SE is making the way accepted answers behave configurable per-site and is looking for input from our side as to what our preference is.

Currently, accepted answers - answers that the asker of the question has explicitly marked as answering their question - always are shown at the top of the list of answers when sorting by votes, regardless of whether there is an answer with a higher score. In the future, it seems we will be able to decide that instead it should just be sorted by its votes, ignoring its status as the accepted answer.

So, what are your opinions on this topic? Would you like the answer to be unpinned, and if so, why? Would you like the behaviour to stay as it is? Again - if so, why?

Note that in the linked meta question there is a deadline by Sept 19th for SE to decide what the default on all sites will be going forward, but this post is about your preference for the behaviour of accepted answers on quantumcomputing.SE, not about your preference for the default for all sites.

Shamelessly copied from the homonymous question on physics.SE


I offer a dissenting opinion and contend that the accepted answer should not necessarily be pinned to the top, especially in the event that the accepted answer has significantly more downvotes than other answers.

I'd guess, without any evidence to confirm but based on my intuition, that the casual user of our site might not know the difference between or the subtleties between "accepted" answers and answers with the largest number of upvotes-downvotes. Rather, the accepted answer might have answered or confirmed the answer of the specific OP; however, the accepted answer might be trying to "push an agenda" of the OP.

Then again, I'm not sure there's a difference between a question having two answers, with the accepted answer having +5 and the highest voted answer having +6. It might be interesting to look at analytics of all multi-answer questions to see how much of a gap there is between the highest voted and the accepted. If it's a handful of cases, and if the gap is not large for the majority of this handful, then it might be moot.

But there are some questions that come to mind where maybe a flag would be warranted to say that the accepted answer is not considered consistent with the mainstream wisdom of the crowds? This may be applicable where there's a big gap between the accepted answer and the mainstream answers.

  • $\begingroup$ some data about questions where this change would make a difference can be obtained from the SEDE. There are, as of now, apparently 171 questions on this site that would be affected. See data.stackexchange.com/quantumcomputing/query/1458833/… (this version of the query is from Emilio Pisantry, discussed on the physics chat around here) $\endgroup$
    – glS Mod
    Sep 12 at 11:42
  • 1
    $\begingroup$ That's a good summary. There are about 10-15 in the "long tail" of having a large gap between the accepted and the highest voted. Of these I suspect many of them may be that the accepted answer was accepted by the OP, who never went back afterwards while newer answers were asked. I don't know if that changes my answer. $\endgroup$
    – Mark S
    Sep 12 at 13:45
  • $\begingroup$ personally, I also don't think accepted answers should be pinned. At the end of the day, I'd say we want to prioritise showing the answers that better reflect what the community likes the most, not what the person that asked the question personally found more useful. Both because the OP might not look back to change the accepted answer after new ones are posted, and because the OP might not be the best suited to judge the technical content of an answer, or in some cases might hold personal "non-mainstream" views, and accept answers reflecting their own views $\endgroup$
    – glS Mod
    Sep 12 at 13:51
  • $\begingroup$ Could we judge this with a difference between upvotes and downvotes? Or maybe put a disclaimer in the accepted answer that it's not mainstream? The first case (maybe the OP never goes back to look) may have the accepted answer have a smaller gap between upvotes and downvotes because the OP never looked at later answers; the second case (the accepted answer is not mainstream) may have a lot of downvotes in addition to any drive-by upvotes. $\endgroup$
    – Mark S
    Sep 12 at 14:16
  • $\begingroup$ sorry for the delayed reply. So, if you mean as an automatic solution, no, anything like that would require fundamental changes in the SE code. For that we'd need to have to propose it to SE developers, have very good reasons for the change, etc etc. I don't quite see that happening. On the other hand, post notices are a thing. They are fairly rarely used though and don't allow custom text. The closest to what you mean would be a "needs citation" post notice, which could be used to signal a sort of "community disagreement" on an answer $\endgroup$
    – glS Mod
    Sep 20 at 9:21
  • $\begingroup$ that said, these are mod-only features, and to be reserved for exceptional cases. I could see these being used in rare cases, as a way to mark that the community doesn't quite "accept" an accepted answer as canonical, but at the very least I'd say we'd have to make a meta post in every such case, to ask if people agree. Otherwise, deciding whether an accepted answer is "canonical" can be subjective, and we (mods) might not want to take the responsability of deciding/acting on it on our own. $\endgroup$
    – glS Mod
    Sep 20 at 9:24

For now (open to debate on this!), I am going to argue that we should have accepted answers stay pinned to the top.

I have read a lot of the arguments in favor of unpinning accepted answers, and I even agree with some of them, but I do think they miss a sort of critical point - ease of access.

Here's what I mean - an accepted answer is a very unique mark of usefulness. Only the OP can truly say, "Yes, that is what solved my problem," and it's notable that when I've searched for answers to questions on various sites, often the most upvoted answer is what the community thinks has solved the problem - occasionally neglecting some sub-aspect of the question, such that even though the accepted answer has far fewer votes, it is vastly more helpful.

Now, if this was unpinned, in the cases where it is less upvoted, it often gets buried in the list of answers. On the other hand, if it is pinned, it's still quite easy to see the top voted answer - scroll down one answer. In other words, keeping the accepted answer pinned barely reduces top-voted answer visibility, whereas unpinning the accepted answer often does reduce its visibility significantly in the edge cases where this distinction matters and which caused this whole debate to begin with.

I'm very open to disagreement on this, but this is my current stance and I suppose this can serve as the 'stay as is' answer for voting.

  • $\begingroup$ I agree that this doesn't make that big of a difference, but I honestly cannot think of a single case of a question anywhere on SE in which an accepted answer was more useful than a non-accepted answer with (significantly) more upvotes. I acknowledge that "often the most upvoted answer is what the community thinks has solved the problem - occasionally neglecting some sub-aspect of the question", but if you stumble upon that post, isn't it more likely that you were looking for the answer to the question as it was interpreted by the community rather than the OP?Can you point to some example? $\endgroup$
    – glS Mod
    Sep 12 at 11:20
  • 1
    $\begingroup$ I totally agree with @AudenYoung here. On this site, we should have accepted answers stay at the top. SE brought up this matter because on StackOverflow, answers from 10 years ago are no longer correct. That's not true here. $\endgroup$ Sep 14 at 16:21

You must log in to answer this question.

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