologize in advance, but the above viewpoint is not uncommon in the scientific community at large.


    I'd just like to copy something over from another thread as it has relevance for this discussion. In replying to Kevin Buzzard's desire to reward a speedy answer, I realised something about how I use MO:

    I want to be helpful to the "mathematical community" beyond my own immediate surroundings (both physically and thematically). But that's quite hard to do in an effective manner. The infrastructure of MO helps me focus my efforts so that I can maximise the help that I can give. But to be helpful, I need to be able to find easily the questions where I can help. An excess of vague questions makes this very difficult and so would, ultimately, lead to me looking for other ways (i.e., not on MO) to contribute.

    Since we're at the level of anecdotal evidence (not having any other kind), I think that an excess of vague questions would lead to me effectively leaving MO ("effectively" meaning that I wouldn't wake up one day and decide never to darken MOs doors again, but just that over time, I would drop by less and less until MO simply dropped off my radar. As evidence that this would actually happen, the time between my visits to mathematical blogs has been getting longer and longer.).

    • CommentAuthorGS
    • CommentTimeJun 9th 2010
    Dear Harry,

    In recent years, the proportion of math Ph.D's in the US awarded to women was approximately 1/4. I am sure you will agree that this is orders of magnitude higher than the proportion of female participation on MO. I did *not* speculate as to the reasons for this disparity---you may have noticed that the paragraph ended with a question, which was not intended rhetorically. It is of course possible that women are avoiding MO because of soft questions on the front page (though as mentioned above I find it surprising that this would be anyone's real reason---my apologies in advance, Andrew!).


    PS- I don't spend any of my time reading questions like "Best math jokes". That's how I know they're easy to avoid.

    Thanks for clarifying. =)

    • CommentAuthorGS
    • CommentTimeJun 9th 2010
    Dear Andrew,

    Perhaps you can be more specific:

    What proportion of the questions would need to be vague in order that it be difficult for you to find the ones with which you can help?



    I can't answer that because it wouldn't be a conscious decision. And I don't particularly want to answer that. One thing I like about the fact that we vote to close (and reopen) questions is that I don't have to be right! I can vote according to what I think, safe in the knowledge that mine is just one opinion and if no-one else agrees with me then the question will stay open or closed.

    Let me say one thing in answer to your PS to Harry. I agree that it is easy to avoid "Best math jokes". But it gets hard when every question is of that type, and if MO is seen as the place to ask such questions then that is what will happen. The numbers are against serious questions: there are far more "vague" questions than "focussed" ones. Also, the danger is less from the obviously vague questions and more from the ones that are just hard to see what's going on. These are the real time-wasters as it's not always clear when one starts reading it whether or not it's answerable. And on my principle of "I want to be helpful", not reading these questions is not an option.

    @Stephen: I have talked with many women about MO, and the reasons they give for not participating much are essentially the same as the reasons I get from men. Here they are, in roughly decreasing frequency:

    1) I don't want to get too involved, because it looks really addictive.

    2) I'm too busy / I have too much on my plate / I don't have tenure.

    3) I actually read [present tense] it a lot, but whenever I see a question I can answer, someone has already produced a ridiculously thorough response, and I have nothing meaningful to add.

    4) If I have a question, I'd rather ask my collaborators/colleagues than broadcast my ignorance on the internet.

    4.5) When I have a question, it is rarely well-formed, and I need to talk to a person in real-time to refine it.

    5) [only heard at the Gross conference] Emerton and BCnrd are an intimidating presence, since they can answer any question I have a hope of answering.

    I agree that the low rate of active participation of women is troubling, but I can't say with any confidence that it has much to do with the frequency of soft questions in either direction. Naturally, it is possible that they are being turned off by the manner of questions being closed, and don't feel inclined to say so to one of the chief culprits.

