3
$\begingroup$

There are topics which keep coming back in different forms such as normality testing of random variables. I thought maybe it makes sense to create a question with the best possible formulation, then put a few answers to it referring to good previous answers. This way we'd increase the probability that whoever is looking for an answer will find this Q&A in Google or SE search, and will be satisfied with a complete comprehensive answer. If someone missed the answer, then we'd know that there's this "best answer" to which we'd refer them.

$\endgroup$
  • 6
    $\begingroup$ What's the mechanism for deciding on "best"? In essence, Stack Exchange is not Wikipedia. There's no mechanism to create official or standard or canonical or best answers by design; the mechanism is just that anything aspiring to that status should receive many upvotes if it's really that good. $\endgroup$ – Nick Cox Jan 2 '15 at 0:22
  • 1
    $\begingroup$ possible duplicate of Common Issues and FAQs $\endgroup$ – Andy W Jan 3 '15 at 15:39
  • 1
    $\begingroup$ I've voted this as a duplicate of a prior question, Common Issues and FAQs. Prior discussion suggested creating a specific faq tag to accomplish this. $\endgroup$ – Andy W Jan 3 '15 at 15:41
  • 3
    $\begingroup$ Good approximations to FAQs are possible by someone just writing a good question and then answering it really well. Others could contribute. Everyone, I suppose, approves of the idea, but in practice, this would be as much effort -- for topics such as stepwise selection, evils of, or normality testing, very limited benefit of -- as writing a paper for a journal. I guess that in practice those capable of doing this would rather write a paper for a journal. There aren't enough rewards in what is, for all of us, just one thing that we do on the side to justify the time and effort needed. $\endgroup$ – Nick Cox Jan 4 '15 at 10:25
  • 1
    $\begingroup$ @NickCox, I was thinking that it must be easy to formulate a good question title, then alternative titles inside the body of the question, then in the answer link to all relevant prior answers. I didn't mean to write an entirely new perfect answer, that would take a lot of time. If one has time, then he could write little annotation to prior answers. From there on, this "best answer" would be the place all other answers would tend to collect. $\endgroup$ – Aksakal Jan 4 '15 at 18:05
  • 2
    $\begingroup$ Your notion of an FAQ is proving rather elusive. Compiling a list of references to pertinent threads would indeed be easier than writing a definitive very high quality answer, but in my view it wouldn't deserve the tag FAQ. $\endgroup$ – Nick Cox Jan 5 '15 at 9:26
  • 2
    $\begingroup$ @Nick Cox In fact the appropriate place for creating "lists of important pertinent threads" would presumably be on the tag-wiki entries, which is something the site doesn't take much advantage of and I'm sure the vast majority of readers never visit. $\endgroup$ – Silverfish Jan 5 '15 at 23:33
  • $\begingroup$ @Silverfish: Since a couple of weeks, I have been slowly reviewing all questions in one of the tags and am collecting a list of "important pertinent threads" along the way. With the current pace, it will probably take me a couple of months (I am trying to restrain myself in order not to get carried away spending all my time doing that). Once I am done, I will post the list on the meta and we can discuss what's the best place for it and if such an endeavor makes sense at all. $\endgroup$ – amoeba says Reinstate Monica Jan 6 '15 at 12:05
  • 1
    $\begingroup$ @amoeba For comparison, look at e.g. the SO tag-wikis for C++ which includes a list of top SO questions. $\endgroup$ – Silverfish Jan 6 '15 at 12:09
  • $\begingroup$ Thanks, @Silverfish. I see that the [c++-faq] has 130+ questions on SO, whereas the wiki lists only 9. So I guess these are like the super-faq. I guess the main question here is how broad the criterion for "important pertinent threads" should be. I guess though that it only makes sense to discuss with the specific examples, and that's why I think my list (once finished) is going to be a good starting point for such a discussion. $\endgroup$ – amoeba says Reinstate Monica Jan 6 '15 at 12:32
6
$\begingroup$

..we'd increase the probability that whoever is looking for an answer will find this Q&A in Google or SE search, and will be satisfied with a complete comprehensive answer.

I guess the result of this idea would not work as expected. It indeed could increase the probability of such posts being found, but would that stop duplicate questions being asked?

My hypothesis is that users who are asking recurrent duplicates do not perform at least a minimum search before posting them.

One partial solution would be to edit the question's title (when it is confusing or disruptive) to reflect more accurately what the post is about. This post contains useful hints for writing SE titles (perhaps we could add something from it to our own faq question post).

$\endgroup$
  • 3
    $\begingroup$ "To edit the question's title (when it is confusing or disruptive) to reflect more accurately what the post is about" -- this is something that should always be done anyway! $\endgroup$ – amoeba says Reinstate Monica Jan 5 '15 at 16:23

You must log in to answer this question.

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