Isaac Moses, the patriarch of Mi Yodeya, wrote something that caught my eye the other day:

[...] the general Stack Exchange directive, expressed in the uneditable part of that section of the About page, to "Focus on questions about an actual problem you have faced. Include details about what you have tried and exactly what you are trying to do." In fact, while this directive is perfect for the behemoth programming site that Stack Exchange grew out of, it doesn't fit very well for any of the more intellectually-oriented, rather than practically-oriented sites. [...]

This is something that's bugged me for a while, and ended up prompting a discussion between Anna, Aarthi and me. The gist of it: is there any simple test that can be used to identify answerable questions asked in good faith when the "problem to be solved" is... simple curiosity?

Obviously, this is something that sites like Movies and SciFi have had to contend with for some time, but now and then such questions pop up even on sites where folks are generally trying to solve real problems. While a potentially interesting thought experiment, you're probably not going to use a distributed hashing algorithm to pair socks next time you do the laundry.

Strictly-speaking, problems like that fail the "practical, answerable questions based on actual problems that you face" test. And yet, I would keep that question simply because I learned something useful from the answers − not necessarily something I would ever apply to socks mind you, but certainly something I might apply to similar problems. I don't believe the question was asked in bad faith or for mindless social fun; yet, I have a bit of trouble stretching the definition of "based on actual problems" to contain it.

Obviously, the fact that we're hosting a growing number of sites dedicated mostly or entirely to intellectual curiosity makes this something of a rhetorical question. The "actual problems" part of the FAQ was added specifically to draw a line between good and bad subjective questions − heading off fantasy battles and their ilk.

But it's a pretty damn good litmus test, and if it's ok to ignore it for some questions − or particularly for every question on a site − then what should stand in its place?

  • 1
    Somewhat related: meta.stackexchange.com/questions/101507/… Commented Jan 21, 2013 at 4:08
  • 1
    I find myself on the "wrong" (?) side of this sometimes, too. stackoverflow.com/questions/5301654/… Commented Jan 21, 2013 at 4:10
  • 5
    Here's what I've said in the past on the issue: My goal is to read the answers to this question, and in doing so expand my knowledge and understanding of all the tools of our profession. Commented Jan 21, 2013 at 4:15
  • Wouldn't it be hard to reach consensus about which questions were deemed "too impractical/ethereal/philosophical" - perhaps, and this is what I've seen, we really tend to feel empathy to the user who earnestly is interested in XYZ and presents a nicely formatted/presented question. To those people, even when I know it's not exactly SO-material, I've always respectfully carried on. If somebody wants to help, well ..OK Commented Jan 21, 2013 at 4:38
  • 4
    @Adel Just a note, this question isn't SO specific. Too philosophical, for example, isn't really a problem for Philosophy.SE.
    – yannis
    Commented Jan 21, 2013 at 4:54
  • 12
    I fear that mentioning the sock question has derailed your question here, because people seem to be putting their SO glasses on, and forgetting the rest of SE. English, Mi Yodeya and Christianity (amongst others) are all fueled by curiosity.
    – Benjol
    Commented Jan 21, 2013 at 6:49
  • 3
    IMO, this ought to be reopened -- the other question (while not explicitly mentioning it), pretty much talks about SO only. This one explicitly mentions other SE sites and is tagged as a network-wide discussion post as well :/ Commented Jan 21, 2013 at 7:36
  • 3
    Sock sorting is a practical, real problem we face every day at our house. It's natural as programmers to want to optimise this tedious task. I currently use a mix of answers given - hash/radix (as a human I have a good context-switch heuristic). You might criticise the question for being off-topic, but you can't criticise it for not being a real problem.
    – AndrewC
    Commented Jan 21, 2013 at 7:57
  • 3
    Sorting socks as a programmer of course
    – random
    Commented Jan 21, 2013 at 15:42
  • 7
    I've been wondering what to call my role at Mi Yodeya ever since my mod-pro-tem status expired. Thanks very much for the awesome suggestion. I have called my stationer to have new letterheads printed. Also, thanks for bringing this to Meta.SO so I didn't have to. :) Commented Jan 21, 2013 at 16:21
  • 3
    This issue was previously discussed on Meta.SciFi: meta.scifi.stackexchange.com/q/1380/3693 meta.scifi.stackexchange.com/q/389/3693 Commented Jan 21, 2013 at 17:22
  • 1
    @IsaacMoses Patriarch? Wouldn't Gaon have been more apropos? It's sort of like reinventing Sura and Pumbedita. And the internet is nothing if not like Babylon.
    – Rosinante
    Commented Jan 21, 2013 at 17:28
  • 4
    @Rosinante, I like that it's an accurate statement of paternity rather than an inaccurate statement of expertise. :) Commented Jan 21, 2013 at 17:47
  • 1
    Here is a recent example of the current FAQ text being harmfully misleading. Commented Dec 17, 2013 at 23:01
  • 1
    Shog, any updates?
    – Double AA
    Commented Dec 4, 2014 at 19:08

