37
$\begingroup$

I think it would be useful to start a series of posts on the weaknesses of Mathematics SE for diagnosis precedes remedy.

In my humble opinion, one major weakness of Mathematics SE is that many users — including many high-reputation ones — simply do not quite know where some difficulties lie. I have worked with former mathematicians in industry and talking to them about mathematics is almost gauche, as once a real-world (non-mathematical) problem has been translated to a mathematical problem, with much exaggeration, it is either trivial or intractable. The difficulties lie in acquiring the domain knowledge required to translate the real-world problem to a mathematical problem in an artful and tasteful manner.

On Mathematics SE, one sometimes reads potentially interesting embryonic questions written by people without much mathematical training, say, on some problem from biology or engineering. If I perceive that the OP is asking the question in good faith, I often try to push the OP towards improvement of the question by asking several questions in the comment section and then editing the question in order to improve it via incorporation of the information provided by the OP in the comment section. Unfortunately, these questions are often closed — say, due to lack of clarity — before the seedling question has sprouted into a budding question that is ready to bloom. Why? An entire class of questions is being "uprooted" before one even has a chance to assess whether it's a flower or a weed. Why not be a bit more patient?

In my humble opinion, users of Mathematics SE should consider reading Bernard Beauzamy's Real life mathematics (2001).

Related:

$\endgroup$
33
  • 7
    $\begingroup$ I appreciate your intentions, but do close some such questions. My thought process is as follows (1) People can continue to converse in the comments and (like I said in the previous question) copy useful comments of OP into the question, each time sending the question into the reopen queue. I assume that people will not shun conversation when the question is closed. (2) Not everyone wants to read developing questions, users want to be reading ready-made ones as well. Answering those properly respects the users that take the time to formulate well written questions. $\endgroup$
    – Sarvesh Ravichandran Iyer
    Apr 8 at 11:09
  • 2
    $\begingroup$ @SarveshRavichandranIyer I don't disagree. I am fine with closing embryonic questions provided that they are opened once they have reached a certain level of maturity. $\endgroup$
    – Rodrigo de Azevedo
    Apr 8 at 11:17
  • 21
    $\begingroup$ @RodrigodeAzevedo ... I see the point. The process goes: close the question, work on it until it is satisfactory, then re-open. An impediment to this approach is the phenomenon of over-eager deleters, who propose deletion of a question too soon. $\endgroup$
    – GEdgar
    Apr 8 at 11:32
  • 8
    $\begingroup$ @lulu, if a question is badly written, it may be impossible for us to figure out what repairs to make; it may be impossible for us to know what OP was trying to ask. Often only OP can tell us that. $\endgroup$
    – Gerry Myerson
    Apr 8 at 13:09
  • 7
    $\begingroup$ I'm thinking of an example from years back...the details aren't important. The very terse question was something about counting words made from a fixed alphabet. But as people pointed out the routine answers, the OP kept remembering rules that needed to be satisfied. The question was hopeless, as the rules kept changing...but there was a core idea. There were some flat laws, implied by Chemistry, and there were some statistical laws observed in nature. I can see some very strong questions along these lines, though the actual post wasn't very useful. $\endgroup$
    – lulu
    Apr 8 at 14:03
  • 18
    $\begingroup$ @GEdgar I wish that model worked better in practice. But closed questions are rarely reopened, whether or not they are improved, and closure is a death sentence for the post. I often wonder if it would work better if the original close voters received a notification when the post was edited, inviting them to take another look. $\endgroup$
    – MJD
    Apr 8 at 14:54
  • 4
    $\begingroup$ @MJD That is an interesting suggestion. I posted a question on Meta about it. $\endgroup$
    – Rodrigo de Azevedo
    Apr 8 at 15:48
  • 9
    $\begingroup$ I think most would agree that patiently guiding an OP through the process of improving their question is better than closing it. It's just a question of scale. With hundreds of questions a day, some will have to be closed without further guidance than the closure message and linked FAQs. That doesn't prevent you from giving individual attention when you do see potential. $\endgroup$
    – Alexander Gruber Mod
    Apr 9 at 2:28
  • 14
    $\begingroup$ It does actually prevent me from doing that, because questions are sometimes closed and deleted while the process is incomplete. Then my time is wasted. And because I don't want to waste my time that way, I am more reluctant to begin the process than I wouldbe if every my interactions weren't under threat of being cut short. $\endgroup$
    – MJD
    Apr 9 at 3:31
  • 3
    $\begingroup$ @JonathanZsupportsMonicaC That is not so easy, as, in my experience, most questions of this class are either abandoned by the OP (which is not Mathematics SE's fault) or closed / deleted before a satisfactory state of maturity is reached. Often, a satisfactory state of maturity is arguably unreachable. $\endgroup$
    – Rodrigo de Azevedo
    Apr 10 at 17:30
  • 2
    $\begingroup$ Well, prompted by your post here, I noticed this question. It seems like one where a non-mathematician is asking for help in "mathematizing" their question, which is how I'm understanding your post here. Maybe start saving off questions that exemplify the problem? These days you can create a list under your "Saves" specifically for just such questions. And I honestly do think specific examples make policy discussions better. $\endgroup$
    – JonathanZ
    Apr 10 at 17:39
  • 2
    $\begingroup$ I agree with @JonathanZsupportsMonicaC's point that evidence would be useful here, and possibly more use of the applications tag would help? This would allow the site to track these kinds of questions, and may solve the underlying problem by encourage users to act differently towards them. $\endgroup$
    – user1729
    Apr 11 at 8:49
  • 8
    $\begingroup$ I agree with user1729 that more focus on applications of mathematics could help. I am an economist, I answered here on MathSE to a question about mathematics in economics, the question was hastlily closed for 'lack of clarity', even if my quite long answer should have suggested that the question was clear to an economist. A bit of patience would have been required. $\endgroup$
    – BakerStreet
    Apr 11 at 13:53
  • 6
    $\begingroup$ For a fresh example, everybody who upvoted this question here would have likely voted to reopen this, which the OP updated today, after 4+ months. No one did, though, which only reinforces @MJD's point that "closure is a death sentence for the post". This was clearly not homework after all this time, and just as clearly it is an applications question from someone seeking math help where they thought they would find qualified math help, only to be rebuffed without cause. $\endgroup$
    – dxiv
    Apr 13 at 6:46
  • 5
    $\begingroup$ @MJD I actually don't think that the core issue is that close voters aren't being aware of closed posts. Only my experience but I've found that there are groups whose view on what's right and what's not are miles apart. Making reopening and closing easier doesn't make things easier when there isn't consensus at the bottom. It will probably lead to more close-open wars. Hence the comments on "inconsistency" of moderation as well. However, I do agree that forcing moderation will make users interact more. After many interactions, I have become far more "moderate" than I was once. $\endgroup$
    – Sarvesh Ravichandran Iyer
    Apr 13 at 11:12

