Fairly often we get a kernel of what could be an interesting, on topic question buried in a not so great post. And fairly often the OP disappears after posting without editing or responding to comments asking for clarification.

Needless to say we end up with more than a few dead closed questions that at their core were asking about something worthwhile.

The first option that comes to mind to perform salvage operations are radical edits, but these tend to be a little tricky around here, because you run the risk of putting words in someone's mouth. Inferring what their goal is can be particularly troublesome.

What I would like to start doing with some of these interesting, yet abandoned, questions is to borrow the interesting kernel and use it to create a self answered question.

This has the advantage of avoiding the problems associated with radical edits, while preserving some of the interesting topics that come to us in poor questions. It could also end up being something of a teaching tool if we took it a step further and cross-linked the previous closed questions with their salvaged counterparts...

Figured it would be a good idea to toss this out to meta before I just started doing it, so what do you folks think? Good idea, bad idea, am I missing anything else that should be considered on this topic?

  • hmm ... what's a self-answered question? What does it mean to cross-link a closed question with their salvaged counterparts? Could you give a concrete example, maybe using a currently-closed question? Commented Mar 6, 2018 at 23:21
  • I like the idea, but it would need to be situationally applied. Commented Mar 6, 2018 at 23:27
  • 4
    @D.Hutchinson A self answered question is what it sounds like; a question you ask and then answer yourself. Cross-linking is just a simple comment posted beneath the closed question linking to the new question.
    – apaul
    Commented Mar 6, 2018 at 23:37
  • 1
    Why do we need to "rescue" these questions? What's the difference between rescuing a question and asking a question just to create a platform for a ranting answer?
    – sphennings
    Commented Mar 7, 2018 at 1:46
  • 4
    @sphennings Way to assume good intentions there friend. How many questions have you asked for clarification on, when you're fairly certain you know what the question is, but feel it would be inappropriate to edit?
    – apaul
    Commented Mar 7, 2018 at 1:51
  • 4
    @sphennings it's not so much about rescuing the question as it is about rescuing an interesting topic from a poorly worded question.
    – apaul
    Commented Mar 7, 2018 at 1:53

3 Answers 3


Potential for misuse of the practice

Once starting this practice, is there a potential for people to take it too far?

For example: ThrowawayUser87 asks a long, convoluted, multi-question, "what should I do?" question.

I see the "kernel" of a question I feel I could answer and create a self answered question. However I pulled a "kernel" that is still not a great for for IPS. I've now just added to the mess generated by the question.

Not an intended use of self-answers

In my mind the self-answer exists to allow somebody to publicly document a real issue they encountered and answered/solved themselves. This serves the aim of allowing others to come to SE with the same problem later and already have it answered. While hypothetical questions are also common across SE, they are usually questions asked to others who may have less than hypothetical knowledge for the answer. Hypothetical questions with self answers feels like a slippery slope for creating questions for the sake of answering, not for the sake of solving a person's future problem.

Related SE.meta answer on what makes a good self-answered question

Self-answers are tough to get right anyway

Its hard to not craft the question to the answer you intend, even when you are asking about a concrete problem you had, using hypotheticals increases this risk as there are no real details to anchor to.

  • 1
    There are potentials for the misuse of nearly every aspect of the site, most are somewhat self teaching, others are handled by community moderation, and the worst cases can be handled by flagging for moderator attention. In this case, if someone were to self answer a poor question, they'd face down votes for the question and the answer as well as close votes, and possibly flags.
    – apaul
    Commented Mar 6, 2018 at 23:47
  • 1
    I agree that community would handle this, but in the case of "rescuing" part of a question that already got closed by the community, I feel we are leaning on the community a bit too much Commented Mar 6, 2018 at 23:52
  • 3
    After giving this some more thought, the list of people who participate on this site who I would trust to do this properly is vanishingly small. So, you're probably correct in thinking that it would likely be abused.
    – apaul
    Commented Mar 7, 2018 at 20:44
  • "it's not so much about rescuing the question as it is about rescuing an interesting topic from a poorly worded question." __ Your idea is quite good if it doesn't insist on "self-answered" @apaul: what about creating a meaningful question from that kernel, linking it to the original for context, and allow it to be answered by members, just like any other question? Commented Mar 7, 2018 at 21:14
  • ^^ @EnglishStudent yes. If we limited the "rescuer" from answering, it would solve my issues that revolve around self-answers, but it also removes the motivations of these "rescues" Commented Mar 7, 2018 at 21:34
  • 2
    The rescuer can possibly be one of the answerers (to retain the motivation) but there is no need to insist on a formally self-answered question. That way it will receive multiple answers, and if the rescuer's answer has the defects you anticipate then it will not be voted as the best answer. Commented Mar 7, 2018 at 21:37

In theory, it sounds nice, but in practice, I don't think it will take off.

I think, since we have a lot of active users and HNQ visitors, it would simply be a matter of time that someone else comes up with a better question with the same or similar subject matter as that of those throwaway questions.

So, my take is.. No such rescue efforts necessary.

  • 1
    Yeah, I recall a user asking a "rescue" question after the original OP refused to improve theirs (basically, new user asked a duplicate but with the requested details) - and was immediately questioned about "is this actually a real problem you are facing?" (this meta thread). So in practice there seemed to be community resistance to the idea of inventing a better version, rather than waiting for someone to ask it better "naturally".
    – Em C
    Commented Mar 7, 2018 at 14:29
  • @EmC You misinterpreted the point of that comment. If you follow that situation to it's conclusion it ended up being a reasonable example of what I'm talking about doing here.
    – apaul
    Commented Mar 7, 2018 at 20:35
  • @apaul I had interpreted that comment as a warning shot, e.g. if the user had instead responded "No personal experience, but I thought it was a good question" I would expect some backlash. So I'm not quite sure what you're getting at, do you mean you agree there should be conditions on who asks the "rescue" questions?
    – Em C
    Commented Mar 7, 2018 at 20:47

Maybe, after a certain time period, those posts could be moved to the a brainstorm forum. And maybe someone could rewrite them and if they got two more votes as quality questions, then they get included in the knowledge database.

I do think that Stack Exchange will have to make a call overall whether their sites are about helping specific people with specific problems or general people with general problems. I suggest going the route of having potential duplicates instead of being too general. General, in practice, is not all that helpful. Everything happens specifically.

  • The hope is that the specific question ends up helping the general population. Granted that's a "hope", but it's the general goal of Stack Exchange.
    – apaul
    Commented Mar 7, 2018 at 6:04
  • @apaul - If it was up to me that's what I would choose. I was in app dev 20 years ago using stack exchange. While not all posts were on point, sometimes i could patch together 2 or 3 to get my answer.
    – Stacy
    Commented Mar 7, 2018 at 6:13
  • Stack Exchange has only been around for about 8-9 years. Commented Mar 15, 2018 at 9:57

You must log in to answer this question.

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