7

Just wondering but I've seen it happen a number of times, but when answers get blatantly obvious duplicate/repeat answers from multiple people. Now, I normally just upvote the first person if the answer is good, but take this question for instance.

What's the etiquette for dealing with repeat answers?

  • 1
    The only "etiquette" I try to follow is comment on down-vote or up-vote a comment when the reason were posted by other user. – Maniero Jan 2 '11 at 0:33
6

On this site, people are incredibly liberal with voting: almost all answers that are remotely related to the question are upvoted, and any downvoting is called into question as being rude, mean, or uncouth. If you want to follow etiquette, you should upvote all the answers.

However, your own personal preference in how you vote isn't going to make or break a person's reputation total. So feel free to ignore the etiquette and follow the overarching principle of upvote answers you think are useful, downvote answers you think are not useful. Of course, only you can tell what's useful to you. If answers duplicating already perfectly suitable answers are useless to you, downvote without pity, or remorse, or fear and upvote only those answers you personally found of use.

To give you my own criteria in the situation you're describing:

I know that a lot of times, even so-called "subjective" questions really only have one real answer that people understand in slightly different ways. It's possible that people act in bad faith when answering (hoping you won't see they just duplicated an earlier answer for quick rep points), but I think most people are thinking that their answer is truly novel because it's not an exact copy-and-paste of an earlier answer.

Because of that, I tend to upvote the answers in which people explain their positions well. This dovetails nicely into the Six Subjective Guidelines which encourage questions to invite long answers that share reasons, experiences, and facts. It's not simply enough that one answers a question: we get enough answers already. For an answer to be useful here, it needs to provide more.

So if a person says something first but provides no detail, and a second person provides the same answer cogently (that is, they provided useful reasons why one ought to believe the answer is true), I'll always upvote the second person and I'll tend to downvote the first person.

If, however, the first person gave a really useful answer and the second person didn't provide any new insights to the answer, I'll upvote the first and downvote the second as the second isn't useful.

If both people have the same answer, but provide really great and different reasons for arriving at that answer, I'll upvote them both. Finally, if both answers lack any explanation, I'll downvote them both.

  • 1
    You are a phenomenom :) – user2567 Jan 1 '11 at 21:48
  • 4
    I will add that your almost 50/50 upvote/downvote ratio is a bit on the high side. Based on what I've seen on the site, I think downvotes should probably be 5%-10% of a person's total number of votes. Most questions that are downvoted should instead be closed, and most answers that are downvoted are going to hang around the bottom no matter what anyway. – Macneil Jan 1 '11 at 22:35
  • 3
    @Macneil in the same respect, you're a perfect example of what I'm describing in the first paragraph of my answer. Your 948 upvotes means you used your entire daily allotment of votes to upvote 31 days out of the 64 you've been active on the site. – user8 Jan 1 '11 at 22:38
  • 3
    You'll also note that I have the "Electorate" badge, which means at least 600 of my votes were on questions. When we were in Beta I had the privilege to close questions. When we went public, indeed, I almost doubled the total amount of downvotes I've made, for the very bad questions. Now that I can vote to close again, I will downvote less. Anyway, I think upvotes and stars serve the purpose of filtering the noise well enough on their own. The downvote is really the "nuclear option" when someone is clearly doing something wrong. – Macneil Jan 1 '11 at 23:16
  • 3
    Let me add also that I support downvotes with explanations, where, if the explanation is satisfactorily addressed, that the downvote should be cleared. – Macneil Jan 1 '11 at 23:18
4

This also implies the question is bad, that it's a flawed "let's make a list of X" type question.

And let's face it, when a question has 3 pages of answers, who actually reads all three pages before answering?

This is also why I favor length as a predictor of answer quality -- if people are sharing specific experiences (eg "the last time I tried peanut butter, {this happened}" rather than "try peanut butter!" then there can't be duplicates in that sense.

  • 2
    So, should SnOrfus's question be revised to be "How have you tried to monetize free software?" instead of "What are the common revenue models for free-software companies?"? – Macneil Jan 2 '11 at 0:17
  • @mac not necessarily, as the answers for that question are good, length-wise. I also don't see 30+ answers for that question. It really depends how many of the six criteria the question meets blog.stackoverflow.com/2010/09/good-subjective-bad-subjective – Jeff Atwood Jan 2 '11 at 0:47
3

You should probably notify the person that their answer duplicates a previous one, and provide a link. They may then decide to delete their answer. If not, then at least you've made a pointer, so that people who would have otherwise upvoted that would upvote the "right" one.

The purpose of voting is to strengthen the signal and reduce the noise. But such activities should also take the whole ecosystem into account: Posting a notice that something seems like a duplicate answer will help those in the community improve. Downvoting a sincere answer can poison the well and we could lose participation from people who would have really helped. (I have seen a similar poisoning of the well on Wikipedia, which has probably peaked in terms of high-quality edits on non-pop-culture articles.)

1

I find it slightly annoying too. I think I was the second person to answer that question, and I remember the first answer posting when I was still finishing my answer. So, I stopped a second and read it.

I revised my answer a bit to be less redundant and opted to expand a bit more on the part of my answer that wasn't redundant.

I typically don't bother with questions that have three pages of answers, unless all of those answers are outdated and I can't simply edit one of the top answers for some reason. One of the reasons why I love SE sites is that the best information typically rises to the top. I hate forums, so I think I naturally avoid going to the third page of anything, much less contributing to the mess myself.

What you can do is answer the question yourself with something that compiles the best answers (with citations) together, then mark your own answer as correct so it floats to the top.

You must log in to answer this question.

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