8
$\begingroup$

This close reason is now live. See: Why are questions off-topic if they invite answers which are not demonstrably correct, or are otherwise speculative?


I'd like to propose a new site-specific close reason, based on some past Meta discussions linked below. I propose the following wording, though this is, of course, open to suggestions and modifications:

This question is off-topic because it is a puzzle question with subjectively-correct answers. Good questions on Puzzling.SE have a limited number of objectively correct answers.

This, I think, covers a category of questions the community already feels we should not allow on the site for a variety of reasons. They're somewhat opinion-based, since one solution is only "more valid" through subjective analysis, and they're too broad, since they can have multiple (or infinite) reasonable answers.

The questions for which this close reason would apply have already received a number of downvotes, indicating the community feels there is something inappropriate about them:

(There are other examples that I've seen over the past couple days, but as far as I can tell they've since been deleted and I do not have the reputation to find them. On this point, feel free to edit in any questions you feel match the criteria. This is an appeal to the Community Managers, so if we want this, we want to make the best possible case.)

We've discussed similar topics:

(Once again, feel free to edit in Meta questions.)

I think this would help us trim down what we do not want, and allow what we do want to have a little more breathing room.

$\endgroup$
  • $\begingroup$ Could you also show examples of the questions with out subjectively-correct answers and questions with exactly one subjectively-correct answer? $\endgroup$ – klm123 Jun 4 '14 at 22:33
  • $\begingroup$ @klm123 The point is to not have any subjective answers at all (the second part of the close reason says one objective answer). Maybe we're creating confusing by discussing limited/multiple and objective/subjective in the same sentence, Emrakul? You know what you mean, and I know what you mean, but maybe the people whose questions we'd close don't. $\endgroup$ – WendiKidd Jun 4 '14 at 22:44
  • $\begingroup$ @WendiKidd Hmm... this is more of a philosophical question, but the way I see it, the existence of a single subjectively correct answer either a) suggests others exist, or b) suggests that the problem is either solvable or not solvable based on opinion. In the case (a), the close reason applies; in the case (b), I think either this or "primarily opinion-based" applies, but I'm not entirely certain. Maybe the word 'multiple' should be removed? $\endgroup$ – Aza Jun 4 '14 at 22:46
  • $\begingroup$ @klm123 Sure, I can do that. All of our best questions have no subjective answers. $\endgroup$ – Aza Jun 4 '14 at 22:47
  • $\begingroup$ @Emrakul, I see your point that the questions with multiply answers are note good for such a site, where only one answer must be chosen. But if you show this chess-question to a person, who never saw chess puzzles and doesn't know definition of "move", he(she) will answer "1 piece". And this would be subjectively correct answer, by definition. The same way anyone here have never saw questions different, but with same rules as the mentioned 4 questions, therefore they can find many different answers to them if they will allow to alter the rules according to their wishes. Do you see my point? $\endgroup$ – klm123 Jun 4 '14 at 22:57
  • $\begingroup$ @klm I definitely understand. The distinction comes from whether one actually has to bend the rules to come to an incorrect/different conclusion. If somebody who is ignorant to the definition of "move" comes to a certain conclusion, that conclusion is still objectively wrong, despite their misconceptions. However, when a question has multiple subjective answers, those answers have no objective status as "right" or "wrong" - this usually occurs when the questions rely on semantic answers as opposed to logical ones. $\endgroup$ – Aza Jun 4 '14 at 23:00
  • $\begingroup$ @Emrakul, then I would repeat the question, which I asked in "Number of houses in a village" - what about puzzles, which are not well known, and therefore doesn't have "objective" terminology to describe them? All new puzzles in new areas are like this. They should not be here at all? $\endgroup$ – klm123 Jun 4 '14 at 23:03
  • 1
    $\begingroup$ @klm One of the first steps in defining a new logic puzzle is creating terms to accurately describe the conditions of those puzzles. For instance, the term "jumbling" is relatively new to the mechanical puzzle world, but it's necessary to discuss puzzles like the Helicopter Cube. Additionally, if the question wants to use a new concept, it must define exactly what it is in the question. $\endgroup$ – Aza Jun 4 '14 at 23:05
  • $\begingroup$ @Emrakul, that contradicts to the description of the tag "brain teaser", which you have here: A brain teaser is a puzzle that involves thinking in unconventional ways, or one that involves disregarding what seems obvious. It typically requires induction rather than deduction." If you define all used concepts there will not be brainteaser at all. Then may be you should remove this tag as well? There are only 9 questions with it. $\endgroup$ – klm123 Jun 4 '14 at 23:10
  • $\begingroup$ @klm That is something I have been considering proposing, yes. But as I've said multiple times, we should really continue in chat. $\endgroup$ – Aza Jun 4 '14 at 23:11
3
$\begingroup$

