Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 158 Stack Exchange communities.

What is meta?
Here's how it works:
  1. Any Stack Exchange user can ask a question
  2. The community provides support, votes on ideas, and reports bugs
  3. Your voice helps shape the way Stack Exchange operates

I've seen a number of cases where a question is closed as a duplicate because the closer doesn't take the time to actually understand both questions, and they jump to the conclusion based on superficial similarities that the question is a duplicate.

Is there any way to discourage this?

Perhaps there should be an appeal process, where the questioner (or anyone else who notices the error) can lay out the facts to explain the distinction between the questions, to a group of users who are much higher-level than those who voted to close, and if it is found that the closers acted rashly then they would be severely penalized, to discourage such rash and careless behavior in the future.

Update: here's a specific example of a question that superficially appears to be a duplicate, but in fact is not: Does PHP offer any way to remove one array from another?.

share|improve this question
5 close votes are needed to close a question. I think that is evough to have 5 people look over it. – juergen d Oct 15 '12 at 15:31
If two questions appear at first glance to be the same, what makes you think the majority of people will take the time to actually understand the differences? It takes a few seconds to look at things and say "oh, those are the same!" and vote to close. It takes actual time to really understand what is being asked. People almost always take the path of least resistance, unless there is some disincentive. People who do realize they are different will not take any action, so there is no natural force working against the careless closers. This proposal would fix that bias. – iconoclast Oct 15 '12 at 15:33
There is an appeal process. Any 5 voters can vote to reopen. Or a moderator can be flagged with a custom explanation. – Bart Oct 15 '12 at 15:33
@iconoclast Do you have a specific example? – George Stocker Oct 15 '12 at 15:33
@GeorgeStocker: Yes, I can give a specific example. – iconoclast Oct 15 '12 at 15:34
@juergend, once in a while, "five closers are needed" is what seems to be driving lower attention. I'm embarassed to say it, but I'm a LOT more careful casting close votes now that my diamond means they're instantly enforced. – Jaydles Oct 15 '12 at 15:35
@Bart: what makes you think 5 people who are unrelated to the question would care enough, and take the time to do the work to distinguish between them? I think it's pretty unlikely in most cases. – iconoclast Oct 15 '12 at 15:36
"if it is found that the closers acted rashly then they would be severely penalized" ... hell no. That would only serve to discourage people from voting to close altogether for fear of repercussions. As a result leaving much more crap around. I would take the occasionally wrongly closed question over that any day. – Bart Oct 15 '12 at 15:36
I'm assuming OP is referring to this:… – John Dibling Oct 15 '12 at 15:37
@iconoclast Not all that unlikely. I don't have figures, but if a question is truly closed unfairly, there is a large chance of it being reopened I feel. I see examples of that coming along daily. – Bart Oct 15 '12 at 15:38
@GeorgeStocker, here's an example from superuser where it appears the "off-topic" closure came from 5 people who saw "tweetdeck" and assumed the issue was a webapp, etc:… – Jaydles Oct 15 '12 at 15:38
@Bart: what makes you think it is only occasional. If the only "appeals process" is needing 5 votes to reopen, then there is effectively no appeals process. People have an incentive to close because they want to clean up (and I'm guessing they get some points for doing so too) but there's no incentive for 5 people to reopen. A question is asked by only one person, and unless it's a wildly popular question, no one will even notice it. – iconoclast Oct 15 '12 at 15:40
@iconoclast You will have taken note of my second option for the OP as well I assume. And I didn't even mention bringing the supposedly unfair closure up here on Meta, which would result in some additional attention for the question. There are plenty of options. – Bart Oct 15 '12 at 15:41
"you are not as likely to see the distinction if you didn't ask the question" ... If that is true, then who on earth is going to distinguish whether or not the closure what too careless? The OP? Because your "much higher rep" users would disqualify as well based on that argument. – Bart Oct 15 '12 at 15:43
So? Ask for it to be more prominently displayed. That I could perhaps support. – Bart Oct 15 '12 at 16:07

Perhaps there should be an appeal process, where the questioner (or anyone else who notices the error) can lay out the facts to explain the distinction between the questions

There is. You're in it, right now. You forgot the links to the questions though, so I don't expect to see much productive review happening. But for future reference, you can always bring up an improper closure here. Tag them .

