What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 127 Stack Exchange communities.

Let's say I asked "How do I do X". Then I found it's impossible. I thought such questions should be closed but there's not "impossible" option on list.

Should I answer myself "X is impossible" and accept own answer or what?

share|improve this question
1  
(joke mode) We should let the community decide. Ask if anyone knows if P = NP and see what happens. –  Super Long Names are Hilarious Jul 29 '10 at 4:57
    
@Super: actually, I saw this solved on a whiteboard on a recent trip to my alma mater. You need only set N=1, then P=P and the proof is complete! Seriously though, I tried this on Apr 1 this year. It was popular, but I forgot to CW it, so it was closed, then someone else asked the same question with CW, and it was even more popular. –  Pops Jul 29 '10 at 14:18
1  
@Popular - How can anyone ask a more popular question than you? If a question is asked by Popular Demand, it's got to be worth keeping. –  Super Long Names are Hilarious Aug 1 '10 at 3:37
    
@Super: to be fair, I asked it on SO, where my name is not Popular Demand. But I appreciate the blanket vote of confidence, and will abuse it wherever possible henceforth. –  Pops Aug 3 '10 at 13:47
1  
Where's Alan Turing when you need him? –  Andrew Grimm Aug 5 '10 at 13:20
add comment

4 Answers

up vote 18 down vote accepted

That's exactly what I'd do. "This can't be done" is a legitimate answer, and it's fine to answer your own questions, particularly if nobody else seems to know the answer. You don't want to close the question, because someone else might be wondering how to do X and search SO, and your answer that it's impossible will help them too

share|improve this answer
10  
you should at least mention how you came the conclusion that it's impossible, otherwise it's a pretty useless answer –  Tobias Kienzler Jul 29 '10 at 8:49
add comment

I would put up an answer stating that you've found it's not possible and why that is. However, I wouldn't encourage closing such a question, because in the future it may become possible, at which point a new answer should be added that states how it can be accomplished.

share|improve this answer
add comment

There are degrees of impossibility. I like the idea of posting that as an answer if you really believe it, but hold off a little while on accepting your own answer. Someone might take it as a challenge, and they might even recognize a solvable special case that, for your purposes, "achieves the impossible".

share|improve this answer
add comment

Someone may have a technique or workaround to achieve the same result.

You can't say "impossible": it's may only be "impossible" for you to achieve without inspiration from an answer.

There are around 500 SO questions with "impossible" with many accepted answers...

Edit:

Of course, there are always impossible requests but I'm trying to answer the "impossible. give up" mentality...

share|improve this answer
2  
OK, then I'll post a question on how to tell, from a priori analysis, whether a given arbitrary program will run forever or not, given large enough storage. Seriously, some things ARE impossible. –  David Thornley Jul 29 '10 at 14:48
    
@David Thornley: you are correct. On reading this a half day later, it's wrong. –  gbn Jul 29 '10 at 17:06
add comment

You must log in to answer this question.

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