Closure while writing an answer – can it be delayed?

Question

I have fairly frequently been in the situation of being in the middle of providing an answer to a question when it is closed, leaving me no option but to discard my contribution or hope that the question can be re-opened.

It can be very frustrating to put in the mental effort to prepare an answer only to have it summarily rejected. Of course, this affects frequent contributors more than others.

I wonder if it is possible to let such "work in progress" be entered into the record when this happens.

Answer 1

The single direct effect of closing questions is that it prevents answers, for reasons such as:

• To give the asker the opportunity to clarify their question before existing answers get invalidated.

• To prevent the inevitable endless discussions answers to some questions provoke.

• To avoid the broken-window effect, i.e., new users getting the impression that they can get answers to certain questions here.

• To avoid redundant content, voting, etc.

• To avoid that askers commit pointless real-life mistakes by blindly heeding the inevitable wrong answer certain questions will attract (applies to many questions closed due to depending on individual factors).

• …

In short, we close questions because we think that answers to them would be bad. Circumventing this defies the entire point of closure. These questions should not be answered on this site, in the comments, and at times nowhere on the Internet. Closure needs the five votes of high-reputation users (or one vote of a diamond moderator). If it happens, there usually is a good reason for it.

So, the best way to avoid your problem is: Do not answer questions that should be closed! Roughly 16 % of your answers were on questions that were later closed. This is a rather high number, among the highest for high-reputation users on our site. (For comparison, it’s 6 % for me. Here is a Stack Exchange Data Explorer query for this.)

That being said:

• If you think that a question was wrongly closed, you can argue against it (on Meta Academia if it takes longer) and vote to reopen.
• If you think that there is a valid question buried under a question, edit it and vote to reopen.
• If you think there is a valid, relevant question to which what you just wrote is the answer, write that question and self-answer it.
• If the asker needs to clarify something, comment to let them know what and wait for them to do it.

While this happens, just save your answer in a regular file.

Answer 2

I agree with Wrzlprmft's analysis, particularly for very "off-topic" questions (too broad, about content of research, etc.). But as another user with ~16% of answers on closed questions, let me offer another viewpoint.

Consider a question like "I'm worried I ruined my life and could really use help, let me post my super long, specific-to-me question" (example). Such questions should obviously be closed as they violate our community norms; on the other hand, it is very unwelcoming to just downvote and close after someone took the time to ask for help. There is really an opportunity to help someone here, even if we know the question will likely be closed and we won't get much reputation in exchange for a good answer.

I know it's a bit heterodox, but in such cases, I have no problem with answering a question knowing it likely will (and should) be closed. As such, I would not be inclined to interrupt users who are in the middle of answering such questions.

Answer 3

This post on the main meta explains why during the next 4 hours of being closed new answers can still be submit, and how to do this. This can be a good balance point for the disciplinary - flexibility issue, because we want two conflict things at the same times.

Not intentionally want to advertise, but I happen to discuss about the way to find the balance point in my article A theory of perspective:

The secret to be flexible without having to compromise stability (e.g. violate rules) is to notice the struggle of others, because mitigating others' struggles is the reason why stability is born at the first place. Instead of worrying whether a behavior is moral or legal or not, we should check its ability to reduces the cognitive dissonances of everyone in long term. In the case of having conflicts, no one will feel that they have to compromise, but sees how their worry is maximally satisfied before it is actually satisfied.