4
$\begingroup$

Stack Overflow is not pinning the accepted answer to the top any longer.

What would we like for our site?

  • Keep things as they are, as the right answer is not changing in aviation as it does for code.
  • or change to the new method, because the best answers are not always the accepted ones?
$\endgroup$
8
$\begingroup$

I would propose to change to the new method:

Unpin the accepted answer and show the top voted answer first.


I added my reasoning below. If you agree with changing to the new method, feel free to add your reasoning below. If you disagree, please post another answer so people can vote on it.


The whole idea of accepting answers does not really make sense on Aviation.SE. For a programming site, the idea is this:

  1. You have a problem and post a question on StackOverflow.
  2. Multiple users post their solutions to the problem as answers.
  3. Users vote on how useful each of these solutions are.
  4. You end up using one of their proposed solutions in your code and accept that answer.

Most questions on Aviation.SE do not follow this principle since there is no specific problem to be solved. We recently modified our What types of questions should I avoid asking? help center page because of this (see Proposed edit to "Don't ask" help page).

Without a specific problem, there is no solution that was implemented. This leaves users with choosing the best answer as the accepted one, which often depends on personal preference.


The tick has always meant most helpful to OP, not correct/approved/best, so the new system in my opinion removes this ambiguity.

Accepting an answer is not meant to be a definitive and final statement indicating that the question has now been answered perfectly. It simply means that the author received an answer that worked for them personally. (Help Center)

$\endgroup$
1
  • $\begingroup$ I upvoted this answer (and TomMcW's), but with some hesitation. I think it's a worthwhile experiment, but if it goes awry, there needs to be an abort lever for the pilots (mods/CMs/sysadmins in this case) to pull. $\endgroup$ Oct 12 at 9:56
1
$\begingroup$

I agree that the new method should be preferred. Sometimes the OP accepts an answer too quickly and better answers may come along.

$\endgroup$
2
  • $\begingroup$ Since it's the same choice as @Bianfable's answer, be sure to upvote it instead, and you can leave that answer as a comment. Will make it easier to wrap up the consensus :) $\endgroup$
    – ymb1
    Oct 10 at 16:08
  • 2
    $\begingroup$ @ymb1 I turned my answer into community wiki so others can add their reasons for changing to the new method. That should simplify the voting... $\endgroup$
    – Bianfable
    Oct 10 at 18:24
-1
$\begingroup$

Personally, I'd prefer to keep the existing accepted answer pinned behavior.

The primary reason for this is that it's quite common for someone to post a superior answer after others have already attracted upvotes. Under the newly-proposed system where accepted answers aren't pinned, the new, better answer would often languish in obscurity beneath possibly several answers that were posted sooner and already had some upvotes.

This effect is especially strong when a good answer is posted more than a couple of days after the question was posted. The person who posted the question will get a notification when the new answer is posted - and possibly choose to change the accepted answer if they think it's better - but most others would never see it if there were already at least a few other answers with some upvotes, even if only a few. This is not helped by the fact that questions tend to get a lot more visibility soon after they are posted.

By moving the accepted answer to the top, the later answer is pulled out of obscurity at the bottom to a position where it's much more likely to be seen (and voted on) by others.

The worst that can happen with the current (accepted answer pinned) behavior is that one lower-voted answer appears above the others, while the remainder remain sorted by votes. Whereas the worst that can happen by unpinning is that the best answer is hidden under a pile of answers that aren't quite as helpful, but were posted earlier.


What I'd really like to see is an option to pin the accepted answer above all answers that are older than it, but, unfortunately, that's not one of our options here, so I think that maintaining the current behavior of accepted answers being pinned is the best of the two options presently available to us. Hopefully SE will come up with a better sorting algorithm that takes age of votes into account later on. Possibly they will come up with something better in 6 to 8 weeks.

$\endgroup$
3
  • $\begingroup$ A month ago I voiced the same primary concern on the main meta, but the staff's response was encouraging (this is only the beginning of fixing things). Try this SEDE query for Aviation, and I'm sure you'll find many examples where the lesser answer is pinned (and that's the main issue: pinning is subjective). The staff's response and that query helped me see the potential of the new system. $\endgroup$
    – ymb1
    Oct 11 at 9:20
  • $\begingroup$ @ymb1 Yes, I also voiced that concern on SE or SO meta (I forget which) and did see that they are planning to eventually come up with something better. As for the SEDE query, I've seen similar query results of higher-voted answers below the accepted one cited as evidence for the change, but that by itself really isn't, as it could equally well be evidence of exactly the problem this answer is describing - vote bias towards earlier answers. I'm sure there are examples both ways, but, clicking through those results, I'm seeing several that were just that - better answers posted later. $\endgroup$
    – reirab
    Oct 11 at 10:06
  • $\begingroup$ Still, there are several that aren't better: the tick has always meant most helpful to OP, not correct/approved/best, so the new system IMO removes this ambiguity. Also possibly if Aviation doesn't join now, we might miss whatever they'll come up with to fix the other concerns. (BTW, -1 not mine :) ) $\endgroup$
    – ymb1
    Oct 11 at 10:31

You must log in to answer this question.

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