2
$\begingroup$

What to do with answers to questions which are obvious and exact duplicates of existing questions?

I'm talking about this review I recently did.

The answer was to one of those limit/inequality/calculus questions which comes up very often (in every calculus course at least) and in this case had an exact duplicate with $6$ answers. The question was furthermore already marked as a duplicate, less than two hours after its posting. The existing question had the exact same formula in the title. I went to the question, had a look at the given answers and quickly found an answer giving the same solution as the one I was reviewing, and another one, pretty much the same.

I then, before reviewing, actually proceeded to google what to do in this situation, as the answer I was editing was clearly redundant, so I found this thread on the meta site. The accepted answer said this:

Does it provide information in a more understandable way than existing posts? Of course some repetition with different wording helps search, but there's a line past which the material can start to get smeared out. Finding the bit you need becomes a hunt through many questions, just like the old sifting of endless forum threads that SO is supposed to obviate. Therefore, I think one should ask: does an answer to this question make info easier to find, or are we past the tipping point?

I then assessed that the post definitely does not add any extra information, and in fact is very bare and offers no clarification whatsoever (which is also evident from the comments of the user who asked the question). The equivalent answer in the other thread also had some additional details, such as the core of the argument, which is that $\frac{1}{\sqrt{n}}$ is $\le$ than all the terms in the sum.

Then there was this strange message:

Our system has identified this post as possible spam; please review carefully

I took it to mean I should be extra diligent and critical in evaluating the answer, as some system has marked it as spam. The system, I assumed, was precisely looking at answers to duplicate questions, and comparing them with the answers to the original question.

Also, the fact that the answerer had > $42\textrm{k}$ reputation means that he should have been more than aware of the fact that he could be answering a duplicate, and didn't look for it before posting. This, coupled with the lack of details in the answer, made it look really bad, IMO.

I proceeded to recommend it to be closed as spam (as the system suggested, since none of the other reasons seemed fit). Turns out, the review was in fact an audit, which I failed. The answer should have "looked ok" to me.

My question is:

What to do with those answers? Was I plainly wrong? The thread on the meta site I talked about, along with this one seem to suggest that the community in general dislikes/frowns upon answering obvious duplicates. Granted, neither threads explicitly suggested flagging and deleting such answers, however downvotes were suggested by some. Probably I was fooled by the "possible spam" so I acted too harshly. This thread also suggests that the repeated answer should at least have some added value.

The problem here is that the system currently does not penalize such behaviour. In fact, the answer in question probably "looked good" because it had $7$ upvotes! The user who posted the question is also not disciplined, as he/she got the $+2$ for accepting the answer in question, even though the question itself was not upvoted.

My concern here is not so much about the audit itself, which was however very unintuitive, and a strange choice for a test. I would expect some inexperienced users (like myself) wouldn't even try to search for the correct action online as it arguably seems obvious. I'm concerned more about why should this practice be allowed/not disciplined especially coming from experienced users?

An answer in this thread about a connected issue of answering off-topic questions, states that off-topic questions shouldn't be answered as it encourages users to ask them. The situation is the same here, as answering duplicates encourages users not to look for existing questions before posting new ones. Also, subsequent good reception of those repeated answers encourages users to answer first, and look for duplicates later, which could harm the overall quality of the site.

$\endgroup$
  • 5
    $\begingroup$ Unless I missed something you did everything right. Unfortunately the review audit system has these flaws. Also, we have too many users who care more about their rep score than site hygiene. In their defence, it is not always easy to locate those duplicates, so we tend to be forgiving. The incentive systems in place on our site simply make it quite hard to remember to spend the necessary amount of time searching. Many of us make this mistake occasionally. The good book says something that I should keep in mind myself. $\endgroup$ – Jyrki Lahtonen Sep 22 '17 at 16:35
  • 2
    $\begingroup$ I agree; I think you did everything right. Given @Jyrki's comment and the Reference he links to, I'll refrain from judging those who answer a question (or even rush to do so) before someone's dupe hammer renders the question closed, or 5 users do so, even when they see a link to a possible duplicate!. But let's just say, you've got many supporters on this matter! $\endgroup$ – amWhy Sep 22 '17 at 20:41
  • 1
    $\begingroup$ Note that we may need to remind answerers, just as we do askers, "How do write a good answer. Note, wrt what not to answer, we have listed: $\endgroup$ – amWhy Sep 22 '17 at 20:54
  • 5
    $\begingroup$ "Not all questions can or should be answered here. Save yourself some frustration and avoid trying to answer questions which... " ...are unclear or lacking specific details that can uniquely identify the problem. ...solicit opinions rather than facts. ...**have already been asked and answered many times before.** ...require too much guidance for you to answer in full, or request answers to multiple questions. ...are not about math as defined in the help center." " $\endgroup$ – amWhy Sep 22 '17 at 20:54
  • 3
    $\begingroup$ Amen to those points @amWhy. $\endgroup$ – Jyrki Lahtonen Sep 22 '17 at 22:32
  • $\begingroup$ @JyrkiLahtonen Ok, so it is good that I reported this audit here on meta? Should I do this every time I come across a dubious audit? About locating duplicates, this thread, among others, says that it is not really users' duty to look for duplicates before answering. However, I acted the way I acted since I perceived the answer as already flagged. Also the nature of the question and the quality of the answer were a factor. $\endgroup$ – mechanodroid Sep 23 '17 at 10:01
  • $\begingroup$ @amWhy Yeah, one could argue that most easily answerable questions were already in some form asked and answered on this site so banning users from answering such would prevent the site from continuing to function. However, existing questions of this type with a formula in the title should be looked up first, especially these first-calculus-course ones. $\endgroup$ – mechanodroid Sep 23 '17 at 10:04
  • 2
    $\begingroup$ Also, I think re-asking well-known questions is perfectly fine, if the user puts in the question something like "I'm aware of these questions: [link] and [link], but they do not answer my additional question $X$ and I would like someone to give more details about $Y$". $\endgroup$ – mechanodroid Sep 23 '17 at 10:08
  • $\begingroup$ mechanodroid: Duty or not? The discussion you linked to was at another site. And even there, a more upvoted answer suggests that this a grey area. That seems to be close to the thinking of users of Math.SE, BUT THERE IS A LOT OF VARIETY. I am known to be almost militantly anti-duplicate. But I have accidentally answered duplicates myself. Particularly when wandering to a tag I don't frequent. That's why I should keep the sermon on the mount in mind :-) $\endgroup$ – Jyrki Lahtonen Sep 23 '17 at 16:57
  • 1
    $\begingroup$ @JyrkiLahtonen I get it, it depends on the question itself. $\endgroup$ – mechanodroid Sep 23 '17 at 17:00
  • 3
    $\begingroup$ Indeed, it does. The frustration comes from seeing high rep users answer essential duplicates dozens of times. Unless they suffer from total amnesia it is hard to avoid the impression that they knowingly answer duplicates to farm more and more points $\endgroup$ – Jyrki Lahtonen Sep 23 '17 at 17:05
  • 2
    $\begingroup$ @JyrkiLahtonen Frustration$^2$: To see the same users justifying ferociously their act and branding as fascistic elitists those who dare to object. $\endgroup$ – Did Sep 24 '17 at 15:08

You must log in to answer this question.

Browse other questions tagged .