13 Answers 13


See also: What "practical" physics problems do you "actually" face??

I agree with Isaac here. On science sites, this rule can actually be counterproductive.

Questions asked out of curiosity generally fall in the following types: "I think I see a paradox here", "What would happen if?", and other conceptual questions. All good, because these are generally conceptual, and get great answers. Thought experiments can sometimes be our best questions.

On the other hand, half the questions asked because of a problem that is faced are homeworky, and not so good. The rest are almost always too localized.

So, while this may be a good idea for SO, why keep it a network-wide policy?

And why do we need something to stand in its place? The close reasons for Too Localized generally covers the Gorilla vs Shark cases on most sites (SO included, as far as I can tell). The possible issues with contrived, not-a-problem-you'll face questions that I see are:

  • They may be unhelpful to others. Already covered by "too localized"
  • They may spark discussion. Already covered by NC
  • They may not really be answerable; such questions may fall into the "Need a whole book to answer" type without the OP realizing it (NARQ)

See? Already covered.

As far as the FAQ goes, why not replace the line with the following?

You should only ask practical, answerable questions that will be useful to others. Chatty, open-ended questions diminish the usefulness of our site and push other questions off the front page.

Or even:

You should only ask answerable questions that will be useful to others. Chatty, open-ended questions diminish the usefulness of our site and push other questions off the front page.

The word "practical" really is meaningless here, and again, may be detrimental to theoretical sites. I'm not entirely sure of this -- the word conjures up an image of hands-on work for me (and is the opposite of theoretical work). I know that it's not intended to mean that.

