12
$\begingroup$

EDIT: After amWhy brought the term "abstract duplicates" to my attention in the comments below, I thought some tidying-up was in order. The original body of the question is mostly intact below; I have removed one paragraph containing a link to another meta question, which does not seem as relevant any more, in light of these:

Coping with *abstract* duplicate questions.
The highest-scoring and containing the largest sample of user opinions on this topic question. Most answers and comments are from 6 years ago. Out of that discussion arose the idea to create wiki-style generalized answers for such questions, and keep a catalog for easy reference here:
List of Generalizations of Common Questions

While in no way implying that this list is the ultimate solution to this issue, I would like to keep this question open, as a place to discuss (among other things) ways to make the best use of it.

[To be perfectly honest, the original intended focus for my question was somewhat different: how to induce posters of new questions to search first. I am considering forking that into a new question, where I will make that clearer. My reasoning for separating that discussion from this one is that not all methods of coping with abstract duplicates will necessarily have a positive effect on the adoption of search-before-asking practices; in other words, there may be a conflict of interest here.]


Original wording of question:

I know there have been a lot of questions here on meta discussing what should be done with questions that provide no context, or loosely speaking, don't indicate that the asker has given much effort to solving the problem themselves. What the upvote/downvote guidelines call "shows no research effort."

Now, it would be embarrassing if this present question itself were guilty of exactly what it complains about, so let me make clear how what I am asking is different from all those (or at least as far as I managed to find). When you go to ask a new question, if that is your first time, you are strongly encouraged to follow these steps:

  1. Make your question clear and precise

  2. Show what progress you have made with it yourself

This second one is what I have seen debated to great lengths (what to do with apparent copy-paste homework questions, CPHQs). But there is another one:

  1. Search for previous questions that might address your difficulties

And this is what I want to discuss here. Let me illustrate with an example. In the probability and combinatorics tags, there are hundreds of questions of the kind "in how many ways can [...] be distributed/arranged/divided so that [...]" or "what is the probability that if random [...] is/are drawn from [...] then [...] is true." Many of those are, while not exact duplicates of things that have been asked before, certainly close enough, so that if the asker looked at some of those previously answered questions, he or she could at the very least get some ideas on how to tackle the problem.

Now, I try to put myself in the asker's shoes to see why they don't do that. I myself, when I encounter any problem I don't know what to do about (not limited to math), always try searching before asking on forums; one of the reasons is that if the answer is out there, I'd have it right away rather than have to wait for someone who can help (and is willing to help) to come across my question. But this "deterrent" doesn't work when there are many users here on MSE who jump at the opportunity to provide full detailed answers (to what for them is a straightforward question) within minutes. So the askers "learn" that there is no need to bother searching: it's faster just to ask, and you'll have your answer in no time.

So my questions are two:

  1. What should be done with questions which are endless variations on the same themes that have been asked many times before? Flag them as possible duplicates? Downvote as "no research effort" (even if the asker did share some thoughts on the problem)?

  2. What can I (or the community as a whole) do to discourage "immediate answer dumping" (that's what I call it in my mind; the word "dump" is not meant offensively, but as a synonym of "output") to these repetitive questions, and thereby encourage the askers to search for similar ones first?

