I've been contemplating the things that make us want to close questions. On many questions, I vacillate between something being opinion based or story based or too vague, and this is an attempt to help nail down which arguments apply to which kinds of questions.

Some questions want to explore an aesthetic, and those are marked as too opinion based. If it's a question of personal taste, then numerous answers will effectively answer the question, and it turns into a poll with heavy incumbency bias.

Some questions just need a few details to become reasonably answerable. Those get needs detail responses.

There is a further category where solid responses do exist, but they require numerous assumptions. As is the nature of assumptions, they don't get enumerated, so the nature of those answers can't be effectively discussed.

This kind of answer does have a solid mathematically based answer that can be described, so it doesn't actually qualify as opinion based. The problem is that asking OP for more details is a limitless task, and enumerating those assumptions in your answer becomes onerous.

With that in mind, I would like to suggest that, when we run into a question with too many "it depends" conditions, we have a question that is too story based.

What is "too many?" I'd say three. Three creates a situation where their are eight or more distinct correct answers, none of which is any better than the others. It suggests that the OP hasn't thought the question out, and is too many for "but what about X" exchanges in the comments to develop an effective question.

Is this heuristic simple and effective enough to be applied as a general rule?

Addendum: There is a fourth category where someone asks "Is this possible in the real world?" for which the clear answer is "No." We keep closing these for all sorts of reasons when we should simply be answering it with a clear and concise "no," or a qualified, "no, but authors do it anyway," or "no, but it's your story."

  • 1
    $\begingroup$ I'm still thinking about this good question, but I'd like to comment on the addendum. There's a practical difference between what we want and what a new user expects (it's always the new users who are the problem with that issue). Worse, the behavior is a consequence of the Stack's Real World Question policy, which I no longer support (I think it's proven to do more harm than good). Finally, we live in a world where (e.g.) teenagers can be filled with an enormous amount of information (*Continued*) $\endgroup$
    – JBH
    Feb 7 at 21:37
  • 1
    $\begingroup$ ... but have not yet developed much wisdom through experience. In other words, they don't actually know that (e.g.) their fantastic creature really can't exist in the real world. They visit innumerable forums where the possibility is discussed as if it's rational fact. In other words, it's a question type that's difficult for us, because in all ways but one the question type is perfectly good for us... except for the OP's expectation that there should actually be a Real Life manifestation so they can have a story that's as "realistic as possible." $\endgroup$
    – JBH
    Feb 7 at 21:40
  • 1
    $\begingroup$ @JBH, I agree, and I feel that it's our responsibility to tell them how far off they are from a real-world question. IMHO, just closing something as story or opinion based is too punitive a measure to have that beneficial effect. This is especially true for things like Ironman power supplies, which is a rule that you have to learn how to break responsibly. $\endgroup$ Feb 7 at 21:45
  • 1
    $\begingroup$ This stack's prohibition against storybuilding questions (to any degree) is intentional. The Too Story-Based VTC reason is something created by our moderators, not by Stack Exchange. It's based on this question and the resulting policy from seven years ago. I frankly don't know why the policy is defended other than it intrinsically violates several SE rules (open-ended, all answers equal, opinions-vs-facts, etc.). But it's skipped by all but the die-hard rules-are-important people (yup, me and others), so I wonder why we defend it. $\endgroup$
    – JBH
    Feb 7 at 22:56
  • $\begingroup$ @JBH, Personally, I defend it because I don't like doing other people's homework. If I can think of a dozen ways that something can go, it should be the story teller's job to tell me which way it should go. Without this prohibition, this site becomes the "tell me a story where..." site. Tell me a story where someone goes back in time and convinces the US to take over the world. Tell me a story where wolves walk on two legs. There are plenty of places where such story prompts can be presented without turning this into another. $\endgroup$ Feb 8 at 0:33
  • $\begingroup$ OK, then why ask your question? Too Story-Based as a concept only exists to avoid idea-generating questions and as such is nothing more than a specialized version of Opinion-Based. Keep in mind that I'm advocating for creating an SE version of the "plenty of places" you're talking about so that it can be trivially excluded here. Unfortunately, storybuilding is popular. Also keep in mind that "solid mathematically based" doesn't work here. Go read the help center again. We do not require questions to have mathematically sound premises or answers. This isn't Physics. $\endgroup$
    – JBH
    Feb 8 at 0:54
  • $\begingroup$ @JBH, my question is an attempt to solidify what these ratings mean. As things stand, a vote to close will have two of one, two of another, and maybe half of them bother to explain what prompted that vote. It looks like votes to close happen because the voter can't come up with an answer. The idea that story-based is no different than opinion-based is exactly. the kind of thing that I'm trying to clarify. $\endgroup$ Feb 8 at 4:05
  • $\begingroup$ The situation is worse than you think. For example, there's a fair number of questions I could answer, but I believe the question breaks the rules and so vote to close. Due to my rep, there are people who simply trust my judgement and vote along with me (same goes with all of us with 50k+ rep). But it sounds like you're challenging what the policy questions have established. You haven't linked to them, have you reviewed them? I'm open to the idea, but we need to eat the whole elephant. $\endgroup$
    – JBH
    Feb 8 at 4:20
  • $\begingroup$ @JBH, I'm going to have to think more on this, but my perception is that the policy questions make a false assumption based on not understanding how narratives work. It presumes that all stories are about people. In sci-fi/fantasy, the stories are often about worlds, with people being more of a filter through which those stories are told. examples: "Red Mars," "Dragon Riders of Pern," "With a Single Spell," most of "The Wheel of Time" or "Game of Thrones". This makes the policy questions misleading. (continued) $\endgroup$ Feb 9 at 17:28
  • $\begingroup$ (cont.) This is more about building a stronger differentiation between SB and OB. This might be too mathy, but OB is an undifferentiated gradient, real numbers, where any point on a line is a good point. SB is like integers, but permutated, where the answer is a matter of selecting between fixed choices. "Give me a list of events by which my heuristic could be accepted" is story based, whereas "What should the rules be" is opinion based. $\endgroup$ Feb 9 at 17:35
  • $\begingroup$ @JBH, I'm reworking this. Should I overwrite this, or make a new one? $\endgroup$ Feb 10 at 18:54
  • $\begingroup$ If you're reworking this, I'd suggest writing a new question. Link back to this one in the new one. $\endgroup$
    – elemtilas
    Feb 10 at 19:57
  • 1
    $\begingroup$ I agree with @elemtilas, create a new question. I work through the bottom of the week and haven't had the time I'd hoped to respond to this question. I'd rather respond to both so that people can see a more complete picture. $\endgroup$
    – JBH
    Feb 11 at 0:26

