-4
$\begingroup$

Have others noticed a tendency for new users of mathSE to be treated poorly by existing users? The worst cases remind of some low-brow college fraternity where new users are expected to show they have earned their stripes, rather than being judged on the merit of their questions.

See, for instance, the complaint posted here yesterday: How can I quit this site? , complaining of exactly this behaviour.

I have encountered a surprisingly large number of interesting, or tricky, or research-level questions (i.e. where textbook solutions do not likely exist) posed by new users, ... and in return for asking an interesting question (rather than their homework), two things happen:

a) An existing user writes a comment quip like: "What have you tried?"

b) Then the gang frat behaviour follows, and the interesting question is closed down.

An example is this question:

The pdf of multiple independent gamma random variables

or this question (now removed):

PDF of the product of a standard Normal and a squared Normal

... still currently CLOSED.


UPDATE

Since this seems to have become a bit embarrassing to those involved, i.e. the inappropriate closing down of an original research question on mathSE, someone has now deleted the question, so that others here cannot see it and make up their own minds.

Here is a verbatim copy of the OP's clarified question:

Let $Z \sim N(0,1)$ and $X \sim N(\mu,\sigma^2)$ be independent random variables.

Is it possible to find a closed-form solution for the pdf of $Y=Z X^2$? I am also interested to find the characteristic function of $Y$.

Here is a $2^\text{nd}$ example:

distribution of Uniform to the power of Uniform

Here is a $3^\text{rd}$ example:

Distribution of X³ / Y when X and Y are uniforms distributions


While this might look like a simple question, for anyone who is familiar with the subject matter, it plainly is not simple, and I am not sure if a 'closed-form' solution even exists. Asking the OP what s/he has tried is frankly insulting and pointless, because it does nothing to progress the answer. That is a comment suited to someone posting their homework: and this is plainly not a homework question. What role is a moderator performing in deleting a question like this?

There are a number of concerns here:

  • The first is the closing down of genuinely interesting or difficult or research questions on this site. That is what I would hope is the real purpose of this site --- not just to see the same homework questions again and again and again

  • The second is the treatment of new users that are ultimately the mechanism for growth and interest

  • The third is that experienced users who are not familiar with a particular subject matter are closing down questions that they themselves could not answer, or do not realise are interesting questions ... and they are shutting down the question presumably simply because the person asking the question is 'new'. That does not seem to be a good reflection on the site either, nor a valid basis for closing interesting questions.


What have you tried?

There are a number of existing questions on meta about people commenting "What have you tried?" See, for instance: What did you try? and What have you tried? Is this homework? You will get more help if ... and more recently A criterion for distinguishing between two kinds of questions

I personally think answering "What have you tried?" is perfectly fine if the question is a homework question. My area of concern is when questions that appear to be simple, but are in fact complicated, get shut down, perhaps because they are being interpreted as homework questions, when they are not. In my opinion, the fact that someone is new is often used as an information signal to suggest that this is likely to be a low-quality question ... but that is not always the case.