Of course, a good first step is to edit the closed question itself and clarify how it differs from the supposed original. Make it obvious, to where the folks reviewing the closure can confidently click "reopen" without spending lots of time comparing the minutia of the two to determine the relevant differences.

You're going for this:

As different as an apple and an orange

NOT this:

I hate these things

share|improve this answer
I think the OP is more talking about real apples vs. plastic apples. Superficially they look the same and get closed by people who see the similarity and the existing close votes, but if you actually bite into them (i.e. read the whole body of the question), they are clearly different. – Michael McGowan Oct 15 '12 at 15:57
@Michael: and I'm saying, due to The Power of Editing, you can make a plastic apple look clearly, obviously different. And if it's in danger of being closed, you should. Consider: if it's confusing the close voters, it's gonna confuse folks looking for answers to those questions in the future as well. – Shog9 Oct 15 '12 at 15:59
Yes, and a lot of readers are seeing something analogous to the second two pictures and saying "oh, witches with owls and cats: obviously they're the same picture. – iconoclast Oct 15 '12 at 15:59
@Shog9: did you read my update? I gave a specific example, and it contained VERY clear explanations as to why the two questions are different. But that didn't result in swarms of people revisiting the question in order to reopen it. – iconoclast Oct 15 '12 at 16:01
@Shog9: "in danger of being closed" seems to assume I'll be able to take action between when the first closer votes and the fifth one. These seem to pick up speed, though: once one person votes to close, others just jump in and join. It would appear that either they're in collusion, or the latter ones are influenced by the first. – iconoclast Oct 15 '12 at 16:03
@iconoclast: Your question has an accepted answer. In addition, when I did a quick search I found at least one actual duplicate. Why should this question receive special attention? – John Dibling Oct 15 '12 at 16:06
@iconoclast There is actually a fair amount of time between the first close vote and the 5th, in most cases. – Servy Oct 15 '12 at 16:06
You also posted this question two weeks ago. It was closed two weeks ago. You apparently let it languish there in closed-land for two weeks when you finally came back to it, saw it was closed and became outraged. Now you finally edited it and posted on meta after 2 weeks of inactivity, but you apparently feel that you have been dealt with unjustly because SO isn't treating your question with more urgency than you treated it with yourself? – John Dibling Oct 15 '12 at 16:08
@iconoclast: your edit more than doubled the length of the question, WITHOUT clearly specifying how it was different from the other (it didn't help that an incorrect answer was accepted to the other question, which happened to match the one you accepted to yours). I've edited your question to illustrate how you can make this obvious without requiring folks to spend even more time and effort on it. Oh - and for the record, there was over a week between the first and last close vote. – Shog9 Oct 15 '12 at 16:15
@iconoclast: I know. That's why I added the line "I'm not looking for the intersection of the arrays." to your question. And as you found, finding the complement is trivially easy in PHP as well. – Shog9 Oct 15 '12 at 16:25
@Shog9: You're right. Sorry... retracting that last objection. – iconoclast Oct 15 '12 at 16:28
@JohnDibling: first, what was the actual duplicate? Second, I never said this should receive special attention, I was simply suggesting there should be an appeals process. Many others are saying I should post specifically about my question and ask for moderation (i.e., ask for it to receive special attention). I didn't know until now that was kosher. Lastly, your sarcastic and condescending tone is not constructive. – iconoclast Oct 15 '12 at 16:35
I can't speak to whether or not this is sufficiently frequent as to warrant special consideration, but I've seen it happen before that a question should have been obviously not a duplicate of the candidate duplicate(s) but it was carelessly closed anyway. This question for instance had nothing to do with integer division in Java but was closed as a duplicate of integer division questions anyway. Now maybe this isn't the best example because it turns out this question was a duplicate of an entirely different question, but the point remains. – Michael McGowan Oct 15 '12 at 16:42

Here's a challenging place to start:

  • Closing questions that don't fit our model is one of the most important ways we've created strong, expert communities that can thrive.
  • Having your question closed is an extremely frustrating, personal-feeling event that turns new users off, and tends to drive (usually incorrect) assumptions that new users aren't welcome.

