17
$\begingroup$

What should I do if I see a competing answer that is submitted around 10-15 minutes later with very similar or the exact same content just rephrased or missing/adding a few intermediary steps? I find the chance that these answers are coincidentally the same very slim considering the time difference. There usually is no new information presented in these similar answers.

I don't think it's really malicious (though some may see it as just another way to grab some quick imaginary internet points), especially if the initial answer already has a few upvotes, but the fact there are "duplicate" answers seems like a cause for concern, since it seems like other places across the network (SO) highly discourage this behavior, utilizing custom flags to remove this content (since duplicate answers that add nothing new are useless).

Should I act or just ignore this?

$\endgroup$
  • 2
    $\begingroup$ can you give us some examples? $\endgroup$ – miracle173 Mar 17 '18 at 7:08
  • $\begingroup$ Some simple questions have a relatively unique answer, so it is likely that in the rush to grab MSE gold that similar answers appear. $\endgroup$ – copper.hat Mar 17 '18 at 17:52
  • $\begingroup$ @copper.hat I understand that, just seems especially nasty when there's already an existing answer that has the same if not more info that's posted well outside the timeframe of coincidence $\endgroup$ – Andrew Li Mar 17 '18 at 17:56
  • 3
    $\begingroup$ It is a little irksome. MSE needs some machine learning to highlight essentially same answers :-). I am often curious why a later answer, essentially the same as another, often gets accepted. Perhaps the OP just selects the top answer that works? $\endgroup$ – copper.hat Mar 17 '18 at 17:59
  • 1
    $\begingroup$ But the internet points are natural! :D $\endgroup$ – SK19 Mar 19 '18 at 20:56
  • 1
    $\begingroup$ I find it annoying when an answer is given that seems to ignore the existence of prior comments/answers, particularly when the content is similar and plenty of time has elapsed. It never hurts to say something like "similar to the xyz comment,..." or to add a line that shows the prior answer appeared while you were typing yours. I sometimes delete my answers that come just a bit late and would be redundant. Sometimes I even delete my answers that come first when I suspect a second-responder will not bother to delete theirs. $\endgroup$ – Michael Mar 19 '18 at 21:37
  • $\begingroup$ @copper.hat Some new users also accept every (good) answer they get, not knowing that only the last one has effect. Other than that, even subtle differences in wording can make a world of difference to someone not already familiar with the topic. $\endgroup$ – Joonas Ilmavirta Mar 21 '18 at 0:17
  • $\begingroup$ @Michael: prior comments? you shouldn't post answers in comments! This is a misuse of comments. you should post answers as answers and that may need more time than writing a comment. $\endgroup$ – miracle173 Mar 21 '18 at 3:42
  • 1
    $\begingroup$ @miracle173 : I don't understand what you are objecting to. I was advocating not ignoring prior answers and comments. Are you suggesting one should ignore prior comments (and/or answers)? I did not mention posting answers in comments. Nevertheless, I do often give hints in comments, and I expect the asker to fill in the rest of the homework themselves, only to find another answerer has jumped in with a full answer, often without giving the asker any time to think on their own. $\endgroup$ – Michael Mar 21 '18 at 4:04
  • $\begingroup$ @JoonasIlmavirta: Many times I have written something (not on MSE) and thought I did a good job only to revisit later and wonder what I was thinking when I wrote it originally. I was hoping to glean some insight from the MSE responses, but nothing obvious leaps out at me :-(. $\endgroup$ – copper.hat Mar 21 '18 at 4:14
  • $\begingroup$ @AndrewLi I think it's uncharitable to ascribe similar answers appearing even many minutes later to dishonesty. I've been in a similar situation before because I typed up my answer on a phone when I was outside my home. When I started, there were no answers. When I was done there were at least two, and one was strikingly similar to mine. Someone actually commented to that effect, and I had to explain that it was hard to type MathJax on a mobile phone to which a somewhat sarcastic reply (to the effect of "real struggles, man") was made. I took it in stride. $\endgroup$ – Deepak Mar 23 '18 at 7:59
  • $\begingroup$ @Deepak I agree; I spend most of my Stack Overflow/MSE time on my phone. I used to write 90% of my answer on my phone the coincidence of answers was common. But I'm talking about situations where coincidence seems highly unlikely, and the "late-duplicate-posting" happens habitually. $\endgroup$ – Andrew Li Mar 24 '18 at 0:31
  • 1
    $\begingroup$ Can editing an answer take longer than 10-15 min? Should one during editing continuously check if other answers were added? As long as you cannot exclude yes/yes one should not take any means, it would impose more pressure to users. $\endgroup$ – Rudi_Birnbaum Mar 27 '18 at 8:07
  • $\begingroup$ I think 15 min is too short time to say that the later answer is a copy. I'm not good at the syntax for writing a nice math formula so I can easily spend more than 15 minutes on answer. $\endgroup$ – 4386427 Mar 28 '18 at 11:50
  • 3
    $\begingroup$ Possible duplicate of Questions with multiple, very similar answers $\endgroup$ – Famous Michael Wang Apr 30 '18 at 14:38
41
$\begingroup$

