I recently placed a bounty on a long-standing zero-answer question (that question is in this link): Does there exist a space filling curve which sends every convex set to a convex set ?

My goal was to promote the question. Did I accidentally do the opposite? If nobody answers within the time limit, does the question descend into a land of oblivion, where it will never be noticed again? I would prefer the question to remain (I did not mean to kill it). (PS: I don't care what happens to the bounty points.) A related followup question is in bold below.


I noticed this 0-answer question 1-2 weeks ago (I noticed it particularly because I searched for unanswered questions). I read it, walked away, and later came up with what I thought was an interesting answer. So I found the question again and typed up my answer. A user pointed out that I neglected a key assumption in the question (which I had indeed forgotten about).

I was going to delete my answer but I was encouraged to keep it, as it was a "partial answer." Of course, keeping a partial answer means the question will no longer appear when people search for unanswered questions. To compensate, I posted a bounty on the question. However, I now wonder if it would have been better to just delete my answer, since perhaps a zero-answer question has more visibility on this site than an expired-bounty question.


My own "partial answer" for this particular case likely does nothing to help answer the actual question. But in general, there may be hard questions for which partial answers can be helpful and/or can inspire others to give more complete answers.

So, ideally, it would be great if there was a way to maintain high-visibility for expired bounty questions and/or "interesting and hard questions" while also allowing for partial answers. Is there an existing mechanism for this, and/or is it possible to develop one?

  • 4
    $\begingroup$ Re: it would be great if there was a way to maintain high-visibility for expired bounty questions and/or "interesting and hard questions" while also allowing for partial answers. Some time ago I have tried to create a chat room for this purpose: Could list of unresolved past bounties be useful? But this attempt was not successful. However, some users are currently planning another bounty related room $\endgroup$ Commented Jan 6, 2019 at 6:27
  • 1
    $\begingroup$ @MartinSleziak : Thanks for your links and your support, it looks like a good idea. One reason I think having longstanding questions with partial answers is good is that it promotes collaborative effort that may eventually lead to a non-obvious conclusion. In the same spirit, there are many modern examples of results developed via blog. If people were discouraged from contributing to the blog unless they had full answers, then, the results would likely not have been developed. $\endgroup$
    – Michael
    Commented Jan 7, 2019 at 16:53

2 Answers 2


Having linked to the question which is the example motivating this question, it reads, in its entirety, as follows:

"Does there exist a surjective continuous function $f: [0,1]\to [0,1]^2$ which maps every convex set to a convex set?"

and then reviewing the meta post How to ask a good quetion helps illuminate problems with the question itself. One sentence. A problem statement question reading as though it was transcribed word-to-word from an assigned homework problem, or an exercise from a text exercise left for the asker to complete, or from a problem sheet given in a course. Note that the question provides no information about its source, or about the asker's mathematics background, current level or course in which the question emerged, or about why the question is relevant to the asker and/or the community, or no information about the asker's progress in answering the question. In short, no context whatsoever is provided.

Even in this case though, it is worth a partial answer (don't waste your rep on a bounty for a question like this), because a partial answer, or full answer, will bump the question to the homepage, and then, since it actually deserves to be closed, users have the opportunity to close a question that somehow went under the radar some time back.

I appreciate your desire to help an asker, and resolve an "unanswered" question, and I appreciate your generosity in your willingness to "sponsor" a question via a bounty deducted from your own rep. But for the sake of answering and preserving far better, yet unanswered questions, I'd encourage you to focus your generosity and your concern more wisely, on answering and/or promoting good questions, as defined in the above link I've already provided.

  • 3
    $\begingroup$ Thanks for your answer (I am just reading it now). It seems you are focusing on that particular question. The Eevee answer also mentined it looked like a problem set question. I doubt it is since it is too deep and hard but it indeed would have been nice to have the user say where that question is from. Perhaps they just thought it up and were curious. I don't know how to answer it and it would be nice to see a solution. $\endgroup$
    – Michael
    Commented Jan 7, 2019 at 2:18
  • 1
    $\begingroup$ @Michael: The user who originally posted this Question is gone now. You would probably do a better job of framing the problem in a new Question, or (alternatively) editing the Question to add context. The Question did not slip quietly into oblivion (it garnered a large number of upvotes when posted in 2017), so the lack of a full Answer reflects its difficulty more than any neglect. Your own efforts to solve it are an indication of that. $\endgroup$
    – hardmath
    Commented Jan 7, 2019 at 17:08
  • 2
    $\begingroup$ Michael, I think @hardmath's comment immediately above this comment from me is the best way to "resurrect" the question. Likely, posting a new and improved question, with a bit of context added (perhaps asking the question as a new question, and adding your "partial answer" for context), would be the best choice to give it new life. $\endgroup$
    – amWhy
    Commented Jan 7, 2019 at 17:19
  • 7
    $\begingroup$ Whatever the general rules and guidelines are, this is a very interesting and challenging question. You will also notice that it attracted some links and context in the comments. I don't understand why you would want to close it. $\endgroup$ Commented Jan 7, 2019 at 17:39

If nobody answers within the time limit, does the question descend into a land of oblivion, where it will never be noticed again?

