Here's the problem... and I'll admit that I don't have an ideal solution.

  • A question gets closed as a duplicate of another question.
  • The question is modified so that it is no longer a duplicate
  • The question is reopened based on that it isn't a duplicate
  • ... but it is now too broad.

(this sequence could just as easily be reversed and layered - starts out unclear, is refined, reopened, turns out to be a duplicate now... and we could do this as another step with it getting changed from being a duplicate to too broad)

The five people who cast the first set of votes can't vote to close the question again. This is probably an OK situation for the big sites where there are scores of close vote reviewers and removing a set of five from that pool doesn't significantly impact the ability to reclose a question when it changes. However, on all of the smaller sites, where you might have a dozen (or fewer!) close vote reviewers who are reasonably regular, this means that the question that isn't a duplicate anymore can't be closed for any reason again (unless you start pestering the diamonds to make calls they are often reluctant to do).

This in turn leads to situations that are less than ideal and difficult to explain to users. More than once I have answered questions about "why wasn't my duplicate reopened - it isn't ${original reason} anymore". An example can be seen in a meta.programmers post where a question moved from career advice to being a poll and wasn't reopened (a mod pointed out that he could in theory reopen and reclose it if it was that much of an issue - the post was eventually deleted).

There are some approaches that seem to have stalled or discounted for one reason or another.

  • Reduce close votes needed (only need 3 to close)
  • Weight close votes differently (if you have a badge, it counts for more)

The above too approaches are essentially making it so that fewer people are needed to close a question allowing for a larger pool of close voters on a second round.

This approach is to let a person cast a vote again for a canned (either off topic or system wide - not custom off topic) reason so that if a question is reopened because the original close vote reason no longer applicable.

  • 2
    Flag the post and ask a moderator to reopen and close it with the correct close reason. Things like this happen infrequently enough that they don't justify the time and expense of a software solution. – Robert Harvey Jan 29 '16 at 15:35
  • 1
    @RobertHarvey what about when its already been reopened (OP's 250 rep reopen vote on their post, and a few other people who disagree with the original duplicate vote and its now reopened)? Flag a moderator to close as too broad? – user213963 Jan 29 '16 at 15:37
  • 1
    For the egregious cases, yes. Posts that get reopened because people disagree with a technicality tend to be edge cases anyway. – Robert Harvey Jan 29 '16 at 15:39
  • On graduate sites, I'd say just move on. If the question is now too broad, someone else will eventually vote to close it. – yannis Jan 29 '16 at 15:47
  • 2
    @Yannis how long to 'eventually'? What about the user experience of users who provide answers to what is now a poll? Or repeating material in a question that went from unclear to a duplicate (and now is fragmenting the information across multiple posts and not having the this is a duplicate linkage)? – user213963 Jan 29 '16 at 16:36
  • Even if we assume that happens @MichaelT, it happens once in a blue moon. The potential for abuse here is far greater than any potential benefit. There's a reason close votes expire, and there's a reason you can't vote to close the same question twice. – yannis Jan 29 '16 at 18:58
  • @Yannis actually... once they expire, they can be recast. Same with reopen votes. At the beginning P.SE three vote experiment a number of people took that opportunity to reopen some questions that were borderline. This has been the case for nearly a year. – user213963 Jan 29 '16 at 19:21

You must log in to answer this question.