$\endgroup$
59
  • 8
    $\begingroup$ For information: "Say I have X∼N(a,b) , a∼N(0,1) and Y=a(X^2) . What is the distribution and characteristic function of Y" is the version of the question that was asked originally and is also the version that got closed (up to formatting improvements made by somebody else). Further "returned to original - still requires clarification from OP" is what you wrote as edit message in a latter version. Given that you ought to agree with a closure. If the question now should be clear, post a request in the reopens request threads. $\endgroup$
    – quid Mod
    Jul 22, 2017 at 16:30
  • 5
    $\begingroup$ There are folks that disagree with you on the particular question to which you link. There is a dedicated meta thread for requests and justifications for the reopening of a question. Go there. Otherwise, you are ranting with only one sample, and drawing conclusions based on that post. In particular, your third point you make accusations about experienced users who close, including "they are shutting down the question presumably simply because the person asking the question is "new"" That's an accusation that just doesn't apply to most users. $\endgroup$
    – amWhy
    Jul 22, 2017 at 16:33
  • 5
    $\begingroup$ There is no way one user can take away any vote to re-open. Both quid and I are referring to the dedicated meta thread to deal with this issue on a case by case format: requests for reopen, undelete $\endgroup$
    – amWhy
    Jul 22, 2017 at 16:38
  • 4
    $\begingroup$ @wolfies I have not missed your "point". I simply don't fully agree with it. Because someone disagrees with you doesn't entail they must be missing your point. You write: Have others noticed a tendency for new users of mathSE to be treated poorly by existing users? ... I have encountered a surprisingly large number of interesting, or tricky, or research-level questions (i.e. where textbook solutions do not likely exist) posed by new users, Then you ought to be providing more links to the "surprisingly large number of interesting questions" which get closed. $\endgroup$
    – amWhy
    Jul 22, 2017 at 17:36
  • 5
    $\begingroup$ @amWhy Unfortunately, the previous examples that came to mind have now been deleted from the site, which is rather the point. $\endgroup$
    – wolfies
    Jul 22, 2017 at 17:45
  • 5
    $\begingroup$ To me, interesting includes 'has not been solved before' -- as distinct from 95% of the questions on the site which are textbook questions. $\endgroup$
    – wolfies
    Jul 22, 2017 at 17:54
  • 4
    $\begingroup$ UPDATE The post still exists, and many folks can still see it via your link. You are the one, who by this post, brought delete votes to it, not to hide it (such a post is never hidden), but because you directed folks to a poorly posted question of not much quality: no effort from the OP existed; the post was a plea for the asker for some answerer to tell them what the answer/result is. That's hardly a research level question. You undermine your concern with such an example. Don't lose your cool and make a fool of yourself via your very recent edit. $\endgroup$
    – amWhy
    Jul 22, 2017 at 18:17
  • 6
    $\begingroup$ Deletions are soft-deltions; users with the appropriate privilege (10k+ or mod) can see all deleted posts. It could be a good idea you get familiar with the basic mechanics of the site. $\endgroup$
    – quid Mod
    Jul 22, 2017 at 18:22
  • 5
    $\begingroup$ @amWhy I think you have described the problem perfectly - there are 5 experienced users of mathSE who voted to close down a perfectly valid original research level question. Say no more. $\endgroup$
    – wolfies
    Jul 22, 2017 at 18:40
  • 7
    $\begingroup$ If I cannot solve a calculus 1 integration problem, does that make it a research level question? Because I can tell you right now, I can barely integrate $e^x$. $\endgroup$
    – Asaf Karagila Mod
    Jul 22, 2017 at 18:49
  • 11
    $\begingroup$ "Research level"? There is nothing "research level" in the fact that, if $Y=ZX^2$ with $(X,Z)$ independent, then the most classical change of variable yields $f_{X,Y}(x,y)=x^{-2}f_X(x)f_Z(x^{-2}y)$, which in turn implies that $f_Y(y)=\int_\mathbb Rx^{-2}f_X(x)f_Z(x^{-2}y)dx$, as is explained in every decent textbook on the subject and has been explained countlessly many times on the site à propos various distributions and various transformations. Now, if one is actually alluding to the task of finding a formula for $f_Y$ in terms of usual functions when $f_X$ and $f_Z$ are normal, then ... $\endgroup$
    – Did
    Jul 22, 2017 at 20:08
  • 8
    $\begingroup$ ... one should note that 1. this is probably not doable, 2. this is not interesting. So much for "genuinely interesting or difficult or research questions"... And we should probably refrain from commenting how ridiculous the gratuitous slander "experienced users who are not familiar with a particular subject matter are closing down questions that they themselves could not answer, or do not realise are interesting questions" sounds when applied to question 2359678, right? $\endgroup$
    – Did
    Jul 22, 2017 at 20:08
  • 7
    $\begingroup$ @MichaelHardy "The closing of this question is an act of cowardice to a degree that is seldom seen." Why is it an act of cowardice? I didn't and wouldn't vtc it, but I can't see anything coward in doing it. $\endgroup$
    – quid Mod
    Jul 22, 2017 at 23:40
  • 6
    $\begingroup$ @MichaelHardy first votes to close this question were cast hours before you showed up. Chill out. $\endgroup$
    – quid Mod
    Jul 22, 2017 at 23:58
  • 6
    $\begingroup$ On your newest edit: did you bother to check who the unnamed user is that in a way was the final cause for quitting? The global situation on the site is a lot more complicated and multi-faceted than you might think. $\endgroup$
    – quid Mod
    Jul 23, 2017 at 12:40