So we're having a lot of discussions about close protocols internally, and will be sharing our ideas for feedback on some of them shortly.

NOTE: We do not intend to do anything that will reduce quality or significantly curb the ability to close posts that don't belong here.

But here are some preliminary highlights still in broad discussion. Again, we'll share specific ideas we're considering as they develop.

  1. Some of the close reasons can be improved in either name or description. ("Not Constructive" and "Not a real question" jump to mind as ones that no one accepts as fair observations on their post.)
  2. Feedback around how to improve a post should be preferred etiquette for all savable questions. We're not talking about mandatory comments ""asdfghjkl", but more of a "one of the five of you should take the time to share things like, "If you explained exactly what you tried, and exactly what you need, this could probably be reopened"
  3. There should be a clearer, easier way to prevent closures or drive re-opens. Currently, if you spot that rare case where close-voters are wrong, you have little ability to help 'save' the post. And we're looking at review-style ways to allow users to look at closed posts that should be re-opened for some reason or another.

Also, here's a tip that seems obvious, but helps a ton:

You should only cast a close vote if you'd be comfortable with your vote instantly closing the question, without knowing whether anyone else voted to close.

I'm a little ashamed to say it, but I'm much more careful voting to close now that my diamond means that my vote instantly closes than I was before. And that's dangerous - if your vote to close factors that others must agree, it means that in some rare cases, the 5-vote requirement can mean that posts are closed with less confidence than a 1-vote one would be.

share|improve this answer
absolutely agree: groupthink in action. – iconoclast Oct 15 '12 at 15:57
Regarding point 3, an often discussed possibility would be to allow casting reopen votes while the question is still open, and just substract them from the number of close votes. – Mad Scientist Oct 15 '12 at 16:47

There are a number of things that you (or others who see your question) can do if a question is getting votes to close as a duplicate of a post that is slightly different.

  1. If you are asking a question and you know that there are other similar but different questions out there you should indicate so in your answer. Say something along the lines of, "I found [this] related question, but my problem is slightly different because of Foo so the solution didn't work for me." Include this information even before votes to close come in if possible.
  2. After the first "proposed duplicate" vote, if it points you to a related question that you didn't notice before, first see if it actually solves your problem. It could be ever so slightly different, but those differences might not affect the answer; if the answer actually solves your problem then the closure is probably correct. If it doesn't, then comment on your question (or edit into the question itself) the reasons why the proposed duplicate isn't actually the same, and why the solutions to that problem don't apply to your example. In many cases this can prevent additional close votes and allow the question to stay open.
  3. If the question is closed anyway, and you are sure that the duplicate isn't actually a duplicate, you first need to edit the question such that it is clear that the two are not the same. Emphasize the differences, discuss the other post such that readers will see they aren't the same. If you do this right away there is a good chance that your question will still have enough attention to get re-open votes.
  4. If, by the time you are able to do all of the above actions to fix up the question, there is no longer enough activity on it to get 5 re-open votes, it is appropriate to flag for moderator attention. Flag your post, use the "other" flag reason, and explain that you have edited the post to explain why it is not a duplicate of the other post. If the moderator who reviews the flag agrees, they can (with a single vote) re-open the question.
  5. As an alternative to flagging for a mod, you could also post a question on the meta site (as you have done here) explaining why the question was incorrectly closed. This will provide attention and activity to your question (as well as quite a bit of activity from mods) who will either determine that the question should be re-opened, or will help explain to you why closing it was appropriate.
share|improve this answer

You must log in to answer this question.

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