I've had two instances today where I've come across bad questions. By bad questions, I mean:

  • Code not included
  • No context
  • poor spelling, punctuation, or really bad sentence structure (irrespective of the user's first, second, or third language)
  • Comments from the user that don't address any of the comments where the user was asked to improve his question

Here are both examples.

Example #1: JavaScript “if” and “else” error?

In the first example that I came across, I left a comment for the user explaining the problem with their question:

We need to know: 1) What the actual problem you're seeing is 2) a short, self contained, complete example that reproduces the problem, 3) what you expect the output to be. (Hint: You should be including the HTML as well)

and I then closed the question, in line with what closing actually means, and in line with what I've said I would do as a moderator.

I received a good bit of flak in the comments for that action.

Example 2: Can't read file with streamreader?

In the second example, the user had many of the same issues (in fact, those issues I listed above where a cross-section of the problems that both questions had some of).

Both questions, I believe, are poor questions. They're questions that are almost too localized by definition: Written in such a way that no future user will ever be able to find that question in connection with a problem they have, and not descriptive enough to be helpful to anyone else if they somehow do find the question. They're also 'not a real question' because not nearly enough information is given to figure out the problem (unless you're a psychic debugger. In fact, I tried to edit the first post a few separate times only to hit 'cancel' because I had no idea what the user was actually trying to do.

My question to the community: Should these questions be closed quickly, providing an incentive for the Original Poster to improve their questions; or should they stay open and just be commented on, in the vain hopes the user will improve their question?

It should be noted that in the first example, the user never came back to improve his question, even after comments were left for him to do so by multiple users. I could probably dig up many more examples, but these are the two on my mind because they just happened today.

  • Jeff had a pretty effective argument for that, talking about SO's "big city" problems. That was around 2000 questions/day. It's now over 6000/day? – Uphill Luge Feb 1 '13 at 2:27
  • Yeah, and Joel was presenting a counterargument in the most recent podcast (some of that may have been Joel playing devil's advocate). – Michael Petrotta Feb 1 '13 at 2:49
  • Maybe consider to not take the fully responsibility to close alone the questions and just vote to close as the rest (I do not know if this is possible for you) and by that way you left and some others to get part of the responsibility to close it. – Aristos Feb 1 '13 at 7:42
  • Couple of quick notes: you can find tons of examples where no improvements are made because you're paying attention, but the ones where feedback does help oft go unnoticed. One other note in this case is that presumably someone could have edited the question to improve it, since it seemed like at least one answerer solved it. My opinion is that anytime you can edit to fix (not saying you could here), you shouldn't be closing. – Jaydles Feb 1 '13 at 14:25
  • @Jaydles I'm a big fan of editing (since you have the numbers, you can see how much I've edited). I always try to edit when I think there's any chance the question can be fixed or if I can somehow discern what they're asking. In this case, I couldn't figure out what the heck they were asking, even after reading the answer. – George Stocker Feb 1 '13 at 14:39
  • I think it's worth contrasting the two examples here, @Jaydles: in the first case, although several people made rather superficial edits, no one has yet bothered to actually state a problem - it remains a case of, "there's a bug here, which I won't describe - find it, and then fix it". In the second case, both the asker and another editor have worked together - taking into account feedback from the comments - to correct the (originally fairly damning) issues with the question. The latter is ideal: any question can be fixed via edits, but not all editors can fix every question. – Shog9 Feb 1 '13 at 17:08
  • @GeorgeStocker, (& Shog) I should clarify that I was in no way criticizing George's editing/commenting instincts - they're dead on. – Jaydles Feb 1 '13 at 18:17

It should be noted that in the first example, the user never came back to improve his question, even after comments were left for him to do so by multiple users.

And even the guy who posted an answer didn't bother to edit the question to state the problem he was solving. It stands there now, with no error quoted, no effects described, just some code and... "What's wrong?"

This isn't unusual. You can't hang around waiting for someone to respond, because many - perhaps even most - of the folks posting half-baked questions won't ever do so.

