Motivated by:

Curve resemblance to Closest Shape - Question on Hold - Advice for clarification needed

The question has two parts - "is there a way to define closeness of geometric objects?" - which is the kind of question that mathematicians ask and answer all the time - and "using that definition, how do we find the closest shape?".

There are many cases of this and it's a contentious issue, these questions often get closed and I don't think it is crazy to do so because what they are literally asking is unclear. However implicitly they are asking "how do I make this question precise, and what is the answer to that precise question?".

How should we deal with such questions?

  • 2
    $\begingroup$ Tag it with the (soft-question) tag? $\endgroup$ – Theo Bendit Apr 3 '18 at 4:43
  • $\begingroup$ The number of my postings to m.s.e. is so large that in some circumstances it seems appropriate that I would take some pride in it, althought that's something I don't normally think about. But the way m.s.e. has treated this question makes me want to say m.s.e. is not good enough for such a question. Partly that may be a manifestation of the bigoted contempt for applications outside of mathematics that has long afflicted the mathematical community. I think that particular preposterous lunacy has begun to clear up and may change substantionally in the coming decades. Maybe$\,\ldots\qquad$ $\endgroup$ – Michael Hardy Apr 11 '18 at 14:52
  • 1
    $\begingroup$ $\ldots\,$history will record that applied mathematics is a field of endeavor that began in about the year 2100. Perhaps the part of m.s.e. that is truly not good enough for this is the process by which closure and deletion is done. It is run in such a way that participaiton is intolerable when one is sober. $\endgroup$ – Michael Hardy Apr 11 '18 at 14:52
  • 2
    $\begingroup$ "[part of m.s.e.] is run in such a way that participaiton [sic] is intolerable when one is sober." Well, that would explain a lot of what I see on m.s.e. $\endgroup$ – Gerry Myerson Apr 11 '18 at 22:56
  • $\begingroup$ @GerryMyerson : $\quad\uparrow\quad$ Specifically, the review queue for closing questions. $\endgroup$ – Michael Hardy Apr 11 '18 at 23:15
  • $\begingroup$ But it's well-known that one should not drink and derive. $\endgroup$ – Gerry Myerson Apr 12 '18 at 0:07

Split them into two questions.

The first question asks "How do I make this problem/question precise?". (It's not clear whether that's going to work well in the Stack Exchange format, since if they're not clear what they want, it's hard to know how we would know, either -- but if they can provide a clear description of the phenomenom/problem, and are asking us for how to formulate it in a mathematical way, that could work.) Hopefully, answers to that question will give them some options on how they could formulate it mathematically, or at least some ideas on how to do that.

With that in hand, they can then ask a second question which is "What is the answer to this precise mathematical question?", where they formulate a precise mathematical question based on what they learned from the answers to their first question.

Splitting it into two keeps it narrowly focused, makes it work better on the Stack Exchange format (where having multiple different questions in a single post often doesn't work out so well well), and makes it more likely that at least one of these questions/answers will be useful to someone else in the future.


I've asked the moderators to migrate this question to stats (dot) stackexchange (dot) com. Apparently the usual mechanisms for migration cannot be used when a question has been deleted.


You must log in to answer this question.