    In regard to your question for Andrew, I don't have a quantitative measure for vagueness tolerance, and I think asking for a number is perhaps a bit demanding. However, my preference is for this to be as small as reasonably possible. For example, I would personally find a front page full of well-formed mathematics questions much more appealing than one that is half-full of of such questions with the rest taken up by noise, even if I could easily block out the noise, because it suggests that the noise is wanted.
    It's worth noting that female mathematicians also seem to be disproportionately underrepresented in the mathematical blogosphere, etc. ("etc." here is meant to include things like the n-Lab, etc.). So whatever contributes to women's non-participation in MO may be part of a broader pattern.

    The only woman I've talked to about MO is my wife (who is a better mathematician than I). What she has to say about both MO and math blogs, etc., is that she simply has no interest in them. I'm not suggesting that she is representative of anyone other than herself, but I bet many other people similarly avoid the entire "doing and discussing math research on the web" enterprise without particularly distinguishing MO from any other part.

    I meant to come to some kind of conclusion here about what this should imply for MO, but now I realize I'm not sure what it should be.
    • CommentAuthorInformer
    • CommentTimeJun 9th 2010


    @Harry: The URL translates to "math girls", and it seems to be a math blog rather than spam. I think the link was posted in response to Mark's claim that women are underrepresented in the math blogosphere, but it is difficult to read Informer's intentions without additional context.
    • CommentAuthorVP
    • CommentTimeJun 9th 2010
    From the looks of it, that is a fantastic blog about mathematics by female mathematicians and mathematics students.

    Well that's what informer should have informed us on (excuse the pun), not all of us are fluent in dutch. Secondly, what's the point of linking to it? Without any context or explanation it basically is as Harry said spam, even though the link is to a good blog.


    From the blog's Who are we? section:

    Mathematics girls Ionica Smeets & Jeanine Daems. They met as doctoral students in mathematics at the University of Leiden. During the lunch break they told each other stories enthusiastic about mathematics. Since March 14, 2006 they put all the cool things they find on this blog. So everybody can enjoy all the fun of math facts they come!

    It's a pretty nice blog. Thanks for informing us!

    [Translation by Google. Sorry for messing up their beautiful language.]

    • CommentAuthorJeremy
    • CommentTimeJun 9th 2010 edited
    @Scott: I think these are great points, I actually think #1 and 2 are the biggest ones for lack of participation. There's really *no reason* for a lot of people to contribute here other than wanting to generally contribute to math... But I (and I'm sure you!) have certainly known plenty of people who are unhappy enough to do this via teaching and would be difficult to compel to contribute here regardless of our standards.

    I think it's interesting that you mention #4/4.5 because most of my discussions with friends and colleagues that end up producing interesting **research-level** ideas are precisely the ill-formed ones that would make inappropriate MO questions by the standards of some here.

    The well-defined exact questions are not really "research level" but are more "understanding what isn't explained in this text book level." Which is good, and still interesting, but I find it odd that there is the idea that we can somehow simultaneously require "research level" and "well-defined" questions.

    Not that I think we should allow *excessively* vague or obviously meaningless questions. But a number of really good soft questions out there have had people complaining that they should be closed in the comments, and I think this is unfortunate.

    Particularly because they are often non-constructive complaints. Asking someone to "make their question more specific," e.g., is not so helpful, because it's entirely possible that they don't know enough to, or the question was purposefully vague so as to not mislead people by technical details unimportant to what they were thinking about. (This is of course giving the benefit of the doubt that they're trying to ask a legitimate question!)

    *Specific* comments like "are you sure you don't mean X", or "did you know this property is called Y", or "are you familiar with Z", or "are you really trying to ask W" etc are much more constructive. (I would point out specific examples to illustrate this more clearly, but I don't want to single anyone out.)

    It does make me more than a little irritated to see some of the non-constructive comments some people like to make. I mean, I'm all for making a vague question into a more exact one, but trying to burn every comment that doesn't have an immediate textbook answer to the ground is going a little too far.

    Again, this is assuming people are asking questions in good-faith; clearly, cases where people are asking vague questions and refusing to clarify or participate, or are asking obvious nonsense are different.

    And we have to remember, too, asking good, clear questions is not really a trivial task for lots of people. I've met some very smart people who could not form a well-posed question the first time to save their lives!

