The other day I answered a question that was quite poorly written that arguably did not include all the details that was needed to give a complete answer to it. If I remember correctly, the OP's code example contained more than one error. One of the errors, however, was quite obvious and my answer (as well as others) pointed out this error. This error was apparently the one that OP's problem in all was solved as he commented on the answers, saying thanks, and I believe that he accepted one of the answers.

My own answer received 5 upvotes (which to me is well above average :)) and obviously I was content with the outcome. Since the question was a bit too vague, it received a few downvotes and eventually was closed and finally deleted. Thus, the rep I gained for my answer was lost.

I am OK with this behavior of SO, of course, but at the same time it discourages me to answer poor questions in the future. Why spend efforts on a question that is likely to be deleted?

So, my question is, should I avoid answering questions that I feel might be closed or should I give it a try anyways if I feel that I can help? Does the moderators take into consideration if any answers help the OP before closing? I feel that trying must be better for the community than leaving the question to die without trying. And if I give it a try, what should I do when I see that the post is attracting downvotes? (I don't have the rep to cast delete/undelete votes). Should I edit and flag if it gets closed? What do others do?

(I am interested in general guidelines, not in this case particularly even thought it serves as example).

  • 1
    Yes. If only to avoid putting hard work into an answer that might ultimately get closed and possibly deleted by the community (and then losing the rep as well); we want all the members of Stack Overflow to be as efficient as possible. =)
    – casperOne
    Commented Aug 23, 2012 at 18:35
  • hehe, true that ;) Commented Aug 23, 2012 at 18:38

3 Answers 3


should I give it a try anyways if I feel that I can help?

I would say that yes, though part of that is improving a bad question.

At the end of the day, however, no one can tell you what to answer and how to answer it. If you feel that a question is too bad to answer, even if you believe you know the answer, there are several things you can do:

  • Comment to the OP and ask them to improve the question in the way you feel it to be deficient
  • Edit the obvious problems in the question - this may well help with such a question not getting closed and deleted
  • Give an answer encompassing all the issues you have identified in the question

My personal approach entirely depends on the question - if I see something that I can kind of understand, I will ask for clarification. If I see something that is an obvious mistake (typo, code that has not been fully anonymized and such) I will edit those out and try to improve the question. I will answer if I feel confident that I understand the issue and that the question is clear.

  • 1
    Improving and answering the question it is, depending on the case, rather than just answering. Maybe it is obvious, but I appreciate the answer and the guidelines. (+1) Commented Aug 23, 2012 at 18:26

I would focus my effort on improving the question (tell the OP what's wrong, edit to fix grammar/formatting/etc., ask clarifying questions to find what's missing) rather than answering. Once you, someone else, and/or the OP has put enough effort into the question to ensure that it is no longer close-worthy, then it's time to consider putting time/effort into answering.

Trying to answer a poor question, without fixing the question, will just least to more problems. You not getting your rep is really the least of them, too.

  1. While you may think you're right, it'll be much harder to know if you're right. For a good question, with a good answer, it's not debatable or questionable. The answer should be clearly right, not a guess.

  2. You are encouraging bad behavior. When people post poor questions and get answers there is no incentive to learn how to post good questions. They won't even care if the question is later closed. While it may cause a little bit of good this once, it's detrimental to the community as a whole to be swamped with poor quality questions. It's also an example to others. When people see lots of poor questions they'll assume it's appropriate. When all of the questions they see browsing the site are great they will have a good example to follow.

  3. Many close-worthy questions aren't entirely unanswerable, they're close-worthy because they tend to elicit poor quality answers. Some examples include: questions asking for product recommendations, lists of things, questions that are off-topic (but otherwise still answerable), etc. For these types of questions, while you could answer them, you are again causing more problems than you're solving. While these answers may answer the question, they will also be delete-worthy.

  • Point 1 is very true. "The answer should be clearly right, not a guess.". I agree with 2&3 as well. Thank you for your thoughts! (+1) Commented Aug 23, 2012 at 18:19
  • 1
    This is exactly what people should walk away with. If you can understand a poorly written question sufficiently to answer it, you can generally spend a few minutes making the question awesome too. Some discretion is needed however, not every train wreck is worth salvaging if you are reasonably certain that it's not going to stay around very long anyway, even if edited. The highly localized or obviously duplicate 'floaters' boil down to asking yourself if you feel generous enough to give five minutes of your life.
    – user50049
    Commented Aug 24, 2012 at 7:23

Closing is based entirely on the question, and is not a permanent end state. See What is a "closed" or “on hold” question?

If you can provide a helfpul answer, definitely do that.

If the question is poorly worded and/or formatted, or if it is unclear, it could potentially be closed. If you want to prevent closure, or get it reopened rather than deleted, then edit the question to improve it. Make it more readable. Don't change any fundamental meaning, or add/remove details that might change the issue. Do what you can, and leave comments asking the user to add any missing information, or correct anything that might not have been transcribed correctly (you said the code have several errors - perhaps the code the user was testing was fine except for the one issue, but was mistyped in the rush to get an ansewr).

  • 1
    Thank for your thoughts, it makes sense. The link to whats a closed question is most helpful. (+1) Commented Aug 23, 2012 at 18:13

You must log in to answer this question.

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