One thing SE struggles from is “trigger-happy” duplicate voters, so I hope that is something tat can be improved here; many times questions marked as duplicates are similar, but because people are either too lazy to investigate the differences (read: trigger happy) they choose the easier option of marking it as a dupe.
Not only that, when your question gets marked as a dupe I find it incredibly difficult to get that association removed - it states you can “edit your question to explain why it’s not a dupe” - but even when you do the association seems to stick around.
To avoid the above, I would like to see a system where if someone explains why it is different via an edit that the question being marked as a dupe vanishes temporarily (or perhaps a different notice) and then mods (or people with high rep on the subject) get notified to verify if it’s a dupe or not).
Now, to avoid any edit notifying the above users, I would assume the user would use a “special” edit button for this purpose which can flag that this is an edit of clarification.
Because dupes aren’t always correct as well as other reasons I don’t really agree with the suggestions of redirecting users from dupes to other questions or integrating the answers from other questions.
I will note though that:
Every “mark as duplicate” vote will need to link to an answer on an other question. Exceptions apply for linking to a blog post/canonical from an other category.
Would likely help avoid false positive dupes, but doubtful it would eliminate them altogether.
I agree, but it is also partially a consequence of that the dup hammer you get when you get a gold tag badge removes the possibility to vote to close as dup. Some users have suggested to make it optional to use it for those cases where they are not sure.
I think this is a superb suggestion. And this special edit should be connected to the specific dup/nondup that you are explaining why it’s not a dup.
Ahhh didn’t realise that. But yep, I would look to improve on that as you suggested; or maybe users shouldn’t be given this ability unless they are “highly regarded” in this particular topic/tag?
Hmmm, I could be wrong but I get the impression that an explanation of what dup hammer means could be a good idea. If I misread you, please interpret it as information for other people reading.
On SO, when you reach 3k rep you get the privilege to cast close votes on questions. One of the close reasons is that it is a duplicate of another question. A question needs 3 (previously 5) close votes to get closed.
But once you get a gold tag badge (at least 200 answers on questions with a specific tag and with a total score (not rep) of at least 1000 on those answers) you can no longer vote to close as a dup. When you try to do that, the question gets instantly closed as a dup. This is often referred to as “gold hammer” or “dup hammer”.
So as you can see, they are not given that ability unless they are highly regarded in that tag. The issue is that your ability to vote instead of using the hammer is permanently removed for that tag. Well, unless the tag gets removed, which it can if you provide enough bad answers or delete previous answers.
Ok got you - I thought you were originally saying once people reach a certain rep (not tied to any one tag) they get the ability to be able to instantly close questions as a dupe - thanks for the rundown
Often dups aren’t closed fast enough. There are two or more sides to every issue. Too often I’ve seen an answer or two sneak in before a question could be fully closed.
I do agree that there were occasional cases (certainly the minority) where something appeared to be a dup at first glance, but after careful reading of both questions wasn’t really. However, this is not the case we should optimize for. Usually the confusion was due to badly worded title or first sentence. Teaching people they need to be more careful with the title and the first sentence is a Good Thing.
Maybe, but only as long as the question still remains closed while it’s in this “disputed dup” state. Otherwise, this just creates an easy way to end-around getting closed as a duplicate.
Nothing people do is always correct. Let’s keep things in perspective. Erroneous dups do occur, but not often. And, when they do, it’s often the asker’s fault for being unclear what is being asked. If they had done their prior search properly, then they would have found the near-dup and could have mentioned it in the question and explained why theirs is different.
Dups scatter knowledge about, making it hard to find, and dissipating volunteer energy. That’s a much bigger problem than the occasional false positive.
I disagree - I have seen many a case where there have been false positives, I personally have been on the end of a number of those myself - so I believe it to be a bigger issue than you suspect.
It is not only annoying, but may lead to users opening yet another question and detailing in there why it is not a dupe since they struggle to get the dupe tag removed off their original question.
It occurs to me that stale content and duplicates are opposite sides of a single coin.
Even if we simply assume away the issue of getting reliable duplicate identification, it only helps if new answers that cover evolving situations get enough attention to rise to the top.
The SE sites I am familiar with have a very checkered history on this.
Some new answer to old posts rise up. But if the OP isn’t around they don’t pass the originally accepted answer in the default sort order.
Some users actively maintain accepted answers, but that means that many of the votes are for the superseded content.
