19
$\begingroup$

Let me start from the disclaimer that I understand perfectly well that questions like "what color an invading alien armada spaceships would be" or "what would my superhero decide to do with his life?" are useless and serve purpose to nobody, and fully deserve to be closed under the "opinion based" reason. But then there are many interesting questions that can prove to be useful to anybody that will stumble upon them, even if the answers to these questions aren't founded in facts and logic. They might still solve the problem or offer a solution proposal to the problem that the question outlines.

Yeah Worldbuilding SE is hosted on Stack Exchange, who looks down on "unproven" question that are based on opinion, but Worldbuilding, contrary to the rest of SE, isn't a technical process with easily quantifiable facts and knowledge - it's a creative one. I'd make a hot take and say that for most questions here it is impossible to not be opinion based, and I in fact observed very many highly upvoted answers to questions that were displaying opinion and\or biases of the answering user. Sometimes quite obviously. Yet they were still valuable and useful to consider.

I think, if many of those "opinion-based" question deserved to be closed, it would rather be for a different reason - them being too broad and undefined instead.

$\endgroup$
14
  • 2
    $\begingroup$ I think this discussion is valid, but it's hard to see whether we've been harsh or not without some examples of questions that you think should not have been closed. $\endgroup$ Dec 27 '20 at 2:34
  • 6
    $\begingroup$ It's rather common that a poor fitting question satisfies more than one close criteria. The implementation we have here is that only one is made visible once the number of needed VTC has been reached, while the voters have the possibility to see the partial count before they vote. $\endgroup$
    – L.Dutch Mod
    Dec 27 '20 at 4:28
  • $\begingroup$ In fact, I was just having such an internal debate w.r.t. this question. Alas, "too broad" is not an option for reason-to-close. I VTC'd as opinion-based because IMHO that question is not useful, but on further thought, realized I'm not sure what constitutes "opinion based" or "fishing" and where the line should be drawn. Quite a few well-received questions are to some extent "fishing", and that's almost the nature of this SE. $\endgroup$
    – Matthew
    Dec 29 '20 at 17:34
  • $\begingroup$ @Matthew so we need to make question that is useful or informative to others in this SE? i though this SE mean to help OP regarding their worldbuild problem, in my opinion it maybe obvious or useless to others, but everyone have their own limited knowledge, and it likely helpful to the OP at least, VTC due to its uselessness seems unfair to me. $\endgroup$
    – Li Jun
    Dec 29 '20 at 23:56
  • $\begingroup$ @LiJun, actually, that is my understanding of WB.SE vs. other forums. Specifically, that is my understanding of the "no story-based questions" guideline. That said, "useful" is maybe not the right word for the question I mentioned; the issue with that question was that it was exceptionally open-ended, had almost no detail, and had no apparent criteria for judging answers aside from whatever happened to resonate with the OP. OTOH, maybe I am wrong; that's why votes to close are votes. (FTR, I didn't downvote it.) In a sense, the reason I'm saying this is because I had second thoughts... $\endgroup$
    – Matthew
    Dec 30 '20 at 1:31
  • $\begingroup$ @Matthew yeah i can understand VTC reasons like its too broad or open ended or not enough detail to draw out a solid answer, thats understandable or fair enough, i just want to say my opinion regarding the reason of VTC base on its usefulness seems unfair, because my understanding of this SE is mean to help OP regarding their worldbuild problem, at least i believe so, considering many meta question here basically point that out often as the purpose. $\endgroup$
    – Li Jun
    Dec 30 '20 at 2:00
  • 1
    $\begingroup$ @LiJun, yeah, don't get too hung up on it; "useful" was a poor choice of words on my part. $\endgroup$
    – Matthew
    Dec 30 '20 at 3:31
  • 2
    $\begingroup$ This question is opinion-based and should be closed :p. Joke aside, it's not a full answer, but often times when it's marked as opinion based, I find it real hard to find a start of an idea and if I do, any schmilblick is a good answer, and therefore there's no real quality sort between answers. Sort of. $\endgroup$
    – Tortliena
    Dec 30 '20 at 15:36
  • 1
    $\begingroup$ Do you have an example of questions that would be useful if left open, but get closed as opinion-based? $\endgroup$
    – Chipster
    Jan 2 at 6:58
  • 1
    $\begingroup$ If you're going to complain about a systematic tool, at least be clear what it is, primarily opinion-based. Not just somewhat dependent on individual judgement/preference, but being solely or largely controlled by it. Seems like a number of answers and voters have missed that distinction entirely, and take issue with something that doesn't exist. Stack Exchange sites have long been open to questions that may not have an objectively optimum answer, and how to decide whether it's a good or bad subjective question. $\endgroup$
    – Nij
    Jan 6 at 5:53
  • 1
    $\begingroup$ Yes, I have had three questions in the last three months all closed because they were opinion-based. And it drives me up the wall because every answer on this site is merely an opinion and we have the Opinion-Based Nazi's making sure that any honest question does not stay open more that a couple hours. Its annoying, its degrading and it is completely unfair. $\endgroup$ Jan 6 at 6:48
  • $\begingroup$ FWIW, as a participant in several SEs, I haven't found the enforcement of the POB close reason here to be particularly notable or frequent. $\endgroup$
    – FuzzyChef
    Jan 7 at 7:39
  • $\begingroup$ The irony is, the vote to close reasons are really only as valid as we the users make it out to be. If none of us vote it as POB, it's not going to close. If our community decides to judge it different than the rest of SE its entirely possible for us to unofficially remove POB from our moderation. $\endgroup$
    – IT Alex
    Feb 22 at 18:59
  • $\begingroup$ @FuzzyChef to be fair you do primarily deal in tech/science based questions. I see the majority of my favorite tag [magic] closed for POB even if 95% defined. Hell a lot of really good answers come from only moderately defined questions where the answer solves the underlying problem generically instead of the focused problem. $\endgroup$
    – IT Alex
    Feb 22 at 19:02
18
$\begingroup$

