110

Similar to this request but narrower, I'm suggesting we make a particular exception for negatively-scored accepted answers (those with score less than zero, or some other threshold like -3), so that they stay in the normal sort order and don't jump to the top.

The accepted answer on this question on bh.se is -5 at time of writing, above an answer with +8 (which is quite a good score on our little site!). It also doesn't help things that the accepted answer is about a mile long so it takes a lot of scrolling to get past it:

enter image description here

15
  • 3
    This is another thing that might be a user preference on SE... But in my opinion, having the accepted answer at the top is the most logical thing. This is after all what helped the OP most, and since at least 10 minutes must pass before an answer can be accepted, the OP will already have considered other answers.
    – MarioDS
    Apr 30, 2013 at 7:42
  • 22
    Wow... just, wow. Apr 30, 2013 at 7:47
  • 3
    I think the accepted answer should be on top; no matter how many votes. If it's not a useful answer, it should be flagged and removed.
    – Kermit
    Apr 30, 2013 at 14:03
  • 6
    Answers get grayed out when they are -3 or lower. For consistency it would make sense to use the same threshold here.
    – hammar
    Apr 30, 2013 at 14:26
  • 28
    @FreshPrinceOfSO - No, it should not be flagged. Moderators should not be placed in the position of having to decide the technical correctness of an answer, and I decline any flags I see like this. The proposed solution here seems to strike a reasonable balance between emphasizing the accepted answer of a user while still allowing the community to judge its correctness. Apr 30, 2013 at 14:51
  • @BradLarson What amount of flags are needed to help a moderator decide the "correctness" of an answer?
    – Kermit
    Apr 30, 2013 at 14:53
  • 27
    @FreshPrinceOfSO - None. We're elected to be janitors, not domain experts. We remove answers that are spam, trolling, nontechnical rants, follow-on questions, etc., but it's up to the community to decide the correctness of an answer via votes. If the community feels strongly enough to downvote and vote to delete an answer for technical reasons, great. We're not going to unilaterally delete an answer based on a flag because someone claims it is wrong. Apr 30, 2013 at 14:57
  • 3
    @BradLarson Janitors? That's depressing. I used to think of moderators as Greek gods.
    – Kermit
    Apr 30, 2013 at 15:03
  • @FreshPrinceOfSO: Sorry to disappoint. Apr 30, 2013 at 19:50
  • 1
    Mods = janitors? That's rubbish. They're custodians/gate-keepers who deal with content, users and crap that most people never see!
    – nickhar
    May 1, 2013 at 4:02
  • 3
    @FreshPrinceOfSO I think this idea is borrowed from Wikipedia: "Wikipedia's administrative tools are often likened to a janitor's mop, leading to adminship being described at times as being "given the mop"." Jul 19, 2013 at 2:38
  • Another view here! I have someone(s) who downvote pretty much everything I do. They've driven some of my answers to -1 (which suggests they're wrong) they are of course not. The tick is all I have left! Don't take that.
    – Alec Teal
    Sep 16, 2015 at 18:22
  • I would be keen to see your thoughts as an answer or comment on my proposal for an alternative/complementary way we could deal with outscored but accepted answers being always first in the sort order forever: meta.stackexchange.com/questions/268666/…
    – PolyGeo
    Nov 1, 2015 at 8:57
  • Referenced in meta post Introducing Outdated Answers project (2021-02-18). Feb 19, 2021 at 13:42
  • Shouldn't this be changed to status-completed?
    – BSMP
    Sep 10, 2021 at 5:10

8 Answers 8

3

As of September 2021, on Stack Overflow, accepted answers are no longer pinned to the top, regardless of score. The team is currently mulling disabling accepted answer pinning on more (possibly all) sites.

As such, this is now on all sites and on Stack Overflow.

As far as implementing this as written, I have the SOUP user script which does this automatically, and I've found it useful in certain cases but not in others. I think that the site setting outlined in the second post above should be able to switch between all three cases (pin all accepted answers, pin only non-negatively-scored accepted answers, or don't pin anything) so that all site-specific cases are covered.