The wording of your proposal seems like it is very easy to misunderstand. As I stressed in this answer I just posted:

The problem is answer validity, not quantity.

Furthermore, the language used in the proposal ("multiple subjectively-correct") can be very confusing even to native speakers. Therefore, I propose the following wording instead:

This question is off-topic because it can have many answers, some or all of which cannot be shown to be valid or correct. Good questions for this site have a limited number of objectively correct answers. See also: Some meta post.

(I also like @WendiKidd's idea of including a link, which I have added in as well.)

EDIT: After a wonderful discussion with @Emrakul in the chat room, we've tried to figure out what the real problem is with these types of questions. After much hair-pulling and exhaustive discussion, we've finally pulled this new proposal together:

This question may invite speculative answers, as the question is not fully defined. The validity of some answers may be based upon opinion. Good questions for this site have a limited number of objectively correct answers. See also: Some meta post.

$\endgroup$
  • $\begingroup$ Hmm... I think you have a good point, but the wording here seems tricky. Lots of answers can be argued (which is how I think a lot of people will read 'shown') to be correct. Maybe "an unlimited number of equally-valid answers"? $\endgroup$ – Aza Jun 7 '14 at 17:40
  • $\begingroup$ @Emrakul Hmm, do you think a better word would be "determined"? Or "proven"? Or "confirmed"? $\endgroup$ – Doorknob Jun 7 '14 at 17:46
  • $\begingroup$ Hmm... there's a slight semantic difference there that makes me feel we're not entirely in-sync on why these questions shouldn't be in the site scope. To me, this looks like the questions should fall under a more specific version of "too broad" or "primarily opinion-based"; maybe to chat? $\endgroup$ – Aza Jun 7 '14 at 17:52
3
$\begingroup$

This is an excellent idea! I want to raise the point that "multiple subjectively-correct answers" might be hard to parse for non-native speakers, but I don't have a better idea of how to say it; it's the most correct and descriptive explanation of the problem. Maybe others can weigh in on this.

I do think that we should, as a community, decide which meta post best explains this problem/our decision to make these questions off-topic (or even create a new meta post, FAQ-style, and explain our reasoning) and add a link to that meta post in the close reason text. I've had success with that on another SE site; when the close reason might cause the OP to have questions, linking to an explanatory meta post can help head those questions off (or at least you can reply to comments with "see that link in the close description?" so you don't have to keep repeating yourself in comments.)

So it'd end up being something like this (minor wording changes made, but that's not the main point):

This question is off-topic because it is a puzzle with multiple subjectively-correct answers. Acceptable questions have a limited number of objectively correct answers. See: Questions with multiple subjective answers.

(Except instead of linking to google like it does now, that would link to whatever meta post we decide to create/choose.)

$\endgroup$
  • $\begingroup$ I've moved a little bit of this wording into the wording above. I like the addition of the page link, and I'm wondering if this page could serve that purpose. It seems almost redundant to create a new question for it, but if we need to, it's not a huge issue. $\endgroup$ – Aza Jun 4 '14 at 22:55

You must log in to answer this question.

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