Stack Overflow and many other Stack Exchange sites have changed the way they sort answers:

We no longer pin the accepted answer (with the green checkmark) to the top of the list of answers. By default, we now sort strictly by votes (descending order by highest score), and the accepted answer's order in the list is based on its score.

I am posting this here so that we can have a discussion about whether accepted answers should be pinned or not on Meta Stack Exchange.

  • 1
    Much of my rationale here for keeping the pin still applies. Oct 10, 2021 at 6:09
  • 6
    I believe they should be unpinned only if mods and staff are given the option to manually pin certain answers. That way, pinning an answer is no longer in the hand of the OP. But if that FR doesn't get implemented, then definitely don't unpin accepted answers and we can keep using that as a workaround, as it has always been.
    – 41686d6564
    Oct 10, 2021 at 6:40
  • 5

7 Answers 7


The status quo of having the selected answer pinned in most cases is the best option under the current system for meta.

In some cases, it's critically important to have a specific answer pinned to the top. Without another alternative mechanism (and there are occasions where staff have had to do.... interesting... things, like change the ownership of a post, or having different staff post answers for pinning purposes), It's going to be difficult when we need to pin the 'canonical' answer where folks can see them.

This is especially true in situations where we need an official answer. As things are, the status quo serves our needs better. It is true that answers can go obsolete, but there's precious little where this helps in our specific case, IMO.

On the other hand, while a 'special' option for being able to pin a canonical post would be nice, and would mean having the selected answer decoupled from the OP. This can be handy sometimes. That said, I'm fundamentally opposed to having a feature for one site that wouldn't be useful elsewhere, which is what this would be, simply cause it wouldn't be maintained over time. Without such a feature, I don't think this would be a great idea. On the gripping hand, If we got collectives like features across the network, I suspect we can do without it

  • 3
    "I'm fundamentally opposed to having a feature for one site that wouldn't be useful elsewhere" That's why this should be a feature on all meta sites, not just Meta.SE (at least that's how I think should happen).
    – 41686d6564
    Oct 10, 2021 at 7:56
  • 1
    @41686d6564 and this answer, and the one I commented about on the question, both argue that it's bad to have them unpinned on all meta sites. Oct 10, 2021 at 9:39
  • @Sonic In case it wasn't clear, I was replying to the part about creating a "special option" for manually pinning canonical answers. In my opinion, that option addresses all the concerns in this answer and your answer to the other post. I could be wrong though; that's just my opinion. The only valid concern that that option (if it's implemented) won't solve is the one in Robert's answer about the number of affected posts whose accepted answer would need to be reviewed to determine if it's worth keeping pinned.
    – 41686d6564
    Oct 10, 2021 at 16:01
  • 1
    By the way, I'm not asking for accepted answers to be unpinned. I don't really have a strong opinion about either side. I'm just saying that if it's decided to have them unpinned, then we really need another way to pin certain answers (by mods, staff, etc.) in that case.
    – 41686d6564
    Oct 10, 2021 at 16:06
  • Don't we already have the "staff notice" thingy which marks an answer as special? I thought now SE staff didn't need to do "interesting things" anymore.
    – muru
    Oct 11, 2021 at 12:26
  • Why pinion ourselves to the status quo?
    – Makoto
    Oct 11, 2021 at 21:22
  • Well I'd welcome good alternative views to why not.... Oct 13, 2021 at 5:41

Looks like this would affect around 4000 questions

We actively depend on the pin for this question to ensure that recently implemented features are displayed first.

We also have official answers here by Stack Exchange employees. Of course we could have some alternative means of pinning them as there is an official answer lock but we'd need to go through the old questions that need it as it's a relatively recent feature.

On non meta sites, someone might be taking the code and using it elsewhere. That's less likely here although it's possible they would still do something with the accepted answer e.g. try to follow its recommendations even when they aren't the optimal ones. Hopefully in the most egregious cases the downvotes such answers are likely to come with should persuade them against that. We're pretty good at voting round here.

Additionally all metas including this one are more likely to have higher rep users participating on it. We know that the accepted answer is pinned and that we might need to look at other answers so we're less likely to be fooled by that.

So there do seem to be good reasons to keep the pin, ones which would leave us asking the developers to special case it for some important questions if it was removed, and I can't think of any questions where the pin is causing any great confusion.

  • 2
    Ow. I... didn't consider the particular problem of just how many posts it would affect Oct 10, 2021 at 8:08

I'd like to try to think of this in a different way, and maybe this'll help motivate a better solution rather than a boolean yes/no response, as easy as that may be.

In the premise of Q&A, the "best" answer is the one that the most people have elevated and have voted to be the most accurate. This works extremely well for Q&A since the OP who actually accepts an answer may not really know that the answer they've accepted is dangerous, wrong or otherwise misleading.

