11
$\begingroup$

It sometimes happens that the same user posts exactly the same question twice in a row.

Examples:

  • 1 2. user asked twice for the probability of choosing 90% of the colors from a collection of $10^{10}$ balls of $10^7$ different colors
  • 1 2. user asked twice for a proof that disjoint polyhedra must lie at positive distance from one another
  • 1 2. user asked twice for the LU decomposition of the same 3×4 matrix
  • 1 2. user asked twice for clarification of lemma 3 from a certain paper of Arthan
  • 1 2, both deleted but visible to 10K users. User asked twice for proofs of the same claim about $2n$ children choosing from sets of $n$ colored balls
  • 1 2, second one deleted. user asked twice for a closed form for $\sum {n \choose k}^3$
  • 1 2, both deleted. user asked twice for a proof that a certain max-flow problem could be solved with Dijkstra's algorithm
  • 1 2. user asked twice for a proof that any problem $A$ in $\mathcal P$ is polytime-reducible to any other problem $B$

It seems to me that the best way to handle these is to flag them for moderator attention, so that the moderators can immediately merge the questions or close or delete the second one. I can vote to close one as a duplicate of the other, but it sometimes takes along while to gather five votes to close, and in that time the following discussion, which should be happening in one place, is split between two. If it is, the moderators could merge the two questions and their answers, which the site members have no way to do, so I think the flag is required anyway.

Sometimes my flags have been accepted, other times rejected. It appears that some moderators see the matter the way I do, but others don't. I would like to hear other members' opinions on this, and if possible I would like a clear statement from the moderators about whether I should raise a flag in this situation.

Related: Closing duplicate questions by the same poster.

$\endgroup$
  • 1
    $\begingroup$ I like this example. The second question got two answers in 30 mins, the first got none in 12 hours. Made me smile. $\endgroup$ – user1729 Aug 13 '13 at 20:24
  • 6
    $\begingroup$ Have you posted this question twice? $\endgroup$ – John Gowers Aug 18 '13 at 23:05
7
$\begingroup$

I think the first thing to do is to vote to close if you have the reputation to do so.

Flagging the moderators about a duplicate is okay if it is definitely a duplicate. If there is some question, then it would be better to allow the community to decide than have a moderator unilaterally decide for the community.

$\endgroup$
  • $\begingroup$ I would like to clarify that I only flag when I believe it is an exact or an almost-exact duplicate, as in the examples I posted. (In at least one case the two versions of the question diverged after they were posted because they were edited and clarified by different people.) $\endgroup$ – MJD Aug 13 '13 at 19:50
6
$\begingroup$

I look through the few cases where your flag was declined. In the ones that I saw, my instinct would have been to decline too: the OP generally have rephrased the question so that at first glance it is not entire obvious that the questions are exact duplicates. (Moving sentences around and such.)

It would help, I think, if you flagged using the free form field for those type of questions, and specify that (a) the questions are asked by the same user and (b) the two questions are the same up to permuting the sentences or that one question is the refinement/clarification of the other or something like that.

Basically, to add to robjohn's answer, moderators will close exactly duplicated posts; but sometimes it is not that easy to see that two posts are exact duplicates of each other. A little justification in your flag will help a lot.