  • 14
    "Will be useful to others" is impossible to quantify. "Problem you actually face" can be quantified. Commented Jan 21, 2013 at 8:11
  • 23
    @JeffAtwood: "problem you actually face" is detrimental to some sites. "will be useful", while not quantifiable, is better than "problem you actually face", at any rate :/ Commented Jan 21, 2013 at 8:13
  • 14
    @JeffAtwood: ? (Not sure I quite get what you're trying to say here) Not all of the SE Q&As deal with "problems". Stick that label on any science site and you have a homework-help forum. Commented Jan 21, 2013 at 8:21
  • 27
    @JeffAtwood: "So the role of science is not to provide solutions to problems in the world, both big, small, and every size in between?" It depends on the science, but in many cases, no. Most science exists so that we can learn about the world around us. Whether that knowledge is useful in any practical way is often irrelevant. Indeed, "learning by accident" is how many of the most useful things were discovered from science: someone was doing something purely to satisfy their curiosity, and then realized there was a practical application once they got the answer. Commented Jan 21, 2013 at 8:34
  • 8
    @JeffAtwood: But we're not talking about "speculative, discussiony content". We're talking about speculative, objective content. Commented Jan 21, 2013 at 8:40
  • 18
    @JeffAtwood: Yes, some "what if" questions are discussion-y/localised. That doesn't cover all of the questions here. Look at physics.stackexchange.com/q/49905/7433 physics.stackexchange.com/q/51757/7433 physics.stackexchange.com/q/48224/7433 physics.stackexchange.com/q/51293/7433 physics.stackexchange.com/q/48211/7433 . Most of these aren't even problems. These have to do with the user being curious about some phenomenon. If we dealt only with problems, we'd be a homework-help+engineering site. Commented Jan 21, 2013 at 12:24
  • 24
    @JeffAtwood, not only is not all science (and not all intellectual pursuits) oriented toward a practical goal, not all of it is even hands-on, as your wife's is. For people who study theoretical biology, philosophy, Judaism, etc., the "problems" people face are not "How do I do this?", but "How do I understand this?". They are real, answerable problems, but they're not "practical." As long as SE is going to serve these disciplines, the use of "practical" in the definition of an acceptable question is misleading. Commented Jan 21, 2013 at 16:40
  • 6
    "Useful to others" is actually pretty close to the original SO FAQ guidance. IMHO, the biggest problem is that it can be very hard to judge, as an asker, whether or not your question meets this criteria - if Google turned up nothing, how can I know that I'm not the only person on earth who will ever encounter this problem? Should I then not ask, and thereby perpetuate the silence? Not only does it hurt "long tail" questions, it damns novel curiosity in deference to popular topics.
    – Shog9
    Commented Jan 21, 2013 at 16:45
  • 3
    @Manishearth Good point. I think it's possible to construe curiosities conceptually as "problems" (and that that's what we do by posing them as questions), but I agree that using this language in the FAQ is also potentially misleading, at least without careful qualification. Commented Jan 21, 2013 at 16:47
  • 4
    "Problem you actually face" can be quantified. Can it? Because I can lie. As a measuring stick that's even easier to bypass than "is it useful"
    – Zelda
    Commented Jan 21, 2013 at 20:25
  • 5
    @BenBrocka, I think it's more useful to require that questions be written from the point-of-view of genuine curiosity, which can be detected, if not necessarily objectively quantified, in the question. What matters is not the actual life or state-of-mind of the author but rather the quality of the question. Commented Jan 21, 2013 at 20:38
  • 15
    @JeffAtwood You seem to be misinformed about what science is. Science is understanding how things work; engineering is applying that information to build something new. And your wife no doubt has excellent bio training, but she is acting as an engineer when designing something new. If you think there should be no questions about abstract concepts, then math, physics, chem, and bio.SE should be shut down right now, because nothing in those fields is about actually implementing solutions to real-world problems.
    – user206562
    Commented Jan 22, 2013 at 10:28
  • 4
    After some thought... although I think all the proposed alternaste wordings have unavoidable problems of semantics (define "practical", to who and when?) -- this is probably about as good as we're ever going to get without radically changing the direction of the network. So I support this alternative wording. At least it makes the asker frame their question in the altruistic terms of "will this ever be useful to anyone but me?" Commented Jan 22, 2013 at 18:44
  • 4
    @JeffAtwood: Great :) Though I myself am not sure if we even need "practical", it really has no meaning here (and may be detrimental, but I don't think it is at the moment). Commented Jan 22, 2013 at 18:57
  • 3
    @JeffAtwood Some guy once wrote: “So long as your question isn't hypothetical in a meaningless "I'm bored, entertain me" way -- that is, it is interesting to other experienced users, that's probably fine.” But what did he know? Commented Aug 8, 2013 at 12:43

A strict literal interpretation of the "practical, answerable questions based on actual problems that you face" clause would eliminate a very large number of questions on many SE 2.0 sites. Many sites just ignore that part of the FAQ, as it was written for SO and just doesn't fit in all cases.

One important type of question that doesn't fit to this problem-focused definition are questions that are based on curiosity, on trying to learn how something works and gaining a better understanding of the subject.

I'll take one question from Biology as an example that I answered myself: Why is uracil used in RNA rather than thymine?. The question is not based on any kind of practical problem, it was asked out of curiosity. It is also a typical question for someone learning biology, I've asked it myself before SE existed. I don't think this type of question is harmful, I think it is very valuable on the science sites and leads to good questions and answers.

There are certainly problematic questions that are asked out of idle curiosity, and I'm not a great fan of such speculative questions and thought experiments as questions on an SE site.

I think that questions that are asked out of curiosity work fine on many sites. What distinguished the "good" kind from the "bad" kind is that the good questions are still focused on trying to gain a better understanding, and are not only based on idle curiosity.

  • 2
    I may have misinterpreted you here, but generally thought experiment questions aren't too bad. We (Physics.SE) have some ridiculous and unanswerable "what if" types, but there are other really great ones that I've seen which are of the "apparent paradox" type (and give great conceptual insight). Aside from that, I agree completely with the rest of your post :) Commented Jan 21, 2013 at 15:21
  • 12
    was written for SO and just doesn't fit in all cases - exactly. Commented Jan 21, 2013 at 18:19
  • 1
    Thought experiments are very important to study in particular the rather theoretical aspects of physics and any other science.
    – Dilaton
    Commented Jan 22, 2013 at 13:02

I'm a moderator on Russian, Chinese and Linguistics. My experience "on the other side of the wall" has taught me that not all questions are a result of an actual problem. No wait, let me bring this to the next level: Not all sites (always) face "actual problems".