Depends on how you compare the three states in which the question has been.

In the state in which you found said question, it was likely just buried deep in the Unanswered queue or somewhere else.

You found it, gave a partial answer - thus removing it from said queue - and then applied a bounty on it, temporarily putting it in the featured queue.

In a sort of way, it does kill the question. Now it's not in the unawnswered queue, and once the bounty fades it won't be in the featured queue either. It will just be one of thousands of questions which end up with answers, but no accepted ones, more than likely. But at the same time, I doubt there is a significant difference in terms of the amount of hits it will obtain as compared to before you found the question in the first place.

At the same time, if there was any "killing," it's because of (a) the inactivity of the OP (was their account deleted or something? I can't access their profile), and (b) the partial answer removing it from the unanswered queue. At the same time I doubt these effects are anything more than marginal, and it is definitely not the bounty itself that would have caused any "killing."

Part of the problem, too, is that this question comes off as a PSQ, so it really should've just been closed long ago, IMO, and I cannot really grasp why it has no close votes and was upvoted so much since such questions usually just get squashed pretty fast. That in and of itself is going to turn a lot of people off immediately.

Is there an existing mechanism for this, and/or is it possible to develop one?

There is not, to my knowledge. Dunno if it's possible to develop one, and I have no strong opinions either way on the topic if it is possible.

  • $\begingroup$ Is it true that it's no longer in the Unanswered queue? When I view unanswered questions I'm shown "questions with no accepted or upvoted nswers", suggesting and many have answers (but one's with zero votes). This suggests it'll remain officially unanswered unless someone notices and upvotes the partial answer. (NB I've not checked this example to see whether they did.) $\endgroup$
    – timtfj
    Commented Jan 7, 2019 at 0:17
  • 4
    $\begingroup$ I removed some comments, as some where unfriendly, and other replied to then inexistent comments. Please try to make your points without attributing bad motives to others. Focus on the actions and their consequences. $\endgroup$
    – quid Mod
    Commented Jan 7, 2019 at 0:41
  • 2
    $\begingroup$ @quid My comments were not nearly as unkind as instantly closing a poor new user's clear question. I would recommend policing that too. $\endgroup$ Commented Jan 7, 2019 at 1:02
  • $\begingroup$ "Is it true that it's no longer in the Unanswered queue?" -- I think it depends on the queue/how you find it. For example, filtering a tag by unanswered questions, I get both kinds (zero answers, and those with answers but none accepted). Using the "Unanswered" thing at the left side, though, I only see those with zero answers whatsoever. I was referring to the latter, fwiw. $\endgroup$ Commented Jan 7, 2019 at 1:38
  • $\begingroup$ @EeveeTrainer : Thanks for your answer (I am just reading it now). What is "PSQ"? $\endgroup$
    – Michael
    Commented Jan 7, 2019 at 2:03
  • 1
    $\begingroup$ @Michael - A "PSQ" is a problem-statement question. It's basically the MSE-post equivalent of a problem you'd find in a textbook: the person basically just posts the problem/question and little-to-no other context. (For example: "How do I solve (this integral)?" is basically their entire post, and they mention nothing about what they tried to do to solve it, what methods are permitted, where they found the integral in question, etc., all of which are much more helpful in aiding the OP to see the direction they need to go.) $\endgroup$ Commented Jan 7, 2019 at 2:06
  • $\begingroup$ Bear in mind that this has no bearing on how interesting/useful the question actually is. A good question could be framed poorly, in other words. The question you answered may well be a deep, curious, and tricky one - I'll be the first to admit it's above my head, so saying either way (it's simple or it's difficult) is beyond me - but posting just the question and nothing else generally makes it looked down on. Such behavior is generally discouraged at MSE (typically resulting in the closure of the question), which also explains my comment as to my surprise that the question wasn't closed. $\endgroup$ Commented Jan 7, 2019 at 2:10
  • 3
    $\begingroup$ @EeveeTrainer : I see. Yes it reads like a problem set question, but I doubt it is. It seems way too deep and hard. Reading the comment trail on that question leads to some related arxiv publications but even they do not answer it. So I am curious about the question, it is an interesting question, and I would be happy to see somebody answer it. It seems to be a very nice challenge. $\endgroup$
    – Michael
    Commented Jan 7, 2019 at 2:13
  • 1
    $\begingroup$ @EeveeTrainer the Unanswered on the left side basically has 2 filters: "unanswered" (no accepted or upvoted answers) and "no answers" $\endgroup$
    – Andrew T.
    Commented Jan 7, 2019 at 4:15

You must log in to answer this question.

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