Am I allowed to say it?

HELL YES!!!!!!

I've often said in comments as well as in other meta discussions that if we really took the "opinion based" rationale for closure as seriously as is normative elsewhere on Stack Exchange, we'd literally have to VTC 95% or more of all our questions.

By definition we deal in the irrealis. That is, things that don't exist, places that aren't, events that didn't or never will happen. Questions about dragon hide resiliency, far future starship propulsion, Elves mating with Trees and what kind of offspring they'd have.

Almost every question we answer here is, to a very large extent, opinion based. And I do think we are, as a community, a little too quick to VTC opinion based queries that are actually perfectly answerable within an imaginative context. And I concur re the colour choices of starmada admirals for their uniform jackets, or the like! Those are opinion based questions that should ought to be closed, and with good reason. But questions that really have no basis in known reality are equally opinion based, but it's simply the case that these are the kinds of questions we need to be addressing here if we are to follow our own mandate to help people build, devise, improve or otherwise tinker with their fictional worlds and settings.

As a matter of my own practice, when reviewing queries in the VTC queue, I almost never agree with opinion based as a closure rationale. I might wish to close it for a different reason (most usually such questions might lack clarity or focus), though. And that is a different matter for a different discussion!

In comments, one person wrote that these kinds of questions (opinion based) "do not belong here". I wonder, since this must by definition include the majority of all questions asked in this forum, where else would they belong?

I hold that to say "those questions do not belong here" is an ill conceived opinion given the nature of this forum. Very few of the questions asked here even cán have a "single best answer" the way we find in other SE forums. Even when an OP delineates very clear criteria for evaluation, this can not guarantee that only one answer will fit. The quoted text from the help centre applies to SE generally, not very well to WB particularly. Our forum is radically different in scope & practice from every other forum here. Some people may do better better jobs of writing questions to elicit a specific answer, while others simply have questions which can not be so easily tailored. I think one of the very best ways to kill this forum is to apply the letter of the law too forcefully and too broadly.

Another commentor brought up our VTC policy. While I agree with the overall policy and with the specific subsection in general terms, I don't believe that every single question asked in Worldbuilding needs to list three or five criteria by which to judge the answer. Sometimes the question simply isn't broken when left criterialess: the OP may be looking to cast a wider net; and there is also the risk of making a question too narrow, to the point where it may no longer be useful to the OP and may not even be useful to anyone else.

Lastly, even when an OP lists criteria, restrictions & limitations (per the excellent linked policy statement) we as a community can still easily come up with dozens of equally good, equally opinion based answers! And this is the essential problem that I deal with here, a problem that puts our hobby / art / activity at odds with Stack Exchange's stated goals. And we see an excellent demonstration of this with the examples in Johnny's response.

Quite simply put: WE DO NOT DEAL IN FACTS --- WE DEAL IN THE UNREAL!