1 Answer 1

Reset to default
13
$\begingroup$

Suggestion: You/we can make sure that such questions are tagged with the applications tag, as suggested by @user1729.

Additionally, we can come up with some boilerplate comment, like

Community: Note that this is a question about an application of math, which is permitted. Please do not vote to close or delete just because it is not purely a math question. Feel free to use comments and questions to the OP to better define the mathematics of it, if you are so inclined.

Or for easier copy-pasting:

Community: Note that this is a question about an 
[application](https://math.stackexchange.com/questions/tagged/applications) 
of math, which is permitted. Please do not vote to close or delete just 
because it is not purely a math question. Feel free to use comments and 
questions to the OP to better define the mathematics of it, if you are 
so inclined.

Rodrigo, would you be interested in "protecting" questions like these with such a comment?

$\endgroup$
5
  • 5
    $\begingroup$ I like this proposal, but I doubt that such a comment will have a noticeable effect on people voting to close or delete. $\endgroup$
    – MaoWao
    Apr 12 at 9:51
  • $\begingroup$ This is potentially interesting. Alternatively, a link to this answer in a header would also be interesting. $\endgroup$
    – Rodrigo de Azevedo
    Apr 12 at 16:21
  • $\begingroup$ @RodrigodeAzevedo - I'm not sure what you mean by "in a header". $\endgroup$
    – JonathanZ
    Apr 13 at 0:15
  • $\begingroup$ @JonathanZsupportsMonicaC This question has a header, for example. $\endgroup$
    – Rodrigo de Azevedo
    Apr 13 at 7:18
  • 4
    $\begingroup$ I have not seen questions of this kind flooding the site in any way, and if any of them improve and become good questions then they're way, way more different and unique compared to the humdrum served up in general. Definitely, if someone left such a comment as above in good faith, that question shouldn't be closed. $\endgroup$
    – Sarvesh Ravichandran Iyer
    Apr 13 at 19:56

You must log in to answer this question.

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