Let me explain this. On sites like SO, or even TeX (examples), you face a problem, e.g. your code shows an error/warning, so you ask about it and eventually you get the code fixed/suggestions on how to fix the code if relevant; after that your problem is solved. On sites like Linguistics, or even the Language-related sites, this isn't often true. If I wonder how a certain expression would be in another language, first I do my research, and if nothing turns up I ask; that isn't an actual problem I face. I could live on without ever searching it. But is it really important in this case?

Another example is Linguistics: if you ask about loanwords, are you facing that problem? Maybe so, maybe not. Does it make any difference? In other words if the purpose of SE is to add content and valuable info to the internet, what harm is there if I ask? Do people really wonder "hey did he actually face this problem"? I think not. On Linguistics I think it never/rarely came up as a problem and with that rule the site might die tomorrow today.

So I ask a question that doesn't derive from an actual problem but it brings something to the site. It's valuable, interesting, it teaches you something. It brings value, it adds to the internet. As long as my question is on topic, well-asked, intriguing, and intelligent, who cares if I didn't face that problem in my life 5 minutes earlier?

  • Is my question less valuable? No.
  • Does it harm the site? It fits the rules, so no.
  • Did it teach me something? Yes.

And again, how do you know if I'm actually facing the problem? No-one can, except me. Well, in case of code-like questions, you can see it's an actual problem, but not all sites treat code. On the contrary, very few sites do. This doesn't mean that the question should overcome the basic rules (on topic, clear, well-asked, research effort), but the "actual problem" factor is really not important and not influent on some sites.

Sorry for the long post, but my point is: SE sites are different from one another. Some basic rules makes sense being network-wide policy. But some sites have different audiences, different needs, different methods and some rules cannot be possibly applied to all indiscriminately.

About the alternate wording, the one that Manishearth proposed is good. If the "that is useful" is prone to be misunderstood, just remove that part like:

You should only ask practical, answerable questions. Chatty, open-ended questions diminish the usefulness of our site and push other questions off the front page.

  • Does "practical," in your view, intuitively characterize the language questions you're describing? Commented Jan 22, 2013 at 14:58
  • @IsaacMoses Are you asking about Languages-related sites or Linguistics?
    – Alenanno
    Commented Jan 22, 2013 at 15:15
  • Given that we're trying to come up with phrasing that works universally, all of the above. I expect that it's more of an issue on Linguisitcs. Commented Jan 22, 2013 at 15:17
  • 1
    @IsaacMoses Well, what's not practical (as in actual vs theory) about Linguistics? Even if we speak "theoretically" about some sounds, it still reflects in real life, since we talk about things you do every day. Every time you speak, write or react to what someone else wrote or said.
    – Alenanno
    Commented Jan 22, 2013 at 15:20
  • 2
    Do the following questions have practical applications? linguistics.stackexchange.com/q/127/238 linguistics.stackexchange.com/q/663/238 linguistics.stackexchange.com/q/315/238 In any case, given that many SE communities invite questions that are inherently not practical, I think the term is misleading in a gatekeeping statement to new users. Commented Jan 22, 2013 at 15:27
  • @IsaacMoses Sorry couldn't answer earlier. I'm not a linguist so I don't have enough knowledge on those subjects to present a complete answer. My take is this: it has similar applications and uses that studying history does. And the two are also closely inter-linked. Perhaps those questions are of no use to you because you don't work in that field, but studying the origins and the development of things helps you understand the present. By the way, that'd be a nice question to ask on Linguistics. :)
    – Alenanno
    Commented Jan 22, 2013 at 16:46

Personally, I think that curiosity is a perfectly valid "problem that you face", and questions should not automatically be closed just because the OP is asking out of curiosity, or with a hypothetical "what if".

Instead, ask if the question is practical for the site, and answerable by the site's targeted community. If yes, then it's fine and should be evaluated the same way you would evaluate any other question on the site.

For example, don't ask "Suppose the White House petition to build the Death Star was approved. How would it work?" on Physics.SE, since it's not really a practical question for that site, but it's probably OK to ask a question like "How does the Death Star work?" on SciFi.

Another reason why I don't like the "based on actual problems that you face" phrase in the FAQ is that I see it used far too often in cases where users are asking about a potential future action in order to make an informed decision, and their question gets closed because the question is not an "actual problem they are facing right now".