    This is aside from the other kinds of "soft" questions that are inherently on the far end of softness, that these complaints don't so much apply to. But these very soft questions make up an even smaller fraction of the total, so I really can't imagine how anyone can think they are a serious problem.

    Also, it's not mutually exclusive to have more "research level" questions and allow "soft" ones. We can certainly have *more* exact, well-posed "research-level" questions without going after everything that isn't one. We just need to (and, I think, **should**) encourage more people to ask these kinds of questions, but this can be done completely independently of discouraging soft ones.
    @Jeremy: I agree that many research-level ideas germinate from ill-formed questions. I also agree that asking clear questions is a skill that requires practice to master. However, I don't think MO is a good place for wrestling ill-formed questions into a well-defined form, even if they may eventually yield interesting research. That kind of shaping is hard work, and I think it is unreasonable to put the onus on the people answering the questions. I think MO functions more smoothly when the question comes with the expectation that there is an expert in the subject who knows the answer more or less immediately. This keeps people from coming away from their interactions feeling that they have been wasting their time.

    I agree that comments could be more constructive sometimes. However, if someone does not appear to have put any effort into writing a good question, that person isn't helping to establish much of a feeling of reciprocity.

    I strongly disagree with your assertion that the well-defined exact questions are not really "research-level". MathOverflow already has thousands of well-defined, research-level questions! How could you make that claim and expect us to take your arguments seriously? At the very least, it suggests that you are working with a definition of the word "research" that is at odds with the way everyone else uses it.

    (Last paragraph edited slightly for tact.)

    I strongly agree with almost everything Scott and Andrew have said in this thread. With regards to the argument "it's not so hard to ignore what you don't like", I want to completely reproduce a post a made a month ago, but I'll just post the "conclusion" to encourage people to go read the whole thing:

    If a "good" question is going to encourage people to come to the site for the wrong reasons (or discourage people in the mood to do some real math from coming to the site), then it will make it harder for the right people to connect. In those situations, I will argue that "MO is not the right website for this" and campaign for closing the question.

    I believe that MO will be maximally useful if its purpose is crystal clear in everybody's mind. It is basically meant to be used to prevent you from getting stuck on "baby steps" in research (where "reasearch" is broadly interpreted to include learning stuff you might use in research). As Scott said, "MO functions more smoothly when the question comes with the expectation that there is an expert in the subject who knows the answer more or less immediately. This keeps people from coming away from their interactions feeling that they have been wasting their time." Reason (4.5) is a very good reason for not coming to MO yet.

    A common response is, "why not make another site for fuzzier or ill-formed questions?" I wouldn't know how to do that, and I have serious doubts that it's even possible. Making such things productive requires a huge amount of feedback which I don't think the internet can provide. Sometimes, even talking with someone in person can't provide enough feedback ... a lot of the process has to happen in your own head. Also, it's usually hard or impossible to make progress in a crowd rather than in a small group. Whether you use the internet or not, I don't think it will ever be very productive broadly announce something like, "Hey, I keep getting this feeling that almost everything is made of triangles. Is there some general principle here?" and then sit back and wait for answers.


    "Hey, I keep getting this feeling that almost everything is made of triangles. Is there some general principle here?"

    This made me chuckle.

    I strongly agree with almost everything Scott and Andrew have said in this thread.


    • CommentAuthorGS
    • CommentTimeJun 11th 2010
    Andrew, Anton, Mark, Scott

    I think we all agree on one goal: maximize the number of good (this adjective requires clarification) questions and answers available on MO. I also think that our opinions about *which* questions and answers are good are not so far apart as you might believe, and that the direction in which we differ is not what you might suppose from this interaction. In fact, I think most questions on MO are either not specific enough or else too close to graduate-level homework problems.

    Where we differ is just in the method of encouraging good behavior---in a nutshell, I'd like to see good behavior rewarded and bad behavior ignored (when reasonable) or corrected as gently as possible. There is an apt British saying: don't get into a fight with a chimney sweep.

    But you are all much more experienced with MO than I am. I hope it continues to grow!