3 Answers 3


1. The basic idea is lost in the background of the post

I understand that you're working to make a point, but the original post is somewhat cluttered with information about other VTC reasons and lacked the technical presentation for a policy that I was expecting. Rule #1 about technical writing (which policy-making is), any effort you make to express yourself that muddies or clouds your message is the wrong way to say it. A title focusing on "Too Story-Based" should lead to a post that's focused only on that VTC reason. If I understand your issue correctly, what you're trying to say is:

In some Too Story-Based cases the question depends on too many circumstances that are not obvious because they can change depending on how the story develops. For example, "which weapon is best?" depends on where the fight takes place, who's fighting, what other resources are available, the defensive capabilities of the enemy, what the weather's like during the fight, etc. In this case, a question should not be closed as Too Story-Based if few enough such circumstances or dependencies are not defined that a reasonable number of answers can be expected. I propose that the maximum number of undefined dependencies be three.

2. Because the post is presented here and not as a solution to the existing Too Story-Based question, it'll be lost

I've been down this road before. No matter how well accepted this post becomes, you'll be forever referring back to it to help remind people it exists. The original Too Story-Based post is linked in the Help Center and the Mods are fanatically against modifying the Help Center, so any solution you come up with here must (not needs, must) show up as an answer there.

Having said that, I've a few problems with your presentation

A) Some Story-Based questions might have limited dependencies, but most don't

Your post is not universally applicable to all story-based questions. In its most general form, story-based questions are either about storybuilding or have too much dependency on storybuilding to be answered according to Stack Exchange's rules.