1
  • Ironically, this answer, which I've just accepted, is one of the cases when pinning makes most sense :) Sep 19, 2021 at 7:45
71

I have seen many instances of downvoted answers being accepted across at at least half a dozen sites. I cannot think of a single instance where having the accepted answer be at the top in these situations was a good thing. They are almost always some combination of outright wrong and possibly dangerous. At the very least they are never answers that are useful to others.

  • On technical sites, usually this means somebody gave out a lazy hack (like alias rm="rm -rf") that—for good technical reasons—more experienced community members know is a bad idea but the OP ran off and did anyway.

  • On less technical sites, they are usually a case of bad-faith questions where the OP had a anti-expert answer in mind and somebody played along and gave them they answer they were looking for (usually that they already knew and where trying to promote in the first place).

Either way, these situations don't deserve the regular answer treatment.

In order to fix this but rule out most possible cases for abuse, I would think the same filter that applies to self-answers could kick in if the accepted answer is < -2. Often these are low quality questions that don't get a lot of attention and if one downvote was enough to trigger different behavior I could see an issue with revenge downvotes targeting 0-scored accepted answers being a problem. In situations were a question does get community attention and there is really something wrong with the accepted answer, even low traffic sites can pull it down a few extra votes.

5
  • I find @shog9's statistic of "237 questions where the accepted answer scores less than -3 and another answer scores more than 0" so hard to believe, I feel like I've seen this so many times. Perhaps -3 is too stringent?
    – hayd
    Jun 4, 2015 at 6:38
  • 5
    @hayd The numbers Shog gave are pretty solid; I don't doubt they are correct. The issue is that these cases when they do happen tend to be absolute train wrecks and as such attract more than the usual slice of attention. Not only do they end up driving traffic corresponding to the commotion, they are so ugly they sear themselves into your consciousness. My argument remains that while this may be a rare-ish problem the level of disturbance is significant and a fix would alleviate the strain.
    – Caleb
    Jun 4, 2015 at 6:46
  • 1
    "Shog gave are pretty solid; I don't doubt they are correct.", I don't their correctness either. They are still hard to believe.
    – hayd
    Jun 4, 2015 at 22:32
  • 5
    I agree with @Caleb on his answer. I think one person should not simply decide which answer appears the first (against the community). Thank you for the rm -rf example, it illustrates well. Oct 17, 2015 at 11:13
  • 3
    @Pierre I am not actually making that case here. I think one person (the asker) getting to decide what answer shows first is a good thing. The exception being discussed is in the case where the community thinks the answer is actively wrong/harmful and downvotes it. An example of not that is shog9's "accepted" showing up above mine here. That's as it should be. The only thing I'm proposing is that negatively scored accepted answers not be sorted first.
    – Caleb
    Oct 17, 2015 at 12:36
42

There's a precedent here for self-accepted self-answers whereby the order is reset to the votes. This seems like a sensible extension.

IMO downvotes are more important than the OP... just because the OP thinks it solves their problem does not mean it is correct/helpful to others.

