This was a particularly crappy question a few days back. Now there is not even a trace of it. This question did not get any answers, but got at least 3 (meaningful) comments. But I've seen a few question in this general theme, at least once every few days. The general theme of this type of question is (which I call coding by crowd sourcing):

  1. Try writing some code or try some example from some site, and when something goes wrong
  2. Don't try to troubleshoot it, post it on Stack Overflow
  3. Find out you're doing something embarrassingly wrong (such as compiling the wrong file, compiling in the wrong directory, not compiling, not saving before compiling, etc), and remove the question

Just in case you think this is a fluke: there are others here, here, here, here and here all from the past few days.

I think that since the system is too lenient letting posters immediately and at no cost to them remove a question, this is a self perpetuating problem. One that encourages coding by crowdsourcing and wasting of the community's time.

Can this be handled by the system better? The real question is: would it be possible to harshly penalize questions that quickly get closed (voluntarily or not), and reward volunteers that make that happen?

My suggestion is, when someone posts, open an "escrow" with 10 rep points (maybe more). For every 10 minutes your question survives, you get 1 rep point back from escrow. When and if it is closed all users that voted to close, and all commentators, that helped point out the crappiness of the question, split the escrow.

  • 11
    If someone keeps doing that, they'll get a question ban. Problem solved. – Mysticial Aug 9 '12 at 4:47
  • @Mysticial With hundreds of thousands of potential users, you'll exhaust your community before you solve the problem. – carlosdc Aug 9 '12 at 4:49
  • 5
    To clarify Mystical's remark a bit: users that post many question and get almost no upvotes will be automatically blocked. Having a large ratio of deleted questions will already cause this ban to be triggered more quickly. – Jeremy Banks Aug 9 '12 at 4:50
  • 2
    I +1'd this because of your concern for the quality of questions, and our shared distaste of the questions you refer to here. I agree with the answers and comments you've received here, though. – Andrew Barber Aug 9 '12 at 16:48

Everybody has the right to fail, nobody should be afraid to be wrong. The ability to remove a question no one needs is an important part of that.

And if a question is really interesting … it will come back in a better form.

  • At the risk of being the minority here. I think the system needs to somehow protect the volunteer time of the community, and probably should not promote coding-by-crowdsourcing. – carlosdc Aug 9 '12 at 4:59
  • 9
    That's why questions that have been answered are so hard to delete, @carlosdc - and why folks who repeatedly delete their questions can find themselves unable to ask more. But keep in mind: a crappy question deleted by its author is a crappy question that doesn't have to be closed, and deleted, by the volunteers in the community... – Shog9 Aug 9 '12 at 5:01
  • 1
    @Shog9: OK I get that. This meant to be a constructive post (one that so far has -6 votes) about more things that can be done. In such a way that potential posters learn from the experience of others. – carlosdc Aug 9 '12 at 5:23
  • 3
    @carlosdc Voting on Meta is used to show agreement / disagreement, it's (usually) not a comment on the quality of the post. Don't worry about the downvotes. – yannis Aug 9 '12 at 5:30

This could be solved by people posting their answers as answers instead of comments. Once a question has answers (or a single answer with a score > 0) you can't self-delete it.

On the other hand, if it happens with the type of question you seem to be describing it's no great loss either. Regarding the title of this question, if people want to remove their crappy questions, my initial reaction is "good"--we've got more than we need of those. Answers-as-comments tend to occur when the answerer is guessing, which tends to be when the question was poorly asked and lacking detail, which is the type of question that should probably be closed anyway as either "not a real question" if the detail is really lacking, or "too localized" if it's a mangled code block with the words "not working, please help" that's not going to help any future searchers.

  • What I guess I'm saying is: the system is too lenient and not protecting volunteer time enough. If you waste say 1/10th of an hour of 4 volunteers, because you posted a crappy question, your rep should take a hit. A nasty hit. I would also let the question be there for a week or so, so other posters take a hint. – carlosdc Aug 9 '12 at 5:19
  • 2
    I think the site does as much as it can in that regard--there's a bad question filter to prevent the worst from entering the system at all, question downvotes are -2 rep, and a pattern of negative score or deleted questions results in a question ban. Leaving these garbage questions lying around does more harm to the site than it would do to the poster, if they care at all. – Brad Mace Aug 9 '12 at 5:24

This is a real problem. Maybe an unpopular one.

Here's another one from a few days back, same symptoms: wrote some code, ran into a problem, did not troubleshoot, posted on SO, wasted 3-5 minutes of 3-5 highly competent volunteers, with a:

while(1){fputs(" ",f);}

and with some particularly outrageous attitude, which can be summarized in: duude, sometimes when I run this and open the output with my l337 TextMate, it is a 2500 MB file!, WTF is EOF broken? I had to redefine it!

Observe that by monitoring a medium traffic tag (C) for 2 hours I saw at least 2 of these. I might want to experiment by monitoring C# or Java or PHP.


  • The system is perfect at filtering this: No it isn't, see linked questions
  • If someone keeps doing that, they'll get a question ban: well that solves the problem for one person, and does so after they become a problem. But what about the remaining thousands of potential posters?
  • SE is doing a lot to protect volunteer time being wasted on crap questions: they may be doing a lot, they can certainly do more.
  • Everyone has the right to fail: yes, everyone has the right to fail, but the system should not aid in wasting valuable volunteer time while you fail in public.
  • We're all well aware of the bad questions that come in. Unless you have a concrete suggestion though, I'm not sure how this is going to be constructive. – Brad Mace Aug 9 '12 at 20:14
  • @bemace I've edited the answer, about the constructiveness part of my question. – carlosdc Aug 9 '12 at 21:46

You must log in to answer this question.

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