# What's the policy on responding to other answers in an answer?

In this question Aeroalias provided an answer, but later edited some extra text as a response to an answer given by Peter Kämpf.

I was wondering if this is the right thing to do? Basically, I think all answers should answer the question and nothing else (like giving a reasoning why any other answers are wrong). The voting system will then show which of the answers is considered best.

I've seen this more often, and so I'm wondering if this wanted behavior or not.

• Voting doesn't quite do what people think it does. First, a decent older answer will almost always have more votes than a better new one and this is perpetuated because, even though the new answer bumps the question and gets more attention to it, a lot of that attention is directed to upvoting that old, decent answer. Second, matters of actual truth aren't subject to democracy. For example, on another SE site, I had an answer voted up to about +10 before somebody noticed that it was subtly incorrect. Their one downvote invalidated all the upvotes even though it only decreased the score by one. Nov 2 '15 at 9:35
• About the truth, unlike other SE sites like StackOverflow where it's easy to verify answers using an execution of the script, here at Aviation a lot of answers cannot be factually verified. The next best thing we have is the peer reviewed answer (i.e. the most upvoted answer). But I agree, that it's a bat thing that a previously correct but invalidated answer still stays up high on the page. Nov 2 '15 at 9:47
• It's still possible that an answer could be invalidated by a factual inaccuracy. But I agree that it's more of an issue on math-y SEs. (For example, "Has there ever been a plane with feature X?" "No, because feature X would be terrible for reasons A, B, C," with all those reasons looking plausible would be invalidated by "Actually, the Cessna Thingummybob had an X. It worked just fine.") Nov 2 '15 at 10:08

I'd much rather see a statement along the lines of "A common misconception is that...but..." and keep it generic and self-contained than referring to another specific answer. If you do feel the need to comment on a specific answer, that is exactly what comments are for.

You also have the option of editing the other answer (once you reach sufficient reputation) as long as it doesn't change the meaning of their answer.

Personally, if there is an answer that I find useful, I make sure to read the comments to see if there are any dissenting views or clarifications that other people have added.

• Along the same lines, if you're referencing another answer it should be cited appropriately (by linking to it: Copy the share link and use that in your citation. e.g. this answer, or Jay's answer) Oct 22 '15 at 16:39

Personally I think it's a great idea to respond to other answers in your answer, but I'm pretty sure I'm going to be in the minority on this one.

Why? Well, the main reason is because often people don't read the comments to see if there are counter arguments to an incorrect answer. Heck, sometimes the system hides comments for us... So putting a specific refutation in another answer (in my mind) is a perfectly legitimate way to overcome this shortcoming.

That being said, I wouldn't be a fan of simply having the refutation as the entire answer. I think if you are going to refute someone else in your own answer, you'd better have an answer as well. Something that is only a refutation should be left as a comment (despite all the afore mentioned difficulties with that system.)

I have offered other, far cleaner, solutions before. But as you can see from clicking the link, there's a lot of hatred towards that particular solution. Pity, because it's the most obvious thing to do if you ask me. (Read the whole post if you want all of my thoughts on why the current system of upvotes and downvotes has some flaws...)

• The problem I have with "specific refutation" of someone else's answer is that is can often devolve into hostile bickering directly contrary to the Stack Exchange "be nice" rule - comments that run afoul of that rule are nuked more often than I'd like, but at least those are comments and designed to be ephemeral. Voting has its problems, but burying Incorrect answers with downvotes avoids devolving into a "discussion forum") Oct 22 '15 at 16:41
• @voretaq7 I see your point, but at the same time...sometimes you've got to point out that someone is wrong. I think that doing so nicely can take a little skill, but I would also argue that it's a skill worth cultivating. I don't think it should stop people from correcting something they know is wrong. Oct 22 '15 at 17:38
• Oh I'm all for pointing out an inaccuracies in an answer, I just prefer it be in the comments, or if the answer is altogether wrong I'd post a more correct one that stands on its own. (When there's a good answer and a wrong answer a good solution is to call forth the Hangar Voting Ring to make sure answer scores are "adjusted" to reflect which one is best. Ultimately votes should sort out on their own but sometimes it's OK to give them a little "help" from the site's regulars...) Oct 22 '15 at 17:49
• @voretaq7 Fair enough, though I still wish they'd just use the system I proposed (and I'm not in any way still bitter about how I was treated just for suggesting it, nope, not at all, no bitterness here.) Oct 22 '15 at 20:13
• @voretaq7 "The problem I have with 'specific refutation' of someone else's answer is that is can often devolve into hostile bickering" That's a problem with hostile bickerers, not with refuting answers. Nov 2 '15 at 9:37
• @DavidRicherby One often begets the other as people can't keep just a little bit of inappropriate snark from creeping into their refutations (see any political campaign for abundant examples). We can't fix human nature, so personally I try to avoid the problem by just saying "It's not a discussion forum, so just let your answers be answers." Nov 2 '15 at 17:29