6:35 AM
Currently it would lead to cycle of duplicate closures of length 3: 706016 $\to$ 23316 $\to$ 95799 $\to$ 706016
@BillDubuque I think that it would be better to close as a duplicate in the opposite direction (706016 $\to$ 95799 rather than 95799 $\to$ 706016). I think that the post which is the most useful should be left open, as discussed on meta. And this question has more answers and some answers are more detailed. I have also mentioned this in chat. — Martin Sleziak 54 secs ago
4 hours later…
11:12 AM
I am still intrigued what would happen if other users voted to close, but maybe the SE software would prevent it.
Loops should not be possible anymore, I expect that the last close vote would fail. If it doesn't and a loop is actually created, retag this as a bug. — Mad Scientist Nov 26 '13 at 20:00
8
Alright, I've added some protection against this via a neat library that danielrbradley created. There's sure to be some issues (which we'll log when a block occurs), but the base cases are covered (and unit tested, oh my!).
11:55 AM
@JyrkiLahtonen I have to admit that it is much easier to write a meta post like the one you're referring to than to apply this in practice to decide whether to close as a duplicate or not.
12:33 PM
@MartinSleziak I agree with the choice of this version as the duplicate target. The principle at work is that the highest voted edition should stay. I will use my supervote to redirect the closures.
As you saw, I have pinged the gold badge holder who voted to close in the opposite directions. (That seemed as a natural thing to do.)
So I guess if he sees some reason why the other one should be preferred, he will menion that in comments or in chat.
« first day (1631 days earlier) ← previous day next day → last day (2867 days later) »