I actually recall a situation on the Workplace a while back where a question asking about the consequences of a potential future action got closed as not-constructive by users claiming that it wasn't an "actual problem being faced", and that if the OP had taken the action first (which would have most likely gotten him fired) and was trying figure out how to deal with the consequences, then it would have been a suitable question for the Workplace.

In many cases, it's a better idea to "ask the experts" if it would be reasonable to take an action before actually taking it, instead of taking the action first and then asking how to fix the mess you gotten into as a result.

How about changing the line in the FAQ to be

You should only ask practical, answerable questions that are relevant to this site's scope and audience.

This implies that the word "practical" is in reference to the site's scope/audience, so I think it would work fine in other SE sites, such as SciFi, Physics, Christianity, etc. In addition, it reminds the user that the site has a scope, and that all questions need to fall within that scope to be considered on-topic.

  • 4
    Without knowing which question you're talking about, I'd have to say that, based on your description, the Workplace example was mischaracterized: the problem being faced was whatever led to the choice of action/inaction, and should have been stated as such. IMHO, such criteria is critical on TWP, since questions like "Should I go to work naked?" are inevitable if you don't require a reasonable explanation for why such a decision would need to be made.
    – Shog9
    Commented Jan 21, 2013 at 17:10
  • @Shog9 I'll see if I can find the link in my history somewhere, but that was just an example. I've seen many other cases where a question I thought was good got closed just because it wasn't based on an "actual problem being faced". A recent example would be this deleted question (pre-edit), and its related meta question about the edit I made to phrase it to be an "actual problem faced" to try and get it reopened.
    – Rachel
    Commented Jan 21, 2013 at 17:20
  • 1
    That question had other problems, but I think your edit demonstrates why "real problems" can be so important - the original revision proposed something that seemed at first ludicrous, like something out of a Dilbert cartoon. Explaining the rationale - describing what circumstances led to the proposed solution - improved it considerably. Even then, the actual question was different from what was being asked - the problem is one of preventing culture clash from rending apart a team, while the original question concerned merely a naive solution to this problem.
    – Shog9
    Commented Jan 21, 2013 at 17:29
  • @Shog9 Yes, I didn't think the original question would get reopened, as I explained in the meta question. But I don't want to get off-track from your original question here. The point I want to make is that I've seen quite a few cases where questions are closed because they are asking about some potential action being considered as a solution to a problem, instead of asking about the actual problem being faced itself. I don't think these should be considered "not-constructive" and closed, as many users are likely to come up with the same proposal and want to evaluate it the same way.
    – Rachel
    Commented Jan 21, 2013 at 17:39

My previous answer, it's true, did not really address the question.*

After some thought I support manishearth's suggested reformulation:

You should only ask practical, answerable questions that will be useful to others. Chatty, open-ended questions diminish the usefulness of our site and push other questions off the front page.

Since it maps well to what we do. So upvote his answer on that basis if you agree.

However, I do want to defend ...

Focus on questions about an actual problem you have faced.

... a little, on the basis that if you care enough about figuring something out, it is in fact a problem to you. You just can't stop wondering why X happens, it nags at you, it keeps you up at night, you're fascinated with learning more about it, you constantly talk to others about it. It's a problem.

Or are you just bored and want to be entertained with some blue sky daydreaming and idle curiosity? If you can walk away from your question and not care too much about the answer, that's no problem at all, is it?

Make us believe your question is important to you, and to anyone else who will ever read it. Because by God, this is a bona fide problem we're all facing here!

Include details about what you have tried and exactly what you are trying to do.

In my mind, there is very little difference between that phrase and asking someone if they have researched their question before asking -- what did you learn? And can you share that with us?

If you can't explain why you are curious about this, why it's interesting, and define what exactly it is you're even asking by framing it in what you already know.. well, the hallmark of idle curiosity is that no effort is expended.

I asked because I was curious.

Is no justification at all, but a get out of jail free card, a recipe for endless no-effort one-liner questions. Explain your curiosity to us. Share what you know, and why you are curious. Convince us this is something we should care about, too.

Why is this problem interesting?

** Though to be clear, the SO community has ruled quite definitively that "I learned something" isn't enough to justify the existence of a question and its answers. The question has to support directed, focused learning, not "random list of stuff that someone somewhere could probably learn something from." That's why "hidden features of X" questions are barely tolerated any more, and even "strangest language feature" which was used as a positive example in the blog, is no longer welcome, though they were encouraged at one point...