$\endgroup$
  • 1
    $\begingroup$ I don't think your suggestion is helpful. In each case, I did just what you advise here, but the flags were declined anyway. For example, my flag of math.stackexchange.com/questions/463610 was declined although I said “The same user asked the same question an hour ago at [link]”. In the case of math.stackexchange.com/questions/446622, my comment said “It is an identical duplicate of [link] posted a few minutes earlier by the same user.” For math.stackexchange.com/questions/209331, I said “The same poster asked the same question yesterday”. $\endgroup$ – MJD Aug 14 '13 at 17:53
  • $\begingroup$ @MJD: (i) I was not the one who declined (ii) be specific. Part (b) of my advice asks you to be clear and say that the questions are the same despite the slightly different wording. For the second question that you linked to, the questions are manifestly not identical, which contradicts your flag message. They may be asking the same question, but the question is phrased differently. While I would probably have stopped and compared the questions line-by-line to see if they actually ask the same thing, I cannot guarantee that another moderator will. $\endgroup$ – Willie Wong Aug 15 '13 at 8:18
  • $\begingroup$ To add to what robjohn wrote in his answer: if the two questions are not carbon copies (or almost carbon copies) of each other, you need to convince the moderator that the two copies are, without a doubt, the same. Some of us are more easily convinced than others, so if you don't want to see your flag declined, please err on the side of being detailed and verbose. $\endgroup$ – Willie Wong Aug 15 '13 at 8:20
  • $\begingroup$ The two posts were manifestly identical at the time I posted the flag; one was modified an hour later by a third party. Please see 1 2. But I guess what I really want to know here is if you are sure there is no disagreement or uncertainty among the moderators on whether to accept such flags when the posts are manifestly identical. If you are sure there isn't any such disagreement of uncertainty, all that's left is for me to try harder to get the point across. $\endgroup$ – MJD Aug 15 '13 at 16:30
  • $\begingroup$ For example, I might always include a message in my flag “If they no longer appear identical, it is because one of them was modified after I posted this flag. Please check the revision history" since it appears that moderators might need to be reminded to do this. But if there is disagreement or uncertainty among the moderators on the proper policy here, then I don't think it's fair to put the entire burden on me as you did here. Do you think there is disagreement among the moderators about the correct course of action when the questions are manifestly identical? $\endgroup$ – MJD Aug 15 '13 at 16:32
  • $\begingroup$ @MJD: I am sure that no moderator will decline your flag when two questions are word-for-word identical. I'm also sure the same if the two questions differ only by minor formatting differences. I'm less sure that moderators will check the edit history to look at the state of the question when you flagged it. (In fact, I can confidently say that for "duplicate" flags I almost never look at the history, but I do tend to look at the current state of the questions quite carefully.) Given that the moderators are spread (somewhat unevenly) across many timezones, I don't think the case where ... $\endgroup$ – Willie Wong Aug 16 '13 at 7:57
  • $\begingroup$ ... a drastic edit is made between when you flagged and when a mod sees your flag is all too common. Have you found in your experience that this happens a lot? Otherwise I would just advise you to save your breath and not bother writing that reminder to check the edit history, and just chalk it up to bad luck when something like this happens. $\endgroup$ – Willie Wong Aug 16 '13 at 7:58
  • $\begingroup$ I posted above three examples of declined flags where the questions were manifestly identical. We have explained one of the three: the moderator was fooled because a third person changed one of the texts after I raised the flag. I am willing to chalk that up to bad luck. So out of eight recent examples we have five accepted, one declined because of unavoidable bad luck, and two declined for reasons not yet explained. $\endgroup$ – MJD Aug 16 '13 at 13:13
  • $\begingroup$ You said you're sure that no moderator will decline my flag when two questions are word-for-word identical, which moves the goalposts; I am asking about questions that are manifestly identical. Are you sure that no moderator will decline my flag when two questions are manifestly identical? Are you sure there is no disagreement or uncertainty among the moderators about this? If you haven't asked, would you please ask, and let me know what they say? And if you're sure without asking, how does your surety accord with the 2/7 decline rate for flags on questions that are manifestly identical? $\endgroup$ – MJD Aug 16 '13 at 13:15
  • $\begingroup$ @MJD: manifest is subjective. Something obvious to you may not be obvious to others. Word-for-word is not. I can vouch for other moderators when the situation is objective. But the same pair of questions may be manifestly identical to one person and not to another. If you insist on asking about how the moderators will act in situations where a judgment call is needed, the best anyone can tell you is that: well, it depends on how the moderator judge the situation. $\endgroup$ – Willie Wong Aug 16 '13 at 13:42
  • $\begingroup$ By the way, I have not moved the goal posts at all. Early on I've talked about exact/carbon-copy duplicates. You asked a question about something, I give you the best answer I can give. While my answer may be unsatisfactory, I would not call it moving the goal post when you are the one who planted the post there at "manifest" yourself. $\endgroup$ – Willie Wong Aug 16 '13 at 13:47
  • $\begingroup$ As to the other cases you are talking about. The first I agree that with a bit more time checking everything, one can come to the conclusion that the two are duplicates. But see the second paragraph of my answer. For the last question, I disagree that the two are manifestly duplicates. In fact, in the second version the user added some important clarifications that help drive at where the confusion is. While it would have been better had the user edited the original question, it stands that the two questions were not exact duplicates, at least not ones warranting unilateral moderator action. $\endgroup$ – Willie Wong Aug 16 '13 at 13:57
5
$\begingroup$

In these situations, if there's no answers yet, I usually handle it by doing two things: voting to close as a duplicate, and posting a link to the original question as a comment. I usually delete the automatic comment that comes up when you vote to close as a duplicate and replace it with something like:

Possible duplicate of drawing at least 90% of colors from urn with large populations.

Exact duplicate of drawing at least 90% of colors from urn with large populations and posted by the same user.

This way other users will immediately know to go to the link to the original question, and it helps divert discussion on the duplicate.

On the other hand, if the duplicate already has answers on it, I think that's the right time to flag for moderator attention, so that the answers can get merged. It also doesn't hurt to vote to close as a duplicate as well, in case a moderator decides to ignore the flag.

$\endgroup$
  • $\begingroup$ To be merged, the answers need to apply to the question into which they will be merged. This usually means that the questions need to be very close, otherwise the answers to one question may make no sense as answers to the other. $\endgroup$ – robjohn Aug 13 '13 at 22:49
  • $\begingroup$ @robjohn Yea, I was assuming MJD meant questions that are identical that the OP asked twice just to get more answers. $\endgroup$ – Ataraxia Aug 13 '13 at 23:02
  • $\begingroup$ I thought my long list of examples would make clear that the questions are not merely "very close" but in most cases identical. It is dismaying that we can't even seem to reach a common understanding of what the scope of this discussion is. $\endgroup$ – MJD Aug 14 '13 at 17:56
  • $\begingroup$ @MJD Just to check, is this to me or robjohn? $\endgroup$ – Ataraxia Aug 14 '13 at 17:59
  • 2
    $\begingroup$ It is a cry of frustration to the Universe. You, however, do seem to understand what I am trying to talk about. $\endgroup$ – MJD Aug 14 '13 at 18:05
  • 2
    $\begingroup$ The example "drawing at least..." isn't a duplicate at all. The user does response and clarify the difference between the two questions. If a user does response and clarify, we should give the user the benefit of the doubt. $\endgroup$ – achille hui Aug 15 '13 at 16:49
  • $\begingroup$ @achillehui I know. I just used that for the sake of demonstration. I obviously couldn't use an actual example of an exact duplicate because it would be deleted. $\endgroup$ – Ataraxia Oct 12 '13 at 18:46

You must log in to answer this question.

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