2 Answers 2

16
$\begingroup$

My area of concern is when questions that appear to be simple, but are in fact complicated, get shut down, perhaps because they are being interpreted as homework questions, when they are not.

I don't* care whether it's homework or not. It's getting shut down since it's simply a problem statement assigned to MSE.

I don't care if the post is an overly terse description of an interesting question — the onus is still on the poster to show the basic research effort. In the case of a "research level question", it's even more important that the poster share the basic research with the reader.

And as a happy side effect, doing so will make the post stand out from the bad questions.


That said, I do think there is some poor treatment involved, but not the way you describe. In my opinion:

  • we don't effectively describe what a post needs to stop being a bad question
  • ad-hoc advice given in comments about how to improve the question is often even less effective at conveying this

So we have well-meaning people trying to help improve the question — and maybe the original poster even does their best to follow that advice — but frustration ensues because it's not done in a way that really effects much improvement (and may even make it worse) and the question gets closed anyways.

Disclaimer: the above description is skewed towards describing the more problematic interactions and may not reflect the average.


Note I neither affirm nor deny that the question under discussion is a research level question. Also, I'm not entirely convinced that the question the original poster was intended is the same question as the one wolfies conceived of after reading the post.

*: Actually, I do care, but have left that topic out of the discussion.

$\endgroup$
3
  • 9
    $\begingroup$ Downvotes on meta express disagreements hence one can understand some downvotes in the present case, as for every post which actually expresses an opinion, but a vote to delete? What can possibly be the point of a vote to delete this? $\endgroup$
    – Did
    Jul 22, 2017 at 20:15
  • 3
    $\begingroup$ This answer seems entirely based on factual error. Maybe I'll post an answer of my own.... $\endgroup$ Jul 22, 2017 at 22:32
  • 4
    $\begingroup$ Hmmm. I see no factual error here. (Though some folks do confuse fact with fiction.) $\endgroup$
    – amWhy
    Jul 23, 2017 at 21:44
4
$\begingroup$

Asking the OP what s/he has tried is frankly insulting and pointless, because it does nothing to progress the answer.

One utility of an OP's sketching of what he or she tried is giving Readers a starting point to explain a procedure to solve the problem. In the absence of this information, guesswork is involved in picking a starting point. Often this leads to an inefficient use of the OP's and the Reader's time. Either the Answer might cover material (i.e. definitions and other basic considerations) already known to the OP, or it might launch into a discussion that sails over the OP's head.

There are (as is often pointed out in Comments) other ways to add context besides sketching an approach. See the previous discussion here, How can this on-hold question be improved?

So "pointless" is not a fair accusation when someone leaves a Comment of the kind you describe. If you feel "frankly insulting" is the right interpretation, then so be it. You are free to place whatever interpretation on Comments that you think is constructive.

Note that you are putting yourself in the position of those who ask Questions that elicit Comments, when you haven't actually asked any Questions yourself. What you consider a constructive interpretation may change when you have to respond to a requested clarification. It can be gratifying when the OP takes a Comment as a sign of constructive interest in their problem and responds in a way that does "progress the answer." I made such a Comment today, which made the OP feel empowered to solve the problem themselves (and as a result, self-deleted their Question!).

$\endgroup$

You must log in to answer this question.

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