BTW, that last part, "according to Stack Exchange's rules," is the big kicker. This discussion isn't simply about how to keep more questions open. It's about how to clearly express what our policies are in relation to those rules. If people don't want site rules affecting how they ask or answer questions, they should be using Reddit or Quora.

A brief and by no means thorough list of story dependencies that make a question Too Story-Based is:

  • Questions about the plot of a story.
  • Questions about the circumstances of action within a story.
  • Questions about the decisions and/or choices of characters and/or organizations.
  • Questions so broad they are asking us to write the story.
  • Questions with any dependency on any of the previous bullets that has not been clearly defined by the post author in the post.

Finally, I'd like to politely refute the idea that if the author doesn't define, as you suggest, three dependencies, that the result is only eight answers. The problem with story-based questions is that they don't fit within the limits of binary logic. In fact, they're often a challenge to answer even when all the dependencies are resolved because there isn't just one answer, thereby violating the fundamental business model of Stack Exchange: one-specific-question/one-best-answer.

Take my earlier example about selecting the best weapon. Even if all the dependencies are defined, the odds of there being one best weapon are slim. There's a reason why humanity has created untold thousands of weapons over its history. Even in one well-specified situation, there isn't one best weapon. Even if we're talking about just one person in one situation and every possible characteristic of the person and circumstantial detail was provided, it's very unlikely that there's just one best weapon.

In other words, letting three details slide doesn't routinely result in just eight possible answers. The question is story-based. There are still dozens (possibly even hundreds) of possible answers. And Stack Exchange specifically wants to avoid questions that lead to too many answers.

B) Besides, relaxing the proverbial speeding laws leads to dead people

I'm using the metaphor of speeding laws intentionally. Honestly, how many of us drive exactly the speed limit all of the time? I don't, and I'm normally a pretty casual driver who isn't in a rush to get anywhere. But I think it's fair to say that pretty much almost everyone violates the posted speed limit frequently enough to rationally wonder why we bother with speed limits.

We bother with them because if you dispense with the law, people start dying in higher numbers and at higher frequencies. It's a frustrating inconvenience we all suffer with because the alternative is inhumane. And it's no surprise that while the police obviously don't catch all the speeders, and sometimes even ignore the speeders, they're still empowered to ticket every speeder and are not wrong to do so.

Now, equating how we close questions to something obviously life-threatening like enforcing speed limits is obviously a bit of a stretch — but the premise behind my use of the metaphor still exists. We don't own this Stack. We are not authorized to ignore the rules of this Stack. We are empowered to create Stack-unique policies that improve the focus of our Stack. And we are expected as good citizens of Stack Exchange to act together to enforce the rules and policies of this Stack.

For every question you present to me suggesting that it should be left open because the result of the question as-written is an on-topic finite list of things and not an off-topic infinite list of things, I can and will prove that the question is, in fact, an off-topic infinite list of things. That's the nature of questions with unresolved story dependencies.

Because in a story, there are never just two possibilities to the outcome of a story dependency, and with every unresolved dependency, you're suggesting we can handle the load of assuming what could be the possibilities.

I've not found that to be true.

And why is it hard to get the OP to resolve the dependencies? I've spoken with a few of them over the years who were honest enough to admit that what they were looking for was a list of options to choose from to forward their story. In other words, they didn't have a question that met Stack Exchange's expectations, what they had was writer's block and they were intentionally trying to get as many answers as they could contrary to what Stack Exchange wants, which is to provide specifically useful answers to specific questions. Helping authors past writer's block isn't why this Stack exists.



I'd say that "story based" has a pretty good / strong hueristic already.

As I read questions, the issue of "it depends" does not equate to the question being story based. Rather, "it depends" seems to be indicative of an OP either not knowing enough about her own world that she can't clarify a matter or else that the querent hasn't quite formulated or doesn't know how to formulate the question in a way that makes sense.

If she asks about the flight characteristics of laden swallows, and we ask for clarification, like which kind of swallow, African or European and we get a response of "it depends"; then I'd hazard the guess that the OP isn't aware of differing sorts of swallows or hasn't yet formulated the question well.

Too many "it depends" probably should be referred to the Sandbox for help.

