I'm sure most of us have seen questions with issues (off-topic, not constructive, etc) which, with a little editing, could be good questions.

In theory, the solution to these kinds of questions would be to close them, give the OP an opportunity to improve via editing, and then reopen them. If editing has truly improved the question, then upvotes after reopening would balance out any initial downvotes. (I've seen this happen many times, so I know it works.)

Except, due to "lounging" or other factors, some questions are swamped with an unusual number of downvotes (-10, -20 or more). Before the question can be edited, the OP loses heart (or recognizes that the chance of regaining their rep is basically non-existent) and the question is simply deleted (when possible) or abandoned (as hopeless).

Is this a common issue? Or, alternately, is it uncommon but "bad enough" (given SO's mission) that it should be addressed? SLaks found one such instance which has since been deleted. I've seen one or two, but I'm not sure what the statistics are.

If so, is there a way to protect these kinds of questions without significantly curtailing the very important task of voting? It's a sensitive issue because any solution would involve the restriction of voting, and that is not a light subject.

One solution proposed a couple years ago was to assign closed questions to community. Votes after closing would thus not affect the author, giving them a chance to either edited the question or for the community to delete it (if it's not salvagable). While that feature-request didn't get much traction (not even a ), it would seem to go part way toward solving the above problem. It, admittedly, would lead to other potential issues.

Perhaps a different possibility would be to put a short voting hold on closed questions to give the author a window to improve the question before it accrues "too many" downvotes. A closed question restricts answers partly for this reason and I don't think it would be a huge leap to extend this to votes. Users who do not have the reopen privelege could always use flags if they felt strongly about the question. Perhaps an hour or two would be sufficient for most cases.

Then again, it is also possible that this is so uncommon and/or relatively unimportant that any automatic or one-size-fits-all attempt to address it, including my suggestion above, would actually cause more harm than it would fix. If that's the case, perhaps mods could be given the ability to restrict voting on a question where they see an issue. (For example, Chichiray voted to delete the question SLaks commented on "because the community made a mess of it". Wouldn't it have been better for a mod to protect the question, rather than deleting it?)

What are the thoughts on this?

  • 3
    Note that "pile on" is a term generally used for people downvoting on a question just because it already has a bunch of downvotes, without even reading it or considering it's quality. (The same works in the other direction with upvotes as well.) You don't seem to be describing that behavior; you're generally describing posts that really are of low quality and that are actually being evaluated by those doing the voting.
    – Servy
    May 21 '13 at 16:25
  • 4
    @Servy: I would add that my general observation is that the number of downvotes a new post receives is usually consistent with the post's quality (or lack thereof).
    – user102937
    May 21 '13 at 16:27
  • 3
    I wonder if "lounging" has thrown off the question ban settings. If the downvote thresholds for a question ban are set assuming the questions get normal amounts of traffic, and "lounging" skews that, users may be being banned in situations outside of the system's original intent.
    – Jeremy
    May 21 '13 at 16:27
  • @JeremyBanks See the fourth paragraph in my answer.
    – user102937
    May 21 '13 at 16:27
  • 2
    -1 I disagree with the statement (or recognizes that the chance of regaining their rep is basically non-existent). OP improves the question to a) get answers b) avoid a question-ban. If he is in it for the rep it is much easier just to answer a couple of questions.
    – user000001
    May 21 '13 at 16:31
  • @Servy - That's the scenario I think I'm describing. A bunch of users collude with one another to downvote a question, and that unusually large downvote in a short period of time attracts attention and creates a snowball affect. This would be an "outside the norm" scenario where a new user isn't given the time to edit the question into something "appropriate" to our site. May 21 '13 at 16:31
  • 1
    @user000001 - true - perhaps I should rephrase as "believes that the chance of getting positive attention (an answer) is basically non-existent" May 21 '13 at 16:33
  • 1
    @Cyborgx37 No, that's an entirely different behavior entirely. "Pile on" downvoting is going to a question, seeing it's a -10, and downvoting without reading because of it's score. What you're describing is a question being linked from somewhere, thus generating a large number of views and, being a very low quality question, being justifiably downvoted by a number of those users. These are very different effects.
    – Servy
    May 21 '13 at 16:34
  • @Cyborgx37 could you find any posts that are heavily downvoted as a direct result of "downvote-collusion" and not just being a bad question? (SLaks post does not qualify since the question had -9 before it was posted in chat).
    – Aboutblank
    May 21 '13 at 16:48
  • @Aboutblank - unfortunately I haven't found anything yet. There are some posts on meta about these kinds of question (not specifically "colluding", though I know of nothing to discouarge such behavior). See meta.stackexchange.com/q/168563/191410. Animuson's answer to that question is that a high number of downvotes should signal that the question is really horrible. But in some cases they aren't horrible - users downvote because the question is a little bit bad (but could be improved) and then start piling on each other in a very short timeframe. May 21 '13 at 16:54

Voting is deliberately anonymous. We don't require people to justify their downvotes, so by definition, votes can be used any way you wish, so long as you don't commit voter fraud (just like the votes you use at the polling booth). To do otherwise would have a chilling effect, and distort the voting system. So we can't regulate the way people vote, even if we wanted to.

Downvotes are not really meant for the seasoned users who know how to ask a proper question. They are meant for the new users who haven't taken the time and effort to formulate a proper question. If you're going to inhibit downvotes for new users, why have them at all?

Downvotes are part of the algorithm that Stack Overflow uses to prevent new users who have failed on multiple fronts from posting again. Protecting questions against pile-on would break that algorithm.

The number of votes that a question gets is in direct proportion to the number of views a question gets. The number of views a question gets is a direct measure of how much time is being wasted by the community on a bad question. In that sense, the downvotes a bad question gets are entirely justified.

So the only remaining consideration is whether pile-on downvotes are a problem for the new user specifically. I would say that, if the user's well-being and self-confidence is dependent on not receiving downvotes, then yes, it is a problem for him. But that is ascribing a meaning to downvotes that doesn't exist; namely, that downvotes are a personal attack. They aren't.

  • 1
    +1 Thanks - that is a helpful perspective. Although I'm talking about edge-case scenarious where questions aren't that bad - that is, a little editing could fix it - but due to some fluke they appear to be receiving heavier than normal negative attention. May 21 '13 at 16:27

You must log in to answer this question.

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