If I am working on an answer, and I see the "New Answer" display, I keep working until my answer is done and then submit it.

If the new answer is correct, and especially if it is similar to mine, I upvote it.

I never remove an answer of mine just because it is similar to another, and I see nothing wrong in similar answers if they were derived independently.

$\endgroup$
  • 7
    $\begingroup$ My answer appearing 15 minutes later can happen. When I read the main page, I open interesting-looking questions in their own tabs. Then I go through the tabs. It may be 15 minutes before I come to the tab with your question; if I answer it based on what I see, then my answer could be 15 minutes later than another answer that I had not read. But in fact sometimes I do remove an answer that turns out to be a duplicate. $\endgroup$ – GEdgar Mar 21 '18 at 13:35
  • $\begingroup$ There is a similar answer to this meta question which has now been delted by owner. Perhaps this serves to illustrate the expected behavior. $\endgroup$ – Paramanand Singh Mar 22 '18 at 10:26
  • 6
    $\begingroup$ That answer is so similar to mine that I thought that it might be a parody or trolling. $\endgroup$ – marty cohen Mar 22 '18 at 16:35
16
$\begingroup$

I am of the opinion that this varies widely from case to case, and is very subjective. "Extremely similar" can be not so crystal clear. Sometimes, two answers with the same content have a different way of discourse which puts them apart. I think there is merit in the opinion that "redundancy is not necessarily bad", particularly in mathematics. This is something to keep in mind when judging these situations.

That said, if you think the answers are quite similar, I think it can be politely pointed out in the comments something like "Your answer seems to be pretty similar to X's. Can you consider adding some new perspective?" or something like that. I've seen this done, with low backlash.

If the case seems bad enough and you feel that the addition of that answer is not useful at all, you can downvote to indicate that (it is the most straightforward motivation for a downvote, as indicated when you hover the mouse over it). And this seems to be effective: if the answerer is just fishing for points, he unconsciously realizes that his answer is not that useful, and a downvote may make the penny drop. If he is not fishing for points, he will try to improve or justify why his answer is useful, and you may change your perspective.

The time gap is also relevant, and the "grace period" dependent on the question. 10-15 minutes later can be too much or not that much, depending on the question. It is really a case-by-case situation in my opinion. But, to give an objective answer and a rule of thumb:

If the answers are extremely similar to the point that the new answer adds nothing of value, consider downvoting the answer.

If the answer is a blatant copy in a literal sense (word by word), then flagging it may be appropriate.

$\endgroup$
  • 3
    $\begingroup$ +1 Indeed "if the answer is a blatant copy in a literal sense", flagging it for moderator attention as plagiarism is appropriate. $\endgroup$ – hardmath Mar 16 '18 at 14:53
  • 2
    $\begingroup$ I don't always check immediately,but sometimes I will see an answer that looks a lot like mine. If it is clear that my answer came second, I will delete my answer. But if both answer were posted "1 hour ago", I have no way of knowing. $\endgroup$ – steven gregory Mar 16 '18 at 17:59
  • 12
    $\begingroup$ @stevengregory: If you hover the mouse over the "1 hour ago", you'll get the exact time stamp as tooltip (for example, you posted your comment at 17:59:39 UTC). $\endgroup$ – celtschk Mar 16 '18 at 18:50
  • $\begingroup$ @stevengregory: You can also change your answer sort mode to "oldest first" by clicking the "oldest" tab above the answers. That way, the answers will appear in the order they were posted in. (And, of course, you can change it back to "highest voted first" by clicking the "votes" tab, if you normally prefer it that way.) $\endgroup$ – Ilmari Karonen Mar 22 '18 at 19:45

You must log in to answer this question.

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