When we say a question is story based, we literally mean that the question focuses on issues of plot, character, motive and narrative. In other words, storycraft. For example, if the OP asks "which kind of swallow should my character use to carry something, African or European?" we know right away that this is pure narrativity. Ultimately, outside of the story itself, it doesn't matter which kind of swallow is used. Or even that a swallow is used. The point is that the story needs something to happen, and the character needs to be able to use some mechanism to move the plot forward. Ergo, narrativity.

The hueristic is thus simple: does this query focus on elements of plot, character, motive or narrative. If the answer is yes, then, technically it ought to be closed. If no, then continue.

Note: We've got a discussion currently ongoing about the creation of Storybuilding.SE, which might also be of interest.


I think the best we can do is "rule of thumb", and everyone is going to have different thumbs.

"Story based" is probably the most difficult to rule-of-thumb to establish. We have a limited amount of guidance on this close reason, and really it's just a specialized form of "opinion based" anyway. We know a lot of the close reasons are highly subjective, and I don't think firm rules can be applied for them.

My rule of thumb for rules of thumb is to try to err on the side of being a friendly community. I see far less risk in "answering too many questions" than I see in "driving people off by being nasty about it". Closing a question is a slap in the face. Which some questions do genuinely deserve. But slaps in the face should not be routine if we still want a friendly, thriving community next year.

[edit] Timely example! What replaces Mr/Mrs/Ms as a title courtesy/respect in a world that cares less about gender?

I see someone gave it a VTC for "too story based", plus a second VTC for "opinion based", however, I see this as being very much like a programming question: there are many possible answers. All of them are objectively correct. It is up to the poster to determine which objectively correct answer is most applicable for their situation.

e.g., it is not an opinion that Japan uses gender-neutral honorifics. Good answer! It's also not an opinion that a society can value something more than gender and therefore base honorifics around that. Good answer! Two good answers! My goodness! Some here would say that means it's a bad question, downvote, vote-to-close, and anyone who answered is a bad person who hurts the community.

But I think it just shows that a question can have multiple answers, none of which are "opinions", nor are they "story based", and therefore it is a fine question, and the VTCers are the ones in the wrong in this case.

  • $\begingroup$ As mentioned in my comment above, I wrote this bit of commentary because we're thinking of SB as being indistinguishable from OB, and I don't think that's fair to the questioners. If someone posts a bad question, then the DBaD approach is to help the person understand why they didn't provide a good question. If we, ourselves, don't have a solid reason for differentiating, then we are, by definition, providing vague input. $\endgroup$ Feb 9 at 17:43
  • $\begingroup$ @RobertRapplean Kinda my point though: I don't think you'll see agreement on what is "a solid reason". I edited in an example where in my view, the asker posed a clear worldbuilding question, soliciting objectively correct answers, but some people are still VTCing it as "story building" or "opinion based", presumably because while the answers can be objective, the selection of the "best answer" will be opinion. But my argument is that this is true for most things, even computer programming. Probably only simple math questions would get singular objective answers -- $\endgroup$
    – JamieB
    Feb 9 at 22:18
  • $\begingroup$ -- but multiple objectively correct answers can be useful for others. Even on Stack Overflow, sometimes the answer with the checkmark and 347 upvotes is not what I need for my situation, but the answer down below with 3 upvotes is exactly what I needed. $\endgroup$
    – JamieB
    Feb 9 at 22:19
  • 1
    $\begingroup$ You argument seems to boil down to "people don't agree on things, therefore clarification would provide no value." To me, that means "I don't see a differentiation, therefore none could possibly exist." Probable failure of horses to drink isn't a reason to avoid leading them that direction. OTOH, having the horses kick you because you're trying to help them usually is. $\endgroup$ Feb 9 at 23:17
  • 1
    $\begingroup$ @RobertRapplean Well, I've had high rep members here call me a troll (and worse) simply because I did not agree with their particular clarification on a particular subjective ruling, and I've gotten no traction when trying to go over the why or the history of decisions. So "best of luck to ya" with them horses! They get a bit bitey, is my experience. $\endgroup$
    – JamieB
    Feb 9 at 23:47

You must log in to answer this question.

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