5
$\begingroup$

Question: What makes an NP-hard problem not to be an NP-complete problem ?

According to his comment, Kaveh deleted two answers posted to the above off-topic question. While I agree that the question is off-topic, I am not sure if deleting answers is desirable. It seems to me that deleting answers to off-topic questions is against the general idea of minimal intervention. At the same time, deletion is kind of reasonable, and I would understand if many think that it is at the discretion of moderators.

$\endgroup$

3 Answers 3

Reset to default
7
$\begingroup$

I was wondering this myself. The argument in favor of deletion is the usual "don't feed the animals" principle. If we don't delete answers to s closed question, then we're allowing an end-run around the action of closing and users might be tempted to "post and hope to get an answer before moderation kicks in".

The cons are as you indicate. It violates the idea of minimal intervention. I lean slightly towards deletion myself but am willing to listen to arguments the other way.

$\endgroup$
2
  • $\begingroup$ If we see the OPs as agents and the reputation as feedback, then we consider answering a offtopic or for other reason to-be-closed answer as a negative action. Why not introduce a loss of reputation for this action, thus discouraging posting answers to question that are likely to be closed? I am not sure how this will affect borderline cases however. $\endgroup$
    – chazisop
    Apr 17, 2011 at 12:25
  • 1
    $\begingroup$ the short answer is that we don't have control over such mechanisms: they'd have to be proposed in meta.stackoverflow.com $\endgroup$
    – Suresh Venkat
    Apr 17, 2011 at 17:28
4
$\begingroup$

Here is what happened in this case:

The questions is clearly off-topic for the site and was closed by Dave. But two answers were posted before the closing. The first one was a one line answer by a new unregistered anonymous user (i.e. didn't have the reputation to comment):

To be NP-complete, it must be in NP.

So I converted it to a comment. But the second answer was similar in essence (by a regular relatively high rep user) and I felt it would be unfair to just convert one of them, so I converted the second one. Then I felt that it would better if I just delete them since they were not adding any information not on the Wikipedia page.

I personally didn't feel good about it after deleting because we don't have a clear policy on this (and the second answerer knew that the questions is clearly off-topic and that we don't answer them on cstheory), my justification for deletion was previous meta discussions we had where IIRC the conclusion was that we should not answer these questions even in the comments and should just forward the OP to the place where he/she could find the answer (Wikipedia, textbook, ...).

I also felt that deleting them would be in line with closing clearly off-topic questions by moderators (i.e. inside minimal mod intervention). I would have suggested adopting a policy like:

answers not containing new information (in addition to what is on Wikipedia, ...) to clearly off-topic questions should be deleted, preferably by OP.

But I think we probably don't need to do this for the following reason:

The feedback from Math.SE guys on adding Math.SE to the off-topic flagging wizard has been quite positive, so I think it we can simply migrate these questions with their answers to Math.SE and there is no need to delete the answers by us if the question is migrated to Math.SE.

ps: I also made a small edit to the general FAQ (Math.SE entry) based on the feedback from Math.SE folks.

$\endgroup$
3
  • $\begingroup$ I understand your reasons, but still think it would have been better to "educate" the resp answerers and prompt them to act accordingly. +1 for moving more questions to Math.SE, when appropriate. $\endgroup$
    – Raphael
    Apr 16, 2011 at 9:09
  • 2
    $\begingroup$ “Then I felt that it would better if I just delete them since they were not adding any information not on the Wikipedia page.” I do not think that this is a good reason to delete something posted here. $\endgroup$
    – Tsuyoshi Ito
    Apr 16, 2011 at 21:10
  • $\begingroup$ @Raphael, I think I agree, it is better to let the OP delete the answer. @Tsuyoshi, yes, that was not an argument by itself, I stated it because I wouldn't remove the comments (or even the answers) if they had non-trivial points not on wiki, but as I said I don't feel good about it, it would be more reasonable if I let the OPs delete their answers (but on the other side, having the answers might defeat the point of not answering clearly off-topic questions).(ps: I have emailed SE admins the request for adding Math.SE to the list. I haven't heard back yet but hopefully they will agree with it.) $\endgroup$
    – Kaveh
    Apr 17, 2011 at 7:22
3
$\begingroup$

Back then when I was struggling with figuring out the appropriate scope, I was angered by people downvoting my answers to OT questions. I can not imagine how I would have reacted were they just deleted. I think deleting is a bad practice in general, and especially in this case. A new, dedicated user might be put off by it. You send the message that if a mod disagrees, your work is for naught.

$\endgroup$
1
  • 1
    $\begingroup$ I guess I share part of your feeling, but I do not agree here. Deleting answers without explanation would be definitely bad, but Kaveh explained why he deleted the answers. One can even argue that posting answers without considering whether the question is appropriate or not is the answerer’s fault to some extent. $\endgroup$
    – Tsuyoshi Ito
    Apr 15, 2011 at 20:28

You must log in to answer this question.

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