A question that was asked recently concerns synchronizing words. Apparently it had been assumed to be on a different and much more trivial topic, graph traversal, and as a result, entertained a shower of downvotes. The people who downvoted this rather short question, not bothering to read its five lines carefully enough, are probably not going to come back and apologize.

How do we discourage such behavior? Who protects the people who ask questions?

  • 4
    $\begingroup$ I undid my downvote and apologised, despite the poster's repeated attacks and insults. $\endgroup$ Feb 12, 2019 at 1:05
  • $\begingroup$ @DavidRicherby I’ve certainly got no complaints against you. $\endgroup$ Feb 12, 2019 at 2:25
  • $\begingroup$ “a shower of downvotes”? I'm coming after the storm, but there's only one downvote now. How many were there? $\endgroup$ Feb 12, 2019 at 7:31
  • $\begingroup$ @Gilles At least three. $\endgroup$ Feb 12, 2019 at 7:32
  • 8
    $\begingroup$ The original question is a problem dump. Downvotes and close votes would have been the appropriate response. The level of the problem is irrelevant. $\endgroup$
    – Raphael
    Feb 12, 2019 at 7:42
  • 1
    $\begingroup$ This old discussion on meta.se seems relevant. The discussion is a bit broader than downvotes, but covers that as well. $\endgroup$
    – Discrete lizard Mod
    Feb 12, 2019 at 9:11
  • 1
    $\begingroup$ @YuvalFilmus Exactly three, I think. $\endgroup$ Feb 12, 2019 at 10:43

2 Answers 2


First of all, I agree with you that downvotes on questions that are fine after minor improvements should be avoided. It is better to close vote/flag instead and even better to edit to improve the question or help the asker do so with a constructive comment.

But I don't think trying to discourage people to downvote is the solution. I think it is better to acknowledge that looking for the best in posts that are usually rather poor requires a lot of patience, which not everyone can muster. This does not justify this behaviour, but it might explain it a bit. Also note that mistakes will happen even if everyone perfectly agrees, so there is no way to completely prevent these situations.

So, what can we do? I have a few ideas:

Provide more guidance to reviewers

Most of the questions vulnerable to being misunderstood appear in the first posts review queue. If anyone should 'protect' those who ask these questions it should be first posts reviewers. In this case 'protect' means to improve the quality of the post and/or be generally supportive of the asker.

Approaching reviewers can be done in multiple ways, we could make a meta post where we can point them to, mention in comments if they can improve or use chat for a more individual approach.

Upvote more

A lot questions seem to receive at most 3 upvotes during their active period. This means that one or two downvotes have a lot of impact on the post score. If questions receive more votes, the impact of 'bad' downvotes decreases.

The problem here of course is how to achieve this. However, it doesn't take many people to upvote questions. Maybe you (yes, you there, reading this) should upvote more?

Close more

This is just speculation, but one reason why people choose to downvote instead of flagging to close could be that questions do not get closed quickly. This means that, from their viewpoint, flagging has almost no effect even after waiting for a while, while close-votes have immediate effect.

Note that I could be wrong here and that questions put on hold are more likely to get more downvotes, but this is something we can consider.

  • 6
    $\begingroup$ Let me second the request to upvote questions. Most questions get hardly any upvotes. I upvote (almost) every question that I answer. $\endgroup$ Feb 12, 2019 at 14:47

I didn't see the question you mentioned before, but when coming across a problem dump I am quite often guilty of voting to close and usually downvoting as well, without leaving a comment. Nevertheless, I try to sense whether the question really is a verbatim copy (in that case, I don't have much sympathy), or whether the questioner is more genuinely asking a "real" question but just is unaware of the practices of the site. In that case, I am almost always too lazy to paste one of the standard answers we have, but try to ask for clarification.

I try my best to view the site as a dynamic beast, meaning that I understand that edits are made, external links might die, better answers might pop up (so the accepted answer can change as well), and so on. So if I (or anyone) has downvoted a question or an answer and it gets improved later on, I try to notice that and then ask myself "OK, in its current form, is the question/answer worthy of a downvote?". If not, I upvote.

The above is my just my ideal, and the site certainly doesn't always make it easy to notify you about such changes. On the other hand, new questions stay within the first two pages for a bit of time, making it more likely that I return as well. But I very much sympathize and agree with the point you have raised. Downvoting is perfectly appropriate (and should not be taken as a personal attack, although it can easily feel as such for a new user), but at the same time we should be willing to reconsider our votes if the question/answer is improved.

So definitely, upvote more if any question/answer deserves that. Downvoting can also be appropriate, but please also try to be willing to reconsider your opinion later, don't use it to punish a user for making mistake. At the time of the writing, the question you linked still has one downvote, which the downvoter might be happy to remove if alerted to it.

  • 1
    $\begingroup$ My point is that this particular question appears to be a problem dump only on first glance. It would be nice if before downvoting you take the time to make sure that the question really merits downvoting, especially if it's by a new user. $\endgroup$ Feb 22, 2019 at 9:09
  • 1
    $\begingroup$ @YuvalFilmus Definitely agreed. My point was that even if you did downvote, it would be nice if you could revisit the question after the edits to reconsider. $\endgroup$
    – Juho
    Feb 22, 2019 at 9:12

You must log in to answer this question.

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