Worldbuilding will never be able to follow the basic POB rule of SE to the letter. We can't even really follow it in spirit! The very best we can do is write up a policy & point to it. But in the end, when I look at that policy, it's just so easy to poke a hole in that particular section, because opinion -- creative opinion -- is the heart and soul of what we deal in!

Even the paradigm question in JBH's example can easily have a number of opinion based answers that all conform to the criteria and guidelines given in the query itself.


And then what?
I'd ask you, though: what do you think needs to be done? It's all well and good to ask the question; and I get to use tall caps and lots of exclamation points in response! But what's the next step?

Me I'd like to consider, if it's possible, a rewriting of the "opinion based" closure rationale to severely restrict its scope. Right now, the blurb reads: "This question is likely to be answered with opinions rather than facts and citations. It should be updated so it will lead to fact-based answers." This is the same blurb that one will find on all other SE forums.

The problems with this rationale as stated, and as it relates to Worldbuilding, are obvious.

  1. Most of our answers are, in point of fact, pure opinions! Even when we bring science to bear in our answers, there is no way to offer that answer as the single and one best non-opinion answer.
  2. Most of our answers derive from creativity rather than from citeable sources of fact.
  3. There are no factual resources we can turn to for an unbiased answer for the vast majority of our question types.

That sort of logic is great in Chemisty & Physics where you can either point to a text book or a reliable webpage or a journal article to get the answer. There are certainly scholarly articles and text books we can point to --- but those are someone's opinion! Not universal fact.

Suggested Rewrite:
"This query asks for a non-creative opinion, each of which is as perfectly likely and perfectly reasonable as any other response. Queries should pose a worldbuilding problem or issue that can be answered by proposing creative solutions."

