Update: I withdrew this policy proposal. It received 5 downvotes and no upvotes.

This is a policy proposal. The proposed policy in the question or its answers will be adopted in the case that after 3 days it received at least 5 upvotes with a 2/3 majority in favor (*). Please upvote or downvote this question or its answers based on your opinion of the policy statement as presented, not just based on whether the premise of the question or its answers is reasonable.

(*) That is, x≥5 and x≥2y, where x is the number of upvotes and y is the number of downvotes.

I propose the following as a policy:

POLICY. Questions that describe a proof of some statement or an algorithm for some problem and ask others to do all the work to check its correctness are deemed to be off topic and closed.

The intent is to give a clear ground to close questions like “Algorithm for Longest Path in Undirected Weighted Graph” in the future. While the question had a specific goal (present a counterexample to the stated algorithm), I find it wrong to ask other people simply to do all the work to check the correctness of proofs or algorithms.

The proposal has something in common with some of the answers to “How to ask a good question?,” namely this one, this one and this one, even if they did not already cover the case with which this proposal is concerned.

  • $\begingroup$ Before giving the vote, can you explain how the policy separates the difference of the question you've mentioned and the question like, e.g. this one? The second one also ask for verifying the reasoning. $\endgroup$ Commented Dec 16, 2010 at 13:32
  • $\begingroup$ In the stated example, the question author invested quite some work after his initial posting as you can see from the comments. So it was apparently not a request "to do all the work"; a rough hint I provided was enough for him. Note also that the questions fulfilles all demands we have had until now. More importantly, and this is why I downvote, the proposed policy is orthogonal to the established scope, not an implication. I wonder what kind of questions remain if we restrict the set of allowed questions even on research level. $\endgroup$
    – Raphael
    Commented Dec 16, 2010 at 13:33
  • $\begingroup$ @Hsien-Chih: I hope that “ask others to do all the work to check its correctness” applies to the question I mentioned but does not apply to the question you mentioned. $\endgroup$ Commented Dec 16, 2010 at 13:38
  • $\begingroup$ @Raphael: (1) The asker of the question I mentioned should have shown the evidence that he was not just asking others to do all the work in the first place. Keeping that question unclosed would set a bad example. (2) The proposed policy is not an implication of what has been stated explicitly, of course. That is why I made a proposal. And please stop insisting that this is the only kind of questions we get. Moreover, TCS SE is a site for research-level questions, and please do not try to argue about it on this page. $\endgroup$ Commented Dec 16, 2010 at 13:42
  • $\begingroup$ @Tsuyoshi: I think I got your point, thank you for the clarification. $\endgroup$ Commented Dec 16, 2010 at 13:49
  • $\begingroup$ @Hsien-Chih: By the way, can you post it as an answer? I think that it should be stated in a larger font than comments, and also I can accept it after I wait for a while to see any other opinions. $\endgroup$ Commented Dec 16, 2010 at 15:22
  • $\begingroup$ @Tsuyoshi: Okay, I'll post it as an answer, and delete the duplicated comments. $\endgroup$ Commented Dec 16, 2010 at 15:26
  • $\begingroup$ (In case anyone wonders about the incomplete conversation in the comments) For the record, I moved a reply to Hsien-Chih to a comment on his answer. $\endgroup$ Commented Dec 16, 2010 at 15:39
  • $\begingroup$ @Tsuyoshi: (1) Read the comments. It has been stated often enough that questions can be reopened after effort has been shown. In this case, the question was closed after effort had been shown. Weird. (2) I do not know where you got the stuff from you "replied" to in the second half of (2), but certainly not my above comment. $\endgroup$
    – Raphael
    Commented Dec 16, 2010 at 23:17
  • $\begingroup$ Tsuyoshi, I didn't have any good choices in the close option list. I'm only closing because the proposal has been withdrawn $\endgroup$ Commented Dec 19, 2010 at 19:38
  • $\begingroup$ @Suresh: Ah, I see. Thanks for the explanation. $\endgroup$ Commented Dec 19, 2010 at 20:23
  • $\begingroup$ @All: I've discovered all this discussion only 20 minutes ago. 1) I've never thought (not even for a second) that the algorithm in my question had any minimal hope to be a $P = NP$ proof (as someone seems to insinuate). I knew that the algorithm was flawed somewhere (it was too simple to be correct), and just asked some help to identify the bug. 2) I absolutely don't want other people to do "all the work" for me. 3) I've put some honest effort in that algorithm (although not a research-level effort, since I'm just a TCS amateur, not a TCS researcher)... $\endgroup$ Commented Dec 20, 2010 at 15:34
  • $\begingroup$ @Walter: Just to make it clear that I did not ignore you, I posted a link to this meta discussion as a comment on your question on December 16. (This is not to say that you are claiming that I ignored you.) $\endgroup$ Commented Dec 20, 2010 at 15:41
  • $\begingroup$ @All: ...4) I'm not a crank. 5) I fully recognize that I committed an error: I was too quick in asking the question. If only I tried the complete graph (as suggested by Raphael), or if only I implemented the algorithm, I would have found the bug and this site wouldn't have one more non-research level question. $\endgroup$ Commented Dec 20, 2010 at 15:42
  • $\begingroup$ @Tsuyoshi: Yes, it's exactly there that I've found the link to this meta discussion. These last days I've visited the site less frequently, that's why I didn't immediately see your comment to my question. $\endgroup$ Commented Dec 20, 2010 at 15:46

1 Answer 1


While problems without a clear demonstration that the author had put effort in are less-motivated for me to answer, the judgement of "the evidence that they have took some work" varies from people to people. It is especially hard to tell the difference when the question is from a new user of the site. They may have good question in mind, but failed to express the ideas in a suitable way, as defined by the site's policy.

For sure we do not want a flood of homework problems and requests for checking the correctness of proofs to some famous open questions, but those have been already discussed and banned in these discussions. And furthermore, we have an alternative choice for questions which do not show their efforts - just ignore them.

In summary, as long as the problems are within scope (that is, research level questions in theoretical CS), I see no immediate need to set a policy for restricting the type of the problems people ask, since that may introduce some side-effects.

  • $\begingroup$ (Moved from my comment attached to the question.) That makes sense, thanks. Indeed, if the asker does not show an evidence of effort, at least I am not interested in answering the question, and everyone can use his/her own judgment, which may render this policy unnecessary; we can see how things turn out. I still think that it was right to have closed the question I mentioned, but the reason for that may be that it was not a research-level question, not because it was asking for a correctness check. $\endgroup$ Commented Dec 16, 2010 at 15:37
  • $\begingroup$ Hsien-Chih, I agree completely. $\endgroup$
    – Raphael
    Commented Dec 16, 2010 at 23:19

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