7
  • 11
    However, I've seen many cases where a heavily upvoted answer was wrong, simply because other people thought it was correct. Turns out, the accepted answer (with minimal upvotes) actually worked.
    – user206222
    Apr 30, 2013 at 20:18
  • 5
    @KnightswhosayNi I've seen this also, but I think it is the exception rather than the rule. The same can be said for other questions (occasionally there is a gem which is non-accepted and with low votes).
    – hayd
    Apr 30, 2013 at 21:02
  • 13
    An accepted answer with lots of downvotes is suspicious. A wrong answer with many upvotes is not.
    – user206222
    Apr 30, 2013 at 21:07
  • 1
    That last sentence is a very good point (and very succinctly put). +1. Apr 30, 2013 at 21:33
  • @Knights, this is a bit of a tangent, but in the case of highly-voted answers that experienced community members know are wrong, what recourse do they have, if any? Should such answers be flagged? Should they draw attention to them on meta in the hope of getting a lot of people to downvote them? Apr 30, 2013 at 21:35
  • 1
    @KyleStrand perhaps see this question :s
    – hayd
    Apr 30, 2013 at 21:42
  • 6
    @KnightswhosayNi -- I think a minimally-upvoted accepted answer out-weighing a heavily-upvoted but not as good answer is fine. But that's not the issue here. There is a difference between a not-upvoted-very-much and a downvoted answer. The former can happen for a number of reasons (another answer got the herd-mentality-upvote, their answer was too long, short, simple, or complex compared to another, etc...) but if there's nothing clearly wrong, it usually won't garner many downvotes regardless of upvote count. A downvoted accepted answer, on the other hand, is usually horribly wrong.
    – Ben Lee
    May 2, 2013 at 20:53
37

First off, excellent suggestion, thanks for writing this up. We discussed this extensively internally, and while I'm declining it right now I don't think it's without merit — it's just not something we think is worth the added complexity right now.

What follows is a rough summary of our internal discussions, heavily biased by my own prejudices and selective memory. Most of it pertains to this question in some tangential fashion.

The meaning and behavior of "Accept"

We heavily discourage thanking people here. Not because we're rude, ungrateful people — that's an unrelated issue — but rather because it is noisy. Instead, we provide the folks asking questions with a built-in way to say "thanks" — the ability to "accept" an answer. This isn't predicated on reputation or experience, but rather on the simple fact that you had a problem and someone tried to help you solve it.

By default, Stack Exchange uses a very simple ranking system for answers. An answer's score is upvotes-downvotes, with higher-scored answers appearing first in the default sort order.

Except when there is an accepted answer. Then it always appears first. Regardless of sort order. This one little inconsistency was added as a way to highlight the importance of an answer which is presumed to have actually helped at least one person solve an actual problem they faced. In practice, other readers tend to agree with the asker in the vast majority of cases.

Note that the accepted answer can be changed at any time, for any reason, if the asker decides to do so.

Issues with pinning accepted answers

Regardless of the stated meaning, acceptance is often presumed to grant some official status as "best" or "most correct". If nothing else, it will be the answer read first by most readers. Therefore, it is somewhat embarrassing when that answer is tragically, woefully wrong.

A closely-related issue involves answers that were once very useful, but in the face of change have become out of date, obsolete, or simply less than ideal. Note that this can also be a problem with answers which were simply highly-voted during the period of time when they were correct, since votes do not age away. A related discussion on the maintenance of such answers.

Proposed solutions to the problem of bad or wrong suggested answers

Well, there's this one — unpinning when the answer score falls below some score threshold. And its slightly more complicated cousin, which wishes for downvotes to be considered by themselves. The only real issue with the former is that it adds complexity to a conceptually-simple system; the issues with the latter are detailed in the answers there.

One of the oldest (and probably most frequently-duplicated) suggestions is to allow trusted voters or moderators to change the accepted answer at-will. The primary issue with doing this is that this waters down the meaning of Accept, while a secondary one is the lack of a reliable means to select a group of users likely to know more about the topic than an asker.

A relatively unobtrusive option would be to just add a small notice to cases where the accepted answer is outranked, noting the existence of a potentially better one nearby.

And some of our devs have suggested that simply time-limiting the pinning granted by accept (say, pinned for 90 days then sorted normally) would at least prevent it from being an eyesore forever.

Rationale for doing nothing at this time

Once you establish significant thresholds (>= 10 point difference, less than 0), this affects a very small number of posts (see Appendix A, below). Adding another rule here increases complexity for new users without much offered in return (note that unpinning accepted answers already confuses folks when they find self-accepted answers).

