19
$\begingroup$

It often happens that (usually somewhat inexperienced) users post several questions on more or less related themes as one single question. To give a stylized example, someone may ask:

Which of the following numbers are even?

  1. $0$
  2. $42847$
  3. $12$
  4. $3$

Sometimes, the connection is even weaker:

Which of the following is true:

  1. $2$ is prime
  2. $4\bmod 2 =3$
  3. $n!>n$.

Such questions tend to have the flavor of being copied from a homework sheet and are among the ones that tend to annoy many users at M.SE. Usually, the person asking the question could formulate an abstract question that adresses the common problem the person has with these problems ("How can I see whether a number is even?") with a little bit of effort. If the person can't, it should really be several questions.

So I propose that we generally discourage or prohibit such lists of questions. Having a uniform standard on such lists would help us in avoiding long discussions and comment threads.

$\endgroup$
  • 4
    $\begingroup$ It's very hard to prohibit something --- perhaps it's impossible. It's even hard to discourage something, when there are so many people with such widely differing views. Perhaps you have a mechanism to propose? $\endgroup$ – Gerry Myerson Oct 29 '12 at 23:39
  • 3
    $\begingroup$ @Gerry: We can always wait until they invent small teleporting devices which will allow us to sneak up behind these people and beat them with heavy calculus books (Thomas & Finney comes to mind)... :-) $\endgroup$ – Asaf Karagila Oct 30 '12 at 0:23
  • 8
    $\begingroup$ @Gerry We could close them "as not a real question" and spend less time on meta discussing whether these are perfectly decent questions that shouldn't be downvoted. $\endgroup$ – Michael Greinecker Oct 30 '12 at 0:32
  • 7
    $\begingroup$ If closed as NARQ, there should be a standard comment explaining something like "This site works best if you ask a single question in each question. Please either edit to ask just one of these questions, or ask about a general technique which could help you solve all these questions. Your question may be closed as "not a real question" because it's several questions instead of one." $\endgroup$ – Noah Snyder Oct 30 '12 at 9:00
  • 6
    $\begingroup$ I think this is especially relevant when some of the subquestions are duplicates. Having them all in the same question makes pointing to existing answers awkward, and might lead to a matryoshka-doll like structure of answered-questions-within-questions. Trying to limit the number of questions in questions makes a lot of sense to me. $\endgroup$ – rschwieb Oct 30 '12 at 16:40
19
$\begingroup$

I would propose the following policy:

  1. Treat questions of the first kind as if they were asked in a more general form and answer them in the spirit of "teaching how to fish". This would entail that similar questions about different numbers count as duplicates.
  2. Close questions of the second kind as "too localized", since the main objection to them, as far as I can see, is that the same combination of questions is unlikely to arise again.
$\endgroup$
  • 3
    $\begingroup$ I reiterate my suggestion that if people agree with this suggestion (which I think is very reasonable) there should also be a standard comment template for each situation which we can copy-paste. $\endgroup$ – Noah Snyder Nov 3 '12 at 12:00
  • $\begingroup$ @NoahSnyder I would suggest, you just suggest this standard comment in the list of templates meta.math.stackexchange.com/questions/4925/… $\endgroup$ – Julian Kuelshammer Nov 3 '12 at 15:06
  • $\begingroup$ I'm not sure that closing as too localised is appropriate. It's quite possible that other people may find one of the question-answer pairs useful. I'd say just comment (perhaps with a link to this question) and downvote. $\endgroup$ – Ben Millwood Dec 17 '12 at 13:32
  • $\begingroup$ @I've made some initial version here, feel free to modify/improve the comment template. $\endgroup$ – Martin Sleziak Jan 9 '13 at 15:32
12
$\begingroup$

There are several problems with multiple questions:

1) It makes it harder to identify the multiple questions as a duplicate or link to it as a duplicate or explanation.

2) It makes it harder to search for a particular part of the question.

3) It makes it almost impossible and/or very frustrating to actually answer the question, especially in the case of "After reading your answer, I want to add the following additional question / condition to my original question list."

4) Basically, the poster diminishes the validation of answers to their question and the value to other people with the same questions (not on purpose, of course.)

I would certainly not close it as too localized, but instead:

1) If the list actually consists of abstract duplicates, then answer the abstract question.

2) If the list consists of several separate questions, answer one question and ask the OP to repost the rest as new questions. Noncompliance after a certain time can be solved by just editing out the other questions. (With a comment of course, pointing out that the OP can still copy and paste the previous version into new questions.)

$\endgroup$
  • $\begingroup$ I think the list of problems you have here would be better placed in the original question, and your answer should just be your proposed solution. $\endgroup$ – Ben Millwood Dec 17 '12 at 13:32
  • $\begingroup$ @BenMillwood That would be me second-guessing the OP. I have proposed a course of action that adresses the problems that I see with multiple questions. Whether or not it addresses the OP's and other people's problems with multiple questions should not be judged by me. $\endgroup$ – Phira Dec 17 '12 at 13:38
  • $\begingroup$ Are you saying that editing should work differently on meta, or would you apply the same policy in general? (The reason I proposed the idea in the first place is because I want to link people to this question, and it's useful to have the problem statement and elaboration together in one place). Would you be okay with the OP editing that content into the original question, and then editing it out of this answer? $\endgroup$ – Ben Millwood Dec 17 '12 at 13:48

You must log in to answer this question.

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