Before you reach for the VTC as duplicate button, I realize that very similar questions have been asked already. I've been looking through those to find a general guideline that can be applied site-wide, and each of them deals with a specific situation, dealing with a unique question with its own inherent problems.

This post is intended to get a consensus on what the general guideline should be, that will apply as a general rule across the board.

Quite often, we see questions that are closed for one reason or another. In many cases, very minor edits can be changed to bring it back in line and make it suitable for re-opening.

The principle of allowing questions to be edited to allow re-opening is built into the StackExchange network. When you get to a certain reputation level, and are allowed to cast "re-open" votes, there's an option to "edit and re-open" on questions that are in that queue.

So, it's perfectly natural to take a problematic post and try to edit it into shape. Many of us do it. I do it, in the spirit of trying to be helpful, but I'm quickly seeing that this is becoming an issue.

In many of the cases, the edit, minor as it is, is often enough to invalidate existing answers. This, naturally, doesn't sit well with someone who invested time and energy in writing an answer to the question, only to have the question changed.

What should be our official stance on editing a question to make it suitable for re-opening, particularly when that edit is going to invalidate the answers?

I'm initially posting two opposing answers to choose from, and anyone is welcome and encouraged to put other/better possible answers.

5 Answers 5


This question came up on Meta.SO just a few days ago, and the consensus view seems to be "yes, it's OK to do this, but do it carefully."


I really think this needs to be handled on a case-by-case basis. For old questions, which were once considered acceptable, it is better to close the question (or lock it for "Historical Significance"). This keeps the (once relevant) question and valid answers around.

For new questions, however, I'm inclined to suggest we delete the now-invalid answers. As David said, if somewhat crudely, in his own answer, if you're going to answer questions that shouldn't be answered, and then complain because your answer is invalidated, that's really your own fault. Training answerers is just as important, if not more so,than training questioners.


We should allow the edits.

The people who answered the question shouldn't have answered it in the first place. They should have voted to close. If you're going to answer questions that shouldn't be answered, and then complain because your answer is invalidated, that's really your own fault.

The people who answered (and shouldn't have in the first place) the option of deleting their own original answer or editing it.

  • That sounds so mean... Commented Aug 2, 2014 at 13:46
  • This will be a useful tool in certain situations, it should not be blanket-banned. Commented Aug 2, 2014 at 13:47
  • 2
    Initially I had reservations about your second paragraph (because those people may disagree that the question even should have been closed), but then I realized they can always ask a new question that would fit their answer. Within reason, of course. Commented Aug 4, 2014 at 15:42
  • 2
    I'm not comfortable with this as a blanket rule. Questions closed as "unclear what you are asking" are sometimes understood more clearly by some than others, and answered effectively. Thus the existence of good answers, in some cases, should be taken into consideration. Commented Dec 11, 2015 at 15:07

One thing to do is to edit the question, and then edit the answers to bring them into line with the new, allowed question. A person with enough rep to do the former also has enough ability to do the latter.

  • 1
    I don't think editing answers for anything other than grammar and other editorial items is good practice. The actual content of the answer is the OP's responsibility. I think commenting "The question has been edited you should update your answer" is good enough.
    – user3961
    Commented Aug 11, 2014 at 17:06

One Option: We should disallow editing questions in such a way that invalidates answers.

This is based on looking at how it's addressed on other META sites, and also based on the overall leaning of the answers to the specific instance questions noted in the preface on the question.

I found this relevant post on the Meta.stackexchange site in this question. (emphasis mine)

If, based on your improved understanding of the situation, you actually have a new/different question that you want to ask to help solve the same actual problem you had before, then you should post a new question, optionally linking to the old one.

If you have the exact same question, and are simply clarifying the question so that future readers will be more likely to understand what you were asking from the start (or at least what you thought you were asking) then you should just edit the existing question.

That "If you have a completely different question" is the heart of the matter here.

As many of us have tried to explain ad-nauseam while trying to explain the Truth issue: "Is X true is off-topic. Does the Catholic Church teach that X is true is not."

I still agree with that, but it's also a completely different question. That tiny little edit, which is enough to bring it in line with the site guidelines also changes it completely.

The original question should remain closed (and maybe deleted) and a new, better question should be asked.

I do foresee some issues with this approach.

  • I can see new users posting variations of the same question until they find one that sticks
  • There's going to be a tendency to see these as duplicates
  • It's more likely to fill the site up with bad questions - unless the originals are deleted

If this is our approach, we will likely have follow-up guidelines to deal with the fallout.

  • This is not how I've handled it in the past, including and up-to yesterday, but I think I've been acting inappropriately all along, so if this is the guideline, I will change here and now. Commented Aug 2, 2014 at 13:39
  • I think the example you used is a bit out of context. The editing there happened before the question was closed and invalidated answers to an open question. Additionally it made the question worse not better and it ended up getting closed. That seems to be a different scenario than you are raising here with a closed question being edited for it to be openable. I totally think it's okay to invalidate answers in that case.
    – Caleb
    Commented Aug 2, 2014 at 13:49
  • This is probably a good general policy, but it shouldn't be the only option. eg. if either the question or (more likely) the answers have a significant measure of merit but some re-scoping is found to be necessary, posting a new question should not be the only option. The issues you raise are valid. Commented Aug 2, 2014 at 13:51
  • @Caleb - Good point. I'm editing that out now. Commented Aug 2, 2014 at 13:53

You must log in to answer this question.

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