As the example given above illustrates, simply deleting very bad answers can be an effective strategy here — this isn't always appropriate, but it does limit the potential for confusion. Of course, in cases where they can be edited without seriously deviating from the original meaning or intent, then that is preferable.

A larger issue is that of whether pinning ever makes sense for problems where there's no immediate, practical, testable solution to be had. As we continue to add more sites on less technical topics, the notion that there's any value in pinning an answer chosen by the asker becomes less sane; it may eventually make sense to disable this particular behavior entirely on some sites.

Appendix A: Accepted answer stats for Stack Overflow

  • 5,386,867 questions on Stack Overflow
  • 4,842,611 questions with at least one answer
  • 3,232,624 questions with an accepted answer
  •    344,600 questions where the accepted answer scores less than another answer
  •    239,804 questions where the accepted answer scores less than another answer after subtracting the score of that answer at the time the accepted answer was posted
  •      12,933 questions where the accepted answer scores 10 or more points less than another answer
  •        5,656 questions where the accepted answer scores less than 0 and less than another answer
  •        4,103 questions where the accepted answer scores less than 0 and another answer scores more than 0
  •           525 questions where the accepted answer scores less than 0 and 10 or more points less than another answer scoring more than 0
  •           237 questions where the accepted answer scores less than -3 and another answer scores more than 0
  • 4,017 questions where the accepted answer scores less than another answer and the author of the question no longer has an account on the site.
  • 97 questions where the accepted answer scores less than 0, less than another answer, and the author of the question no longer has an account on the site.

Gallbladder B: Accepted answer stats for Software Engineering

  • 15,982 questions with an accepted answer
  •   2,644 questions where the accepted answer scores less than another answer
  •   2,488 questions where the accepted answer scores less than another answer after subtracting the score of that answer at the time the accepted answer was posted
  •      494 questions where the accepted answer scores 10 or more points less than another answer
  •        15 questions where the accepted answer scores less than 0 and less than another answer
  •        15 questions where the accepted answer scores less than 0 and another answer scores more than 0
  •           8 questions where the accepted answer scores less than 0 and 10 or more points less than another answer scoring more than 0
  •           3 questions where the accepted answer scores less than -3 and another answer scores more than 0
  • 52 questions where the accepted answer scores less than another answer and the author of the question no longer has an account on the site.
  • 0 questions where the accepted answer scores less than 0, less than another answer, and the author of the question no longer has an account on the site.

(I would link to a SEDE query for this, but… SEDE is throwing fits today due to some ongoing maintenance.)

12
  • 5
    Gallbladder? I usually go with Epiglottis, myself.
    – mmyers
    Jul 19, 2013 at 3:36
  • 27
    Perhaps ironically, this question is currently accepted (and thus appears first) but scores 19 points less than the most highly voted answer and is hosted on a site where "there's no immediate, practical, testable solution to be had".
    – JDB
    Nov 7, 2013 at 15:18
  • If a downvote on a exempted answer stop it being exempted, there they may get more down votes. Aug 6, 2014 at 21:25
  • 4
    There is one nuance to the sorting presented here: if the OP accepts their own answer, it is given no "special" treatment; answers are in order strictly according to votes. Aug 25, 2014 at 11:22
  • @PaulDraper: I knew that I'd noticed accepted answers sometimes ranked below other answers. I just never realized what the condition was that lead to it happening. Now I know - thanks for sharing! May 17, 2015 at 15:51
  • @JDB: Not only that, both of the other answers (at this time) have more upvotes than the accepted one, and yet the accepted one shows up above it. May 17, 2015 at 15:53
  • 3
    On "increases complexity for new users without much offered in return" 1. Most SO visitors either have no account or have reputation ~1 point -- I doubt they know or even care about the current sorting rules 2. The benefit is that these visitors may find a more useful answer at the top. Personally, I like the expiration idea (perhaps based on number of views: the more views the less OP's opinion should matter)
    – jfs
    Jul 1, 2015 at 1:06
  • 1
    I only just read this in your answer "And some of our devs have suggested that simply time-limiting the pinning granted by accept (say, pinned for 90 days then sorted normally) would at least prevent it from being an eyesore forever.", and wanted to let you know that before seeing that I wrote up a proposal for "Keeping special status for Accepted Answers without sticking them to top forever?" along similar lines.
    – PolyGeo
    Nov 1, 2015 at 9:07
  • 2
    This subject keeps bothering people, has anything changed internally in the past four years? Dec 25, 2017 at 16:01
  • 2
    Lots. But, uh, nothing that would help here.\
    – Shog9
    Jan 4, 2018 at 21:35
  • That word tripped me up, and I can imagine others can get tripped up as well. A joke like that one definitely doesn't belong in a factual post. It's out of place. Jan 25, 2018 at 6:37
  • 4
    If unpinning downvoted accepted answers would confuse users, then clearly we should also not unpin accepted self-answers, right? I understand from UX folks that users are most confused by inconsistency. They learn the "answers in order of votes" rule and the "except accepted" exception, and then have to learn the 'except self-answered" exception to that. So either that's too much complexity or adding "or bad" to that last exception seems logical, no? Jan 25, 2018 at 16:18
5