And as a moderator, you have hundreds of other questions begging for your attention, some of them asked by folks who do care enough to respond.

So close and move on. Frankly, even leaving a comment is more than should be expected of you - the description for "not a real question" pretty much sums it up.

There've been a lot of requests over the years for tools that would make closing a bit more responsive to folks who did come back and improve their question without requiring someone to sit and watch it... I think such a tool could come in handy in some of these cases.

But until such a thing exists, don't let folks waste your time - Stack Overflow doesn't ask a whole lot of the folks asking questions, but requiring a complete, understandable question isn't something anyone should really disagree with.

  • Devil's advocate (but only sort of, because fundamentally I agree with your message): some folks come back and play the "well, I know what I asked and someone answered it, so clearly it's answerable" game. A comment asking for clarification and/or providing instructions for the next steps to take to get the question reopened can help mitigate the fallout from that. – Adam Lear Feb 1 '13 at 3:26
  • If someone comes back, you're already kinda ahead of the game. While they're here, they can fix the question... (I don't have a problem with leaving comments if the issues with the question are obscure, you're bored, or you're aiming for the pundit badge - but in George's first example, it's pretty blatant - I kinda wonder if the comment didn't end up drawing more attention away from the problems with the question itself) – Shog9 Feb 1 '13 at 3:27
  • Is there not something that exists already? Closed questions that get revised are automatically put in the re-open queue (no idea if all of them). Putting something about this in the closed box might alleviate some of the comments. – ben is uǝq backwards Feb 1 '13 at 8:49
  • 1
    I don't agree that a comment was generous. While I wouldn't advocate requiring them, I think if you're telling yourself that closing is usually an just an opportunity for the poster to improve, you should be willing to tell them how. – Jaydles Feb 1 '13 at 14:21
  • @Jaydles: although it may not be entirely obvious, folks can and do address moderators who've closed a post in comments on that post, and a notification results from this. I tend to think this is a more efficient form of communication, particularly when there's a pretty good chance no one will respond. Note that in this particular case even though George did proactively comment, including specific instructions... No one, not the asker, not the answerer, has bothered to fix the post. That's a waste. – Shog9 Feb 1 '13 at 17:03
  • 1
    @Shog9, I don't follow - are you saying it's better to close without comment, and let the closee comment? I'm saying it's better if closures generally came with comments when possible (which we both agree George did). – Jaydles Feb 1 '13 at 18:16
  • @Jaydles: I'm saying that commenting to elaborate on the problems with a question, while nice in theory, can (and in this case did) end up being a huge waste of time. Not Constructive is probably an exception since it's so open-ended, but NaRQ is fairly self-explanatory. – Shog9 Feb 1 '13 at 18:24
  • 1
    @Shog9, I fear we're going to need to clean up our own comment thread soon, but I don't agree. I do not think that most of the reasonable people who get NARQ know what they're supposed to do to improve their question. – Jaydles Feb 1 '13 at 18:44
  • 1
    @Jaydles: then we need to improve the guidance in that close description. – Shog9 Feb 1 '13 at 19:05

I'm well known as a deleter and also closer. I believe this question should have been closed, quickly.

However, I do believe that we're driving away certain users who do not understand [so] well enough to understand what closing actually means. These users take it to mean, "I'm not wanted here", or at least, "I won't bother to clean up my question".

Rather than expect the closers to leave a helpful comment, I wonder if they couldn't leave a link to something like a specialized version of the "About" page. Such a page might be able to show them, via animation, what we expect them to do. Start with an example of a question like the referenced question, show how someone can ask for clarification, then show the OP adding details, and other users editing the question for clarity and formatting, etc. Show an actual collaborative process for whipping questions into shape.

Be certain to show that the question gets closed while it's being cleaned up.

Also, show how the improved question gets better answers.

  • Well known deleter/closer.. :) – Krishnabhadra Feb 1 '13 at 5:16
  • +1 Your idea of a specialised about page is brilliant. Perhaps you should post that as a feature request. – Duncan Jones Feb 1 '13 at 9:03

You must log in to answer this question.

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