  • 7
    This. I don't care at all if the problem is actual or practical (by whatever definition of practical), what I care about is the asker being serious about it, at least serious enough to have done some minimal research before s/he resorts to a community of volunteers. Intellectual curiosity is as great motivator as any, idle curiosity is a different beast entirely, one that, if entertained, will ultimately suck the life out of the community.
    – yannis
    Commented Jan 23, 2013 at 6:26
  • 4
    Yep, this is how we have been interpreting "problem" on our sites, so I agree :) I guess the intention was the same, just that the wording was highly misleading. I completely agree that most "I was curious" questions can be crappy (even if not of the make-a-list type) Commented Jan 23, 2013 at 12:54

I propose:

Answerable questions based on relevant issues that you want an explanation for


  • Answerable questions, I think we can all agree, sums up what the SE model is built to handle.

  • relevant incorporates the on-topic requirement and also points toward "useful to others." Hopefully inspire in this asker's mind the question "Is this relevant to anyone but me?"

  • issues is a more general term than "problems" that includes both practical problems and points of curiosity, inspiring the question "What is at issue here?"

  • you want an explanation for1 determines whether an on-topic question is "worthwhile" or not. In practical disciplines, practical problems faced by the asker prompt a desire for an explanation. In theoretical disciplines, interesting-to-that-field issues that the asker is curious about prompt a desire for an explanation. This also excludes questions that couldn't have come from genuine curiosity/need. Finally, "explanation," as opposed to "advice," "ideas," or "discussion," describes questions that can be answered objectively.

1. This phrasing was inspired by Jeff Atwood's answer to the question of whether SF&F, although not about "how to," can work on SE.

  • I'd prefer some alternative wording like "Practical" instead of "deserve an explanation". I can see that easily getting misused and users leaving comments quoting the FAQ and saying things like "this doesn't deserve an explanation". I like the term "relevant" as an alternative to "useful to others" though, and the point you made about it incorporating the on-topic requirement.
    – Rachel
    Commented Jan 21, 2013 at 17:51
  • 2
    @Rachel, the point of this exercise is at least partly to replace "practical," which can be misleading in the context of theoretical disciplines. I agree that "deserve" may be emotionally loaded, but there is, like it or not, a value judgement to be made here. Do you have another idea? Commented Jan 21, 2013 at 17:54
  • Not right now, but I'll let you know if I think of anything. I think you have a decent start here though :)
    – Rachel
    Commented Jan 21, 2013 at 17:59
  • @IsaacMoses: I agree regarding "practical", that would discourage half the good questions on Physics. Instead of "deserve", how about "merit" (just an idea, not really sure if it's better. Commented Jan 21, 2013 at 18:02
  • Merit is as subjective as deserve. Transferable may be a better word. @man
    – random
    Commented Jan 21, 2013 at 18:04
  • 3
    @Rachel, et al., I've replaced "deserve" with "you want," which is more parallel with the original "problems you face." Commented Jan 21, 2013 at 18:19
  • @IsaacMoses I'm not sure if that's a suitable alternative to some SE sites... "You should only ask answerable questions based on relevant issues that you want an explanation for". I don't think it would go very well for objective sites like SO, where you are most likely going to get a simple answer to your code problem, not a complete explanation. What about "You should only ask practical, answerable questions that are relevant to this site's scope and audience"? This implies "practical" is used in terms of the site's scope & audience, which sounds OK to me.
    – Rachel
    Commented Jan 21, 2013 at 19:36
  • 3
    @Rachel, I still think "practical" is misleading. On Physics.SE, it might be taken to mean "experimental/applied only; no pure theory." On Judaism.SE, it might be taken to mean "practice only; no text interpretation." In both of these cases, it looks like you're excluding whole areas of study with canons from which to derive objective answers. Commented Jan 21, 2013 at 19:44
  • @IsaacMoses Hrrm good point. I tend to think of "practical" more in the terms of "realistic", such as don't ask things like "If were were all aliens that could do X, how would we do Y?". Still, couldn't the potentially unpractical on-topic subjects, like pure theory in Physics.SE, be covered elsewhere in the editable parts of the FAQ, such as the "What kind of questions can I ask here" section?
    – Rachel
    Commented Jan 21, 2013 at 19:54
  • 1
    Why not just pick one of these wordings which is inevitably vague in some way and link those words in the FAQ to a fully explanatory meta post?
    – Double AA
    Commented Jan 22, 2013 at 3:48
  • 1
    @DoubleAA, I think it's optimistic to expect many readers to click through to the commentarie. It's important to pick a wording that will be closest to what you're actually looking for in the minds of most readers. Commented Jan 22, 2013 at 5:35
  • 1
    @Rachel: We already have that, and that just makes it look self-contradictory. Commented Jan 22, 2013 at 5:36
  • 1
    I agree with @double that there is no magical one-sentence formulation that is going to capture all this. They are all vague in their own way. Define useful. Define relevant. Define practical. Define problem. To whom? How? Answerable seems strongest since it can be quantified somewhat, if there are dozens or hundreds of answers, clearly that bar is not being met and we have a discussion. Commented Jan 22, 2013 at 18:31
  • 3
    @JeffAtwood, this is an exercise in instruction, not technical specification. We should do our best to make the point we want to make as clearly and accurately as possible to the target audience, which is new users in a wide range of disciplines. Commented Jan 22, 2013 at 18:44