I have a small remark on the stats shown in the accepted answer. If you check them today you can see that the questions with a lower scoring accepted answer are also the better viewed answers.

  • Questions: 13970299
  • Questions with accepted answer: 8570161
  • Viewcount for questions with accepted answer: 25575565677
  • Questions with lower scoring accepted answer: 716425
  • Viewcount for questions with lower scoring accepted answers: 3910020065

The questions with a lower scoring answer take up 8.3% of the questions, but when you compare viewcount they take up 15.2%. I guess thats why people think it happens more then that stats show.

If 8% is not enough, then perhaps 15% of the questions is significant enough to take some more action?

1
  • Thank you for this query. This issue came up again recently on one of my sites (Mi Yodeya), where your query shows similar proportions. Jul 13, 2018 at 15:58
2

Alternatives that seem to not have been mentioned:

Notify OP about contesting answers

One alternative I did not see mentioned here is to raise awareness of OPs about the situation of their accepted answers. I.e. when alternative answers cross some threshold, give a message to the OP to re-confirm which answer, if any, should be the accepted answer.

Allow multi-accepts

Simply allow OP to select multiple answers as solving his problem / being most helpful. All accepted answers go to the top, top-voted first. Reputation spread evenly, somehow. This might be limited to 2 answers.

This also reduces the unfairness of accepting rewarding one person who put in effort, but not a second person putting in similar effort.

Consider highest-voted as "community-accepted" in addition to "OP-Accepted"

Similar to previous, allow a question to have 2 accepted answers, one by the OP, the other as highest voted (with many votes) by the community (Probably using some threshold to prevent junk from being marked as accepted). This also solves the case of rewarding reputation where the OP cannot be bothered to click accept.

De-prioritize based on flags

Another option here is to disable the priority of accepted answers for questions flagged with "primarily opinion-based", and possibly other flags. If the community thinks this is an opinion-based question, why should there be one accepted answer based on OPs opinion trumping other opinions?

This can be extended to add additional flags with a similar effect, such as flagging accepted answers with "Contested answer" or "Wrongly accepted", removing it's sorting privilege (but nothing else).

1

Instead of pinning the accepted answer unconditionally ar the top it will be better to just add certain bonus number of points (e.g. 10/15/20) to the answer and then order answers by points number normally. The actual number of added points can be shown next to the tick. The number of points can be different for different sites(for small sites the number may be less).

Update:the similar idea was mentioned in Modified answer view based only on votes, not accepted status

0

Can order of answered be customisable per user? I.e. user should have option to specify “in list of answeres show “accepted answer” first” and be able to untick it.

Update:the similar idea was already suggested in incorrectly closed as duplicate Allow higher scoring unaccepted answers to sort above the accepted answer

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