$\endgroup$
18
  • 6
    $\begingroup$ The problem arise when the OP does not define the evaluation metric for picking a "best answer" $\endgroup$
    – L.Dutch Mod
    Dec 27 '20 at 5:50
  • 5
    $\begingroup$ @L.Dutch-ReinstateMonica -- Sure! That is indeed a problem. As I see it, it's just not thìs problem! Even when a querent lays out a really nice metric for picking a "best answer", you can still get half a dozen "best answers" that all account for the metric, but approach from different perspectives. The question itself is still "opinion based". It's just that there are criteria for judging those opinions. $\endgroup$
    – elemtilas
    Dec 27 '20 at 6:13
  • 5
    $\begingroup$ @L.Dutch-ReinstateMonica Some questions don't have one best answer. That doesn't matter. $\endgroup$
    – Johnny
    Jan 1 at 8:37
  • 2
    $\begingroup$ @Johnny, those questions do not belong here. $\endgroup$
    – L.Dutch Mod
    Jan 1 at 9:49
  • 3
    $\begingroup$ @L.Dutch-ReinstateMonica That is not an answer, that is an arbitrary opinion on whether they belong here. If so, you should delete the three most popular questions this month, as there is NOT one best answer for any of them. $\endgroup$
    – Johnny
    Jan 1 at 9:56
  • 2
    $\begingroup$ @Johnny, take a read at the help center, where it states clearly that questions where every answer is equally valid are not a good fit for this community. $\endgroup$
    – L.Dutch Mod
    Jan 1 at 10:05
  • 3
    $\begingroup$ @L.Dutch-ReinstateMonica Take a look at the community. Particularly the top three questions this month. Apparently, their views are not shaped by the help center. $\endgroup$
    – Johnny
    Jan 1 at 10:07
  • 3
    $\begingroup$ @L.Dutch-ReinstateMonica -- That I have to strongly disagree with, that "those questions do not belong here". Very few of the questions asked here even cán have a "single best answer" the way we find in other SE forums. Even when an OP delineates very clear criteria for evaluation, this can not guarantee that only one answer will fit. Keep in mind that the quoted text from the help centre applies to SE generally, not to WB particularly. Our forum is radically different in scope & practice from every other forum here. This legal literalism is a bit rich from someone who has asked... $\endgroup$
    – elemtilas
    Jan 1 at 21:23
  • 3
    $\begingroup$ (cont) ... at least a couple queries that really don't have one single best answer! Some people may do better better jobs of writing questions to elicit specific answer, while others simply have questions when can not be so easily tailored. I think one of the very best ways to kill this forum is to apply the letter of the law too forcefully and too broadly. $\endgroup$
    – elemtilas
    Jan 1 at 21:24
  • 2
    $\begingroup$ @elemtilas Well said, Elem. You should consider adding that to the post. If you haven't already, you should take a look at my post, as well. I outlined several questions that really seem to mess up any idea about no opinion based questions, and JBH was kind enough to tell me whatfor in the comments of my post. I feel worrying about the main stack's rules which don't apply to WBing is a waste of time. Which should focus on rules and standards which actually have utility for WBing, which suit the community. Far from legalist, I consider the current standards horribly vague and open to opinion. $\endgroup$
    – Johnny
    Jan 3 at 22:47
  • 2
    $\begingroup$ @Johnny -- I actually did read your response (and duly upvoted). Those are some great examples of questions that, even when the OP lists criteria, restrictions & limitations (per JBH's excellent linked policy statement) you can still easily come up with dozens of equally good, equally opinion based answers! And this is the essential problem that I deal with here, and that you demonstrate with examples in your own response: WE DO NOT DEAL IN FACTS --- WE DEAL IN THE UNREAL! WB.SE will never be able to follow that basic rule of SE to the letter. We can't even follow it... $\endgroup$
    – elemtilas
    Jan 3 at 23:00
  • 2
    $\begingroup$ (cont) ... really in spirit! The very best we can do is write & point to JBH's policy. When I look at that, it's just so easy to poke a hole in that particular section, because opinion -- creative opinion -- is the heart and soul of what we deal in! $\endgroup$
    – elemtilas
    Jan 3 at 23:02
  • $\begingroup$ @Johnny -- Just so you realise: what you just told me is publicly viewable as it's here in comments! That said, two things: if you believe there is a consistent pattern, I would first flag the questions (yours) and click on "in need of moderator intervention" and explain what you think is going on; then, I'd ask a question here in Meta where you can link to your questions that you believe are being wrongly scrutinised and also link to other queries that you believe are not. You won't be banned for this, though I would make certain that, in any interactions you have going forward, you are... $\endgroup$
    – elemtilas
    Jan 4 at 21:55
  • $\begingroup$ (cont) ... clear in your mind that he is or is not actually insulting you personally vs simply saying something -- perhaps snarky, perhaps rude, perhaps simply frank & forthright -- about your questions. I've known JBH for quite a while here, and while not above frustration, I've never known him to be actually insulting of a person. I just want to get straight what you actually mean when you say "insults and extremely condescending manner". I want to be fair with both of you, which is why I think you should open a Meta question rather than just us chatting in comments! $\endgroup$
    – elemtilas
    Jan 4 at 21:59
  • 1
    $\begingroup$ (cont) ... Sometimes, as in prompting this Meta query, the rules get applied too rigidly I think. Personally, I find it utterly absurd that anyone would seek to close questions in this forum because there's no factual basis for the question or the answers. But the fact remains that "scrutiny" and "threat of closure" are two of the basic functionalities of all SE forums. If the community thinks a question is bad, for any reason, it can be closed. If a question is at all interesting, just wait! There's always a good chance it'll end up in the reopen queue! And lastly, yes JBH is a prominent... $\endgroup$
    – elemtilas
    Jan 6 at 17:00
12
$\begingroup$

We are required to face some realities

  1. Stack Exchange is not a discussion forum.

  2. While we tend to work around them, we are bound to Stack Exchange's basic question-writing rules as found in the help center. The most notable problems are (a) every answer cannot be equally valid (it must be possible to choose a best answer) and (b) questions can't be open-ended (most opinion-based questions are).

  3. The "Primarily Opinion-Based" VTC reason and its text are controlled by Stack Exchange, not us. It can't be suspended or modified. In other words, whatever solution we come up with must deal with the fact that we're stuck with it.

Over the years I've proposed a number of ways to help people work around these problems.

A. I've posted the following comment, when appropriate:

You're asking what we call a high concept question, which isn't a good fit for our site. Please review these links and help us by improving your question.

B. I've posted the following comment, when appropriate:

We need to convert this from an off-topic infinite list of things to an on-topic finite list of things.

C. I provided a policy solution: A question is POB if there are no listed criteria for judging a best answer

D. I created an alternative Welcome page: Welcome to Worldbuilding.SE!

I have used or pointed people to all of the above in my (usually vain) effort to help people reformulate their questions to meet basic Stack Exchange guidelines. As much as we might all wish it were otherwise, this isn't a free-for-all. Those aspects of the Help Center that we cannot modify are rules we are expected to live by.

Why not just ignore them? Because the result is chaos. People will point at the rules and say "you can't do that!" and others will scream "but we all agreed to ignore that!" and then we're all hunting for Meta posts for policies that few, if anyone, read.