$\endgroup$
  • 1
    $\begingroup$ 1) depending on how closely the variant matches, close as duplicate (or not) Point out the similar question(s) in a comment to the user. Remind them to search first next time. 2) There isn't much to be done to change other people's answering behavior. And sometimes the answerers as well as the posters have trouble finding duplicates. It's within everyone's purview to withhold upvotes or maybe downvote, depending on their believe system. (I simply wouldn't upvote a solution that was a complete duplicate of a solution at a duplicate question.) $\endgroup$ – rschwieb Oct 3 '17 at 19:25
  • 9
    $\begingroup$ Finding duplicates before you've posted a Question is often a matter of luck. Experienced users, esp. those who have answered similar posts, might spot the duplication immediately (or at least suspect it) where new users may be doing all they can to articulate the problem. Good tags play a role in the system generated Related links (as do words from the title), and these are precisely the choices that new users may have difficulty making. So I tend to sympathize with new users who post even "obvious" duplicates in the eyes of higher rep users. $\endgroup$ – hardmath Oct 3 '17 at 21:04
  • 2
    $\begingroup$ @hardmath I am not talking about exact or obvious duplicates though. Questions where the basic principles are the same. As in: there are a lot of people who prefer to give a hint in a comment rather than provide a complete solution - which is great! - but they end up giving the same hints over and over and over... and today I came across this question where the poster had actually found a relevant previous question, got confused by something in the question itself, but didn't bother to read the comments to that question which explain it all rather well! $\endgroup$ – Nick Pavlov Oct 3 '17 at 21:13
  • 4
    $\begingroup$ @rschwieb I feel that what you recommend borders on enabling. The burden of at least attempting to find relevant help that already exists should be on the asker, not on the responders. And pointing out that a search could have been done, when dealing with a lazy asker, will inevitably be met with the claim "I did, but I found nothing." Maybe some of them did, indeed, but I'd bet the vast majority didn't, because they don't see the point. So what I see is a lot of people who don't care to take responsibility for their own learning taking advantage of a lot of other people who value ... (tbc) $\endgroup$ – Nick Pavlov Oct 3 '17 at 21:38
  • 3
    $\begingroup$ @rschwieb (continued) ... reputation points more than their own time. Fine, that's their choice, we can't change that, but as a teacher, I can't just accept that it's normal for people like my students to get accustomed to the idea that there's always going to be someone around to do their work for them. $\endgroup$ – Nick Pavlov Oct 3 '17 at 21:39
  • $\begingroup$ @NickPavlov I fail to see how suggesting to posters they should search first enables people to post without searching, but I guess you just meant it didn't go far enough. My advice certainly can't hurt... if one lazy asker out of many picks up the searching habit, I count that as a win. It certainly isn't making lazy posters produce lazy posts any faster. I get the feeling you imagine we can influence posters behaviors in ways that I have trouble finding feasible. What would you recommend instead? $\endgroup$ – rschwieb Oct 3 '17 at 21:51
  • 1
    $\begingroup$ @rschwieb Well, I suppose what I am trying to do is raise awareness. I've only been a member here for a short while and I am astounded by the multitude of these similar-but-not-duplicate questions. And I understand that we can't make people search before posting. I guess I was looking for any ideas on how most efficiently to bring attention to this. It seems that people are much stricter requiring posters to demonstrate an effort in solving the problem, but not so much in searching for old answers. I haven't seen a single comment suggesting to a poster that a search might have been in order. $\endgroup$ – Nick Pavlov Oct 3 '17 at 22:09
  • 2
    $\begingroup$ @NickPavlov I find myself admonishing posters to add context and/or search first with about equal frequency. $\endgroup$ – rschwieb Oct 3 '17 at 22:12
  • $\begingroup$ If it's that obvious the question is a duplicate, the best option for everyone concerned is to flag it as such. Otherwise the asker is all like "those mean bastards are just downvoting me and not answering my question. Don't they understand my homework is due tomorrow morning?" $\endgroup$ – Robert Soupe Oct 8 '17 at 1:43
  • 2
    $\begingroup$ @RobertSoupe You say that as if it's a bad thing. To adapt rschwieb's argument from one of the comments above, if even one of these last-minute-homework askers starts to question his strategy and ponder the reasons why he's not getting answers, I call that a win! $\endgroup$ – Nick Pavlov Oct 8 '17 at 16:58
  • 1
    $\begingroup$ This is a concern talked about in the post, where users arrived at the term "abstract duplicate". So if someone (very basic question) asks how to solve $ax^2 + bx + c = 0$, or asks for the roots of $ax^2+bx + c$ all answers entail using 1) factoring if it is relatively easy to do, or 2) using the quadratic formula. For example, see another example described here. I believe there is, or was, a list of abstract duplicates, with general answers. $\endgroup$ – amWhy Oct 9 '17 at 17:41
  • $\begingroup$ The post I linked to immediately above addressed the same issue, but it's been a couple of years since it's been addressed, and six years since it was asked. So I don't see this meta question as a duplicate. $\endgroup$ – amWhy Oct 9 '17 at 17:45
  • 1
    $\begingroup$ Also search for "abstract duplicate questions" on meta. $\endgroup$ – amWhy Oct 9 '17 at 18:02
  • 1
    $\begingroup$ @Nick. True. We need a FAQ on content on main meta, how just "about" how to operate within and navigate the main site. But all too often a asker of a question cannot see the forest, for the trees. $\endgroup$ – amWhy Oct 9 '17 at 20:13
  • 2
    $\begingroup$ @amWhy That was a point hardmath brought up in a previous comment, I accept that in situations where the duplicate is hard to find. But the blind insistence that "my question is different because it's about sheep and pastures, not balls and boxes" or "because the numbers are different" I don't want to condone. We have to challenge the askers to at least attempt to distinguish superficial from crucial similarities and/or differences - that's the secret to learning. It seems there's not enough emphasis in education these days on that extra step from "how it works" to "when does it work" $\endgroup$ – Nick Pavlov Oct 9 '17 at 20:43

You must log in to answer this question.

Browse other questions tagged .