In the premise of Meta, this falls flat on its face, since Meta is not Q&A. There are Q&A elements to it, but the questions and answers that are used here are not traditional Q&A.

For instance, a feature to keep answers at the top of other answers in spite of their score indicates that they have some elevation or importance at a policy level rather than a community curation level. That distinction shouldn't be ignored.

So instead of just making this question boolean, I'd actually propose something different: a Meta Post type.

Meta Posts can violate some conventions of Q&A since they're not traditionally Q&A, which could include:

  • Arbitrary ordering of answers
  • Longer post lengths (to accommodate those kinds of "Q&A"s we have that are just massive questions with tons of answers to it)

...among others.

In the face of just the question, I would want the accepted answer to be unpinned from the top on Q&A-style Meta posts. For Meta Post-style posts, we need to have a better way to manage this.

There's only so long that it makes sense to shoehorn this functionality into something that it really wasn't suited for before you run into breaking functionality across the board.

  • So we would need a [Meta] tag on the networks Meta site?
    – Luuklag
    Oct 12, 2021 at 14:13
  • 3
    No, I'm thinking it's something that is a completely different content type. We're limited right now to questions and answers which doesn't quite fit or conform well enough with what Meta posts really are. The point of this: Meta discussions aren't Q&A and we shouldn't be trying to shoehorn them into Q&A.
    – Makoto
    Oct 12, 2021 at 14:36

Since Meta Stack Exchange has an entirely different Q&A format than other sites, where sometimes it's a standard Q&A, and other times it's an announcement with feedback, or a welcome / goodbye for an employee or moderator - we need a third option:

Per Question Selectable:

Mockup of a toggle button, that chooses between an OP selected pinned answer and the highest voted answer appearing on top.

Initially the poster of the question would select whether they wish to choose to have an answer of their choosing or the highest voted answer appear on top, the setting could be changed by moderators and staff but not by editors or other users.

This would allow the best setting for each question and to always have the correct setting available regardless of the type of post or any shortcomings of either choice being forced on to all questions.

Things would remain as they are until the OP edited their question and changed the setting or a mod / staff decision was made to toggle the setting.

This offers: the least change and the most flexibility for every situation.

  • 6
    that would likely be an ideal, and useful on any meta (hint hint) Oct 10, 2021 at 12:31

Keep the accepted answer pinned on MSE because:

  1. In all threads I recall -where a technical criteria can be applied- the accepted answer tended to be the most relevant/accurate.

  2. MSE has lots of "opinion threads" that should be read as such. The OP's choice of the pinned answer does not significantly skew overall vote distribution, it in fact acts as a reminder of plurality by highlighting the importance of individual choice.

  3. The pinned answer would only become a problem on "consensus threads" where the community tries to define guidelines on matters with considerable latitude (like editing, etc). However, such threads tend to be aggregated into FAQs over time, thus the reader is still required to search for the right thread and put each into proper context.


Practically - MSE works differently and having the 'selected answer' pinned has very little value.

In many cases - for example questions introducing new features for feedback, feedback on official announcements over the blog. What's the most 'important' answer can be somewhat arbitrary - in some cases an 'official' answer, in some cases the 'best' solution for a problem on the platform, and in some cases what the user felt was most useful. There isn't an immediately 'apparent' pattern to it.

Where we use rely on 'selected' answer first order to emphasise a specific post, that information could be included in the question instead, even if it's a bit of a bodge, and use faux tags to keep track of individual answers in new features/bug report posts as we do now.

As such the current status quo doesn't have any particular value, except tradition.

  • 4
    as an aside, I don't actually agree with changing the status quo. I just wanted a half decent dissenting opinion, and don't mind taking a reputational hit for giving people options Oct 10, 2021 at 11:37
  • 1
    I really like your 2nd paragraph but I'm also in favor of keeping status-quo for the accepted answer so I can't upvote the counter-argument.
    – bad_coder
    Oct 10, 2021 at 12:47

The highest scored answer should have top place

The top-voted answer should appear on top, even when a different answer has been accepted (this is the new behavior on Stack Overflow and many other Stack Exchange sites).

  • The mandate of SE in general is to create a “library of reusable knowledge”, and each site attempts to do that in its subject area by attracting and retaining experts. These experts become “the community”, and the community expresses its will — and expertise — principally through voting. The highest-scored answer is the direct expression of the community.

  • The community as a whole is often better suited to determine what the best answer is to a question than the majority of question authors.

  • It is also true that the highest scored answer is more likely to help future visitors who are either looking for a quick answer or thinking of asking the same question.

  • The green check is supposed to indicate that the author of the question found a particular answer more useful than others. It has been observed that many question authors here (a) lack the maturity to judge a "good" answer, and (b) are inclined to give the green check to the first person who answers. (taken from here)

If you want the highest voted answer to be on top then please upvote here. Please avoid downvoting as this confuses the count.


You must log in to answer this question.

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