Should answerers be held responsible for updating their answers when the questions are revised?

Question

Here's a typical situation. I see a question. I write an answer that draws upvotes. Then the OP revises the question. My answer starts getting downvotes. The critical comments are generally that the answer is "out of sync" with the (revised) question.

My typical response is that my (timestamped) answer is valid for the question that I answered, not necessarily for the question in its current form. Is this posture reasonable?

More to the point, as a voter, when I see an answer out of sync with a question, does it make sense for me to take note of the "time stamps" on the (current) question, answer, and original question?

And then there are questions whose body is different from the title. How should those situations be handled? (Typically I choose one or the other and answer it.)

Answer 1

Questions should not be revised in ways that invalidate existing answers. Edits to questions that do that should be rolled back. The new question should be posted as exactly that: a new question.

Answer 2

I've found myself in the same position more than once. It's also worth noting that it's not always the OP that changes the question.

---

I agree that your answer should only be judged against the question as it was framed when you wrote your answer. I would prefer that if the OP has a new question (often based on the information that you have provided in your original answer) then they should ask it as a new question. We live in an imperfect world.

If I see an answer that appears to be out of sync with the question, I certainly try to look at the times the question & answer were posted / updated. Personally, I'd probably post a comment rather than downvote, but that is just me.

[Actually, I don't downvote many answers, I prefer to upvote good answers and let them rise above the rest. I'm more likely to downvote poor questions, where I can reverse the downvote if/when the question is improved. (I prefer to downvote questions, rather than closing them, where possible. It seems to me that our VtC process is something of a "trapdoor function").]

When I've found myself in this situation, I've usually edited my answer to include a note that I was answering the original question, rather than the edited version. Otherwise, I could end up chasing the question indefinitely! If people still want to downvote, that is their prerogative.

---

As for your last question, where the question in the body differs significantly from the question in the title, in most cases I'll post a comment asking which question they'd like me to answer. If I do post an answer, I will simply state explicitly at the start of my answer which question I am answering.

Answer 3

OPs, when editing their own questions, should definitely take into account any answers that have already been posted. Not to do so is clearly unreasonable and shows a lack of respect for people who have taken the time to respond to their (original) question.

If you do not feel inclined to modify your answer, I think you would be well within your right to edit your answer and add a sentence at the beginning stating something along the lines of The answer below was posted before the OP edited his/her question and then briefly summarize what the original question was. Of course, you shouldn't be put into a position of having to do this, but I can't see any other way.

When OPs' main question doesn't seem to match the body, I'll usually post a comment. If there is no response to that comment (and the time stamp shows the OP has logged in since I made the comment) then I usually vote to close as 'unclear what you're asking'.

As someone who posts questions regularly, I believe I've always edited for one of the reasons stated in the list below, which could perhaps serve as a guideline for OPs as to what is acceptable (in my humble opinion).

1. Correct typos and / or language errors (spelling, grammar etc)
2. Rephrase parts of the post to bring out a point or points more clearly. This is usually in response to answers which seem to be ignoring part of the question (which may be because my wording wasn't clear enough). That said, if someone answers one part of my question and acknowledges the other part but is unable to provide an answer for it, that's fine.
3. In response to comments. Sometimes someone makes a valid point in a comment which needs addressing with an edit.
4. Highlight a part of a question which isn't being addressed or isn't getting enough attention. I usually do this in bold.
5. 'Refresh' the question. It's slipped down the list in a busy period without getting much attention. In this case, my edit is usually done by highlighting key parts.
6. Research update. Added extra information / context that I've found in my research since I posted the question. This is aimed at helping others to reasearch
7. Perhaps a little controversially, I have at times added a 'sub-question' closely related to the main post (but I have never changed the main question). The only other option I would have would be to post it as another question, but then it would probably get flagged / closed as 'duplicate'.

Answer 4

I have dealt with this two ways.

• First, if my answer was still at least partially relevant, I will edit in a note on my answer that it was answering the original question, explaining the difference if necessary.

• Second is self-delete. I had one answer with four votes, nothing special, question changed completely, so I deleted my then irrelevant answer. Not much else to do.