Sometimes you see new questions like “How does foo affect bar in 2020?”. If you’re lucky the poster at lest links the old version and explains why those answer may not eb applicable any more.
The suggestions elsewhere that acceptance need not pin an answer indefinitely could help with the stale accepted answer. That or some other mechanism for community over-riding of acceptance pinning.
The info text for gold badge says “You must have a total score of 1000 in at least 200 non-community wiki answers to achieve this badge.”
And at this link: https://stackoverflow.com/help/privileges/close-questions it says " It takes 3 close votes to reach the closing threshold (except for duplicate questions, which can be marked as duplicate with a single vote from a user who has earned a gold tag badge in one of that question’s current tags, unless that user participated in editing the tags)."
Well, we all (including me) suffer from confirmation bias. I have seen a lot of examples with a scenario where someone posts a poor question with no research effort that it is VERY clear to ANY experienced user that it does exist a duplicate. Very often it also shows clear signs of not following other rules, like creating MRE, multiple language tagging, poor indentation and such. Someone comments “This have been answered over and over again. Please search.” Someone else close votes with a suitable dup. Someone else also votes for that, or another close reason. And then, dispite the fact that the question is downvoted, closevoted and commented, someone posts a halfbad answer, and someone else upvotes it and the questioner accepts it.
For those who does not know, on SO, it’s much harder to delete bad questions with upvoted or accepted answers.
I agree here. It’s way to much “Noooo, you should NEVER delete and repost a question. Edit it instead. When you have edited it, people will retract their downvotes.” Yeah, right. That’s definitely a situation where SO completely have ignored human behaviour.
The problem you are addressing here is because of a wide misconception of what “acceptance” means. It should only mean the OP saying “this is the answer that helped me the most in solving the problem”. It also marks the whole question as “Thanks, I’m done”. It doesn’t not mean “this is the best answer”.
Both those separate mechanisms are useful. Only the OP can decide what helped him the most, and whether the question is resolved to his satisfaction. There should still be a mechanism to mark a question as resolved, possibly pointing to one or more answers.
However, this has nothing to do with quality of the answers. This is the part SE got wrong. Sort order should continue to be by quality, as judged by the masses voting. I wouldn’t have a problem doing away with accepted answer status altogether, and I don’t think rep gain from such a thing is useful.
Sure. Acceptance is about solving a particular problem while voting is (at least can be, if the site culture encourages it) about value as a on-going resource.
So two possible goals were conflated from day one on SE.
I’m probably missing some context, but I’m not sure which points that post supports.
It was closed off topic on MetaSE because it should have been asked on MetaMathSE. It is about a question that was closed as off topic some time ago which has since been updated and reopened.
Whether or not the post was valid to be closed is not the point; my point is the user is struggling to get his post reopened, has no idea what is happening with it, is it being reviewed or what - so he resorts to meta.
It was posted, and then it was edited and went to the reopen queue (review) when it was reopened.
The version that was closed was https://math.stackexchange.com/revisions/3248613/2 which is entirely a title of: “Prove ceil(z)=(z+0.5)-(arctan(tan(pi*(z+0.5))))/(pi) when z is not an integer” with the body of:
I have tested this many times but cannot come up with a proof.
At 16:21 the review was created to reopen it started from the OP’s reopen vote. It was reopened by YuiTo Cheng, kimchi lover, Lee David, and Atticus. Those votes came from the reopen review.
The discoverability of functionality is one of the bigger problems for StackExchange. The how do you get a post reopened is difficult had has significant friction to it.
That said, the OP jumped the gun in “why isn’t this reopened now?” rather than waiting for the reopen review process to proceed as normal.
I feel this question (and the corresponding MSE post) are poor examples of “explain why this isn’t a duplicate” because it wasn’t a duplicate - it was a low quality post that consisted of one sentence.
But lets keep in perspective that all this resulted from the OP writing a bad question in the first place. Yes, the re-open process can be difficult sometimes. But, you only get into that situation by dumping crap on us. Take it as a lesson to be more careful next time.
We definitely should make it easy for the OP to find them without a special privilege — there are a few questions that I have asked e.g. on Apple SE that were closed as duplicates and deleted but I can’t find anymore. That it is hard to find deleted questions by yourself until you have 10K reputation is one of the most annoying easily-fixable technical problems with SE.
That it is hard to find deleted questions by yourself until you have 10K reputation is one of the most annoying easily-fixable technical problems with SE.