If you want to avoid closure as Primarily Opinion-Based on this site:

  1. Focus your question on the smallest problem possible. Literally take the time to wash all the extra fluff out. If you don't take the time to do this, don't expect us to do it for you.

  2. ADD DETAILS I can't tell you how many times I've had an OP in comments say, "I intentionally didn't add details because I wanted people to come up with a variety of ideas." With all due respect, that's 100% contrary to how Stack Exchange was designed. Page after page in the Help Center explain the need to be specific and detailed.

  3. Explain how your best answer will be judged. An OP's refusal to do this is 99% of the problem, and it's almost always because what they're doing is fishing for ideas.

The harsh reality of Stack Exchange is that it was intentionally designed as a place where people could go to get specific and useful answers. Consequently, it's a terrible place to go when you have writer's block because it's intentionally NOT designed for raw idea generation. Over the years I/We have tried to come up with explanations to help people work around this — but the limits imposed by Stack Exchange that keep these efforts from becoming easily found makes them basically worthless.

Which was a long-winded way of saying that I disagree with my learned colleague. But, gratefully, there aren't a lot of us who are fanatically policing the rules. So, with good fortune and providence, we're slapping hands enough to get people to write better questions but not so much that we're driving too many people away.

$\endgroup$
11
  • 1
    $\begingroup$ OP suggestion to use undefined instead, seems more make sense to me. most of the time i don even know how or where should i start the answer be, due to the the question, despite its quite focus or detailed. $\endgroup$
    – Li Jun
    Dec 29 '20 at 0:30
  • $\begingroup$ @LiJun the primary problem is that we can't change the POB VTC reason or its text. We're stuck with it. Adding another, competing reason would only make things more difficult. I'd better add that to the answer. Thanks. $\endgroup$ Dec 29 '20 at 0:36
  • $\begingroup$ the best answerer there did mention that "If you want communities to be able to change the text for POB you need network wide changes. By making a feature-request you could theoretically get the attention of the StackExchange staff." so seems like its still have a way at least, so not all hope is lost, it require to make a better structure to convince the staff regarding the SE problem and solutions, and luck. $\endgroup$
    – Li Jun
    Dec 29 '20 at 0:48
  • 1
    $\begingroup$ Understandable about the rationale being SE controlled and unalterable by us (officially at least). That doesn't mean we have to follow it to the letter. All it means is we have to have a sensible community interpretation of those words. $\endgroup$
    – elemtilas
    Dec 29 '20 at 1:59
  • 2
    $\begingroup$ You might disagree with me, but I think we will find ourselves on common ground on many of your points. They are all, essentially, good question writing tips that are broadly applicable. And let's face it, there is something of a learning curve when it comes to writing good questions. I certainly don't intend to say that we should leave poorly written opinion questions unedited or uncomplained about. But neither should we behave so aggressively towards one type of question, even when sufficiently well written. $\endgroup$
    – elemtilas
    Dec 29 '20 at 2:07
  • 1
    $\begingroup$ @elemtilas Actually, the only place I think we're not on common ground is the H1 print. I felt that the policy proposal in "C" is an adequate interpretation (and I've seen it used by L.Dutch) in that we put the burden on the OP. an open-ended, list-type, or opinion-based Q is acceptable if (a) it's not too broad and (b) the OP provides the criteria for a best-answer (necessary limits, conditions, and expectations). Since SE won't change the text or make it Stack-dependent, it's the best we can do. $\endgroup$ Dec 29 '20 at 17:23
  • 1
    $\begingroup$ There are many cases where there is no single best answer in real life, or no clear best answers in general... so why do we assume only questions with a single best answer should be asked? $\endgroup$
    – Johnny
    Jan 1 at 8:39
  • 2
    $\begingroup$ @Johnny Whether or not there actually is only one best answer is irrelevant. (a) Stack Exchange dictates the model, one-specific-question/one-best-answer. Declaring the owners of this engine to be irrelevant (and hoping everyone will ignore the help center) is insensible. (b) Posting a question with the expectation that there could be and should be a single best answer will always produce a better question with better results. As has been mentioned by both me and others, if what someone wants is an environment like Reddit or Quora, they should use Reddit or Quora. $\endgroup$ Jan 2 at 5:42
  • 1
    $\begingroup$ @JBH Look at the top questions this month. Tell me, what's the best answer to how aliens can be killed with nerf darts...? Pretending to follow rules isn't sensible, it's hypocritical and leads to incoherently arbitrary standards. Quora and reddit have very different environments from each other.... And yet, like Stack Exchange, you could say it's roughly the same. Rules don't change people that much, I guess. $\endgroup$
    – Johnny
    Jan 2 at 6:18
  • 2
    $\begingroup$ @Johnny did you miss my point? Go to my answer and read the link for (C). You're stuck on the idea of the precise meaning of what SE's text says. 90% of the questions we get fail that precise interpretation, that's why we shifted the policy to, "if you, the OP, can tell us how you'll judge a best answer, the question is no longer Opinion-Based." $\endgroup$ Jan 3 at 21:28
  • 1
    $\begingroup$ @JBH You mention in another comment that quality standards drop when these rules, "aren't enforced." I'm not sure how this is, when the rules are so vague. You didn't even give a full example in the post you linked to, when you could've used a real example, or invented one. You also used a 'bad' example not unlike some of the popular questions of late. Since the standards for opinion and description aren't good enough to write an example, I have difficulty imagining what the standard for "worse" is. Do you have any example of how things get "worse" when you "stop worrying about it"? $\endgroup$
    – Johnny
    Jan 3 at 22:00
6
$\begingroup$

