What to do about posts that don't answer the rewritten question?

Question

A controversial question was substantially rewritten, but not before a number of answers were posted and discussed at great length. Is it appropriate now to flag the existing answers that don't answer the question in its rewritten form (i.e. flag as "not an answer")? If not, what should, or could, be done instead?

If moderators already provided guidance about this in the following related Meta question, I apologize and would be glad if someone could please remind me what the conclusion was (it was a very long thread): Why are we challenging the premise rather than answering the question (question on potential sexist remarks)?. (Is thread the right word?)

(Not sure whether this other question is relevant: What to do about old questions whose answers are now invalid)

Answer 1

(takes off moderator hat, puts on regular user hat.) I support the following, as an approach that helps people keep their contributed content roughly as they intend it to look:

• Keep the question as is. Clearly, the current form is what the OP wants, and most of the answers do address the general question (even if some also refer to the specific example that has now been removed).
• Not forcibly deleting any answers. Authors of answers are of course still free to delete their own answers, if they want.
• Add a little bit of context to answers that refer to previous versions of the question, so that the answers make sense. For example, include in the answer a brief quote of the relevant part of the previous version of the question, when it is necessary to understand the answer. But keep the content of the answer as intact as possible while editing.

(If this approach wins substantial community support: Any user with edit privileges can help with this, and add a comment referencing this meta discussion. No need to flag for a moderator!)

Answer 2

There has not been a strong consensus on the linked meta post on what to do with the existing answers. Many people think that the answers should stay as they are; many argue that they should be deleted. A large number of users have argued that the edits to the question should be rolled back; a large number of other users think the question is better in this current form. It's hard to tell where people stand, also, because some answers to that meta post contain multiple points, and people voting on them may agree with some parts and not others.

Given this disagreement, unilateral moderator action seems inappropriate, so don't flag them.

If at any point you are able to show a strong consensus on meta for deleting them, then they could be deleted. A "strong consensus" would be if you

• Posted an answer here saying they should be deleted (and not containing any other proposals),
• and it got a large number of upvotes and few downvotes,
• and there was no competing answer suggesting to keep them, or a competing answer got much less support than yours.

Answer 3

This site is supposed to be about questions and answers, not feelings and emotions. That is the reason behind the very robotic up/down tick system and the UX of the Q&A. Input in, answer out. The site is not a free-range forum, if it was the comments wouldn't be wiped, especially those that provide more insight.

There are several sites this meta question would never come up and an example is mathematics. Imagine a person posting an equation that needs to be solved (overly simplistic):

2a + y * 4b -7c = r

Suppose they ask if "a" can equal "1" while giving the number possibilities to the other variables. And then people answer... No it can't be one.

Then the person starts rolling back one variable at a time until "a" could theoretically be "1". Well the question is nonsense. It should be rolled forward with the intent it was made. I am positive these edits on mathematics would be rolled back.

I am not sure what the deal is here. Is it the word "sexism" that keeps it from being rolled back? Has this site taken be nice too far? Was leaving a question alone given more concern than the answerers? I don't know. But common sense says you don't let the OP divulge less information as the question timer clicks down when they didn't get the answer they wanted. If the OP wanted the current question answered that would be a new question, which should be closed due to lack of information.

My proposal = Roll back the question to the point where it had the most information.

Answer 4

With the question being as strongly edited as this one, I would consider that the new version is effectively a new question to be treated as such.

A rollback with the suggestion to post a new question with the revised text is my recommendation (regardless of the quality of the original question).

Answer 5

It depends: there are rewrites, and there are rewrites. Rewrites which improve a clear, on-topic question (which is not an XY question) are fine. Adding (or preserving) a reasonable level of context is good, too. Rewrites which delete all context, mutate the question into something it never was, and thereby invalidate answers, are not fine. If the rewrites actively contradict each other, that's even worse.

That question is one of the worst examples of rewrites mutating the original question beyond recognition. The original #1 edit essentially said "Student objected to my teaching methods and not my male colleagues, I feel this is sexist". OP's #5 edit changed this to "IF A STUDENT COMMITS A CLEAR SEXIST BEHAVIOR IN FRONT OF YOU...". Totally moves the premise. The OP's own version was not clearly anything, other than an annoying weak student raising bogus objections, by email. Hence, solutions would presumably address those behaviors, not the chain of assumptions about motive. To do otherwise is an XY problem. "commit" is loaded language. Then #9 says "I never said that it /was/ sexist; it occurred to me (based on previous interactions with this student etc) that this was probably sexism.". Again, changes the question, and more missing context. We're now up to revision #12 on the mutating question, and we have all sorts of topics being injected like racism. The question has become a wish object for people to inject whatever perceived type of bias they want to conjecture about or analogize to. No relation to the original question! It's an irredeemable mess. (I hope it's no longer on HNQ and attracting wildly offtopic answers from newbies).

I have to say that the OP's behavior in edits #6, #8 and #9 damage her credibility and don't invite us to take her conjecture at face value. So we end up with three different conflicting versions of a he-says-she-says, which might in any case be irrelevant to addressing the original question #1 that was asked.

v#1 of the question should have been quickly put on hold for being unclear; or certainly by edit #5 which deleted the facts, changed the context and invalidated all the answers. If either of those had been done, this situation wouldn't have arisen. It's now way too late to fix the question.; it should belatedly be put on hold. Blaming the answers is an irrelevant sidebar.

Answer 6

As someone who arrived in that thread after the edits, the first few top voted answers verge on nonsensical.

I view the Question as the thing of paramount importance, and as such I'd support removing answers that were effectively made obsolete by refinement.

Answer 7

For what it's worth, good answers should usually be standalone in the first place. At worst they should become obsolescent, but never wrong. If crucial context is added to the original question, hopefully, the answers are self-explaining enough so as to sound like good answers to the more general question missing the crucial context.