The question Is The Examiner a reliable news source? was of dubious quality, and several commenters suggested that it should instead ask, "Do the Celestial Seasonings teas contain dangerously high pesticide levels?" A helpful third party edited the question.


  • The original poster accepted an answer to the original version of the question.
  • The helpful editor got consensus from commenters but not the original poster.
  • The new version invalidates the accepted answer, which is now attracting downvotes.
  • The original poster has neither rolled back the question change nor unaccepted the answer.

This puts me in the bad position of having an accepted answer that doesn't actually answer the question (although it does answer the original question). What's the best way to address this?

  • Roll back the question? (But it wasn't a great question.)
  • Delete my answer? (But it solved the original poster's problem.)
  • Change my answer to match the edit? (But that's too radical a change.)
  • Do nothing?

I think the root of the problem here is that the edit to the question was too radical a change to make without agreement with the original poster.

Note: Changing my answer to match the question edit would seem ideal, except: 1. Even if my edits are total rubbish, it would still look like the OP accepted them. 2. I am not actually interested in answering the revised question.

  • I agree that the edit changed the meaning of the question.
    – user5582
    Jul 4, 2013 at 0:37

8 Answers 8


I am not so sure that your original answer addresses the point of the original version.

The original version has a couple of big problems, and in particular that the whole question is based on bad logic. I believe this is called "fallacy of division"

the reliability of a source in general does not guarantee the reliability of any claim it makes

Examples: a broken clock is right twice a day and even the Lancet published bogus papers.

Since the question was referring to a specific claim, but then derailed asking for the wrong evidence, I believe your answer should be, in any case, fixed to actually address the claim directly, or at the very least point out the major flaw in the question!

The aim of this site is to make the Internet a better place. I do not think that poor logic fits the bill. Your answer is sound, if incomplete, but it validates bad logic.

  • 1
    Excellent points. I mentioned the “broken clock” problem myself in comments to my original answer. I'll revise my answer to acknowledge that it's not an appropriate way to settle a dispute. Jul 4, 2013 at 8:16
  • 3
    +1 for finally ending a couple of days of searching what that fallacy is called.
    – user5341
    Jul 8, 2013 at 17:05

In general, edits that invalidate existing answers should not be done. The exceptions are if the edit saves the question from being closed, then such radical edits are acceptable.

In this case, the original question about the reliability of a specific newspaper is off-topic here. So editing it is acceptable even though it invalidates an existing answer, as the alternative would be to close the question.

  • Just to confirm, is it okay to change the meaning of a question, without confirmation from the original asker, in order to save the question from being closed?
    – user5582
    Jul 4, 2013 at 15:19
  • Yes, if the question would otherwise be closed such drastic edits are acceptable.
    – Mad Scientist Mod
    Jul 5, 2013 at 7:34

As one of the down voters of your answer, an ideal solution from my point of view would be (in order of preference):

  • You deleting your existing answer and answering with a correct answer to new version of the question.

  • You deleting your existing answer (without answering)

  • The question being deleted altogether, since the edit was too radical. Instead, ask the new version as a separate question, so as to not reward the original poster with reputation for a crappy question.


Honestly, in a situation like this I would just delete my answer and then reply back to let us know if you were awarded the Peer Pressure, Disciplined, or both badges since you have -3 and +3 votes on your answer. :)

The others have pretty much presented all of the relevant arguments and philosophy behind how the Stack Exchange network sites work and this question appears to have been one of those edge cases in that normally such a question would have been closed as off-topic as opposed to being edited in such a way. The loss of reputation is a bit annoying, but it generally doesn't take too long to earn it back.

  • Ha! Good point about the badges. Unfortunately, I can't delete the answer myself, because it's accepted. If "delete" becomes the consensus here, I'll flag it for a moderator to delete. Jul 11, 2013 at 6:04

What to do?

Don't answer badly written questions

This was a rubbish question when it was first asked. If we all refrain from posting answers until the question is a proper question within the scope of this site, then this situation won't arise.

In this case, your answer was for a badly-written question. So, you've two options: delete it, or edit it to answer the question as it now stands. If you don't edit it, it will get flagged as "not an answer", and then I expect it's likely to be deleted.

  • 1
    It's still an answer – just a partial one now. The user's original question is still in there, it's just less visible after the edit. Also, the question isn't all that different from folks asking whether Wikipedia is a reliable source – it just probably belonged on Meta rather than the main site. Jul 4, 2013 at 7:18
  • Option of moving a question to meta is possible.
    – Wertilq
    Jul 4, 2013 at 7:29

Roll back the question and migrate it to Meta

The impression I get from discussion here and on the question itself is that:

  • The question shouldn't have been edited like that after answers were posted, but
  • The original question was a better fit for Meta than the main site.

Therefore, the best solution may be to roll back and migrate.

  • Meta or main site, the question, as it was originally, was bad...
    – nico
    Jul 8, 2013 at 19:18

I am the one who edited the question, so I feel I should point out why I did that.

In the original question, the OP was asking about whether The Examiner was a reliable source because he was investigating a rumour on Celestial Seasoning teas.
Now, I think we all agree that asking for the reliability of a source is off-topic for Skeptics. Personally, I think that, although possibly not off-topic, it would also be a poor question for Meta, for the reasons that other people have already pointed out.

I then thought that the best option was to actually try and edit the question so that it would solve the problem that the OP was trying to investigate, that is, find the information about the tea.

Note that I specifically asked the OP to revert changes if he did not like the edits, but since there has been no answer we have to imagine he did not visit the site again, lost interest in the question, or simply was happy with the edits.

Now, what should you do with your answer?

The idea of Stack Exchange sites is to create a high quality repository of both questions and answers.
If you feel that your answer does not answer the question anymore simply delete it. Otherwise the community will decide by voting whether or not the answer is worth staying there.

Having an answer is really no reason to leave a poorly written question standing there as it is. Proof is, the question had 3 negative votes before the edit and now has had 4 positive ones on top of that.

Another option would be to copy the original question on Meta (although maybe it would still need some edit...), and transfer your answer with it, while leaving the edited version on the main site with no answer.

A final option, although I don't know whether that is technically possible, would be for a mod to un-accept the answer, although at that point I would see deletion as a better option, since you are not interested in editing it.

  • If you downvote PLEASE SPECIFY WHY. We are here to discuss, and just leaving a downvote with no explanation does not serve any purpose at all.
    – nico
    Jul 8, 2013 at 20:26

Edits to a question should NOT happen if you already have an answer(or answers), unless you edit all answers too. If an answer is accepted and the original questioner of the question is NOT around, then you should NOT edit the question. Not really any exceptions. Of the reason you already mentioned. It's irrelevant how rubbish the answer is, it's still the accepted answer. Only other option is like delete the answers, then remake the question.

But for that it's better to just remake a new question, than just destroy a current one.

I think that edit was really poor sport, even if it made more sense to ask that from the claim, it was without consensus of the questioner, it broke an accepted answer. And while the original question was a bit hard to answer, the new question was not overly interesting either. It focuses on a small fraction of what the original one asked.

  • Should I suggest an edit to roll back to the original version of the question then? I suspect that might lead to the question being closed, but that might be better than the current situation. Jul 4, 2013 at 7:20
  • I believe so yes. I think it would be better if the question was closed than what was done to question now.
    – Wertilq
    Jul 4, 2013 at 7:28
  • Perhaps a rollback and migration to Meta would be the best solution? Jul 4, 2013 at 7:36

You must log in to answer this question.

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