The irony is, it's ALL opinion. With such loose guidelines as to what is "opinion based," posts are arbitrarily defined as opinion or not, by moderator opinion.

Worse, this largely falls to a few, as you have to go out of your way to get involved with the review process or the meta discussions, and it's outside the reward scheme of SE, giving little incentive.

And while we throw about opinions about what's opinion, most answers and questions are STILL opinion based. Let's take a look at some of the current high ranking questions and answers of the month:

"Semi-plausible reason why only NERF weaponry will kill invading aliens"

...That one is that the top of the list for the month. The 76 point answer was suggesting the aliens play dead, because it's a game show for them. The 39 point answer suggested allergies to nerf darts, and a 7 point answer which the querent picked had it that the alien shields so happen to malfunction when hit by nerf.

So, after reviewing that, the most popular question this month with many of the most popular answers... how exactly does one avoid opinion-based answers, when deciding how nerf darts can kill aliens?

It appears to be a fear that, "the problem arise when the OP does not define the evaluation metric for picking a "best answer"", that, "every answer cannot be equally valid (it must be possible to choose a best answer)". ...So what was the best, non-opinionated answer here!? The one that got a green tick with seven votes? The one that suggested the gameshow? Or one of the many other radically different answers (the 3rd-most popular suggested NERF was an acronym for a weapon)?

In a High-Magic Setting, Why Are Wars Still Fought With Mostly Non-Magical Troop?

2nd most popular question this month. It seems open ended questions are the most popular. As if the aliens wasn't bad enough, now we're dealing with not even pseudo-science.

And it's not that this is a bad question... people enjoy it because there are MANY good answers that can be given, that it requires creativity to answer. Thus, we get another selection of answers that don't really respond to any clear metric.

Why do people still live on Earth Like Planets?

It just keeps going.... And this brings us to a good topic on scientific opinion. Because, though people are bringing up science, even though these may be the same arguments brought up by experts... it's still just OPINION.

They don't KNOW why people would stay on Earth like planets when given other choices, because the futuristic scenario being discussed has NOT happened yet, so we can't say for sure. AKA: We can only give an opinion on how it might be.

Conclusion

So, while some opinions are better than others, enough that people seek out the best opinions on sites like this... they are still opinion. The way we sort answers and questions is, in fact, a polling of community opinion.

So clearly, saying that a question is bad because it is opinion based is ridiculous and hypocritical, we'd need to delete 90% of the site if that was a genuine problem.

Why are "opinion-based" questions a problem?

Even the example by the OP, what colour should a spaceship be, someone might point out it could be painted white to reflect light so as to reflect sunlight, and another might suggest painting it like decorative WW1 biplanes. I'm not sure if that question is less relevant than the one about nerf darts.

I guess the real question, is whether the question has been answered before--you only really need one question about the colour of a non-specific spaceship.

The one about what a superhero does in daily life... is that less open ended than the one about nerf darts? Or the other two? Non specific questions are likely to get non-specific answers, which are less helpful, but what level of specificity is required, and how do we measure it?

In the end, it seems a lot of worry and hassle over something which can't be legitimately decided... which would likely be better handled by community voting (downvoting questions is specifically for filtering low quality ones).

$\endgroup$
1
  • $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$
    – L.Dutch Mod
    Jan 5 at 9:32

You must log in to answer this question.

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