The current text…

Focus on questions about an actual problem you have faced. Include details about what you have tried and exactly what you are trying to do

…seems to demand that questions are grounded in practical reality, but I suggest that the real benefit is a side-effect. What is good about questions grounded in practical reality? Is it that they are not theoretical? Is it that they are the only kind of questions that fit the format? No, it's that they are not a moving target.

If you ask a question about a problem you faced today, and someone comments asking for more information, you can provide it without ever changing the nature of the real underlying question: you are now merely communicating it better.

On the other hand, if your question was never grounded in reality in the first place, you are free to 'evolve' it as your thinking develops, and this is a disaster for the Q&A format.

I suggest we try and change the wording to keep the 'stationary target' requirement but remove the need to root this necessarily in a historical event ('you have faced', 'you have tried').

There are probably many ways of approaching this, and all might have their shortfalls. It may be better to have a short centralised list of texts from which the most apt can be chosen by each site. Here is my best effort for what it is worth:

Ask questions that are 'real'—that you understand fully now. Getting an answer might make you want to ask another question, but it shouldn't make you want to change your question!

  • 2
    +1. I don't think your proposed copy is particularly useful, but your point that one of the major genres of questions we'd like to curb in the first place is the moving targets that result when a question is based on some sort of theoretical need rather than an actual one is very salient. For any real genuine question, there is going to be a satisfactory answer. For all others, they are likely just to morph as the user wanders along whatever pathways are allowed.
    – Caleb
    Commented Mar 16, 2013 at 13:48
  • @Caleb I've tweaked my proposed copy to make it more focused on the 'stationary' issue. Do you think it is improved or am I not addressing what you see as the problem with it? Commented Mar 16, 2013 at 14:07
  • That is improved. One issue I see with it -- at least in the context of C.SE -- is that quite often it is really geniune questions that the OP does actually fully understand the question until they start getting answers. "Fully understand the question now" is not something we usually start with when people are asking about doctrine or tradition of churches. Very real genuine solid questions often start with some uncleanness on what the question is. Often times if they knew exactly what the question was, they would already have their answer and it would no longer be a "real problem they face".
    – Caleb
    Commented Mar 16, 2013 at 14:16
  • @Caleb that's bonkers, I pity you c.se mods ;) Commented Mar 16, 2013 at 17:57


You should only ask non-trivial, answerable questions that demand expertise in our site's subject. Chatty, open-ended questions diminish the usefulness of our site and push other questions off the front page.

The current FAQ standard, while quantifiable, depends entirely on the judgement of the asker. Long experience suggests that once a person asks a question on a Stack Exchange site, they are no longer unbiased about it. Therefore, it's not very helpful to ask them if it's an actual problem that they face; people will go to great lengths to post-justify a question once asked.

Besides, motivation isn't really the problem here. The problem is described in the second sentence: bad questions displace good ones. In truth, whether the asker is sincere hardly matters compared to whether the question does a good job engaging the sorts of people who provide excellent answers. Simply having a problem is no guarantee that your question will be interesting. Homework questions tend to be poor even though they are "practical, answerable questions based on actual problems that you face" because they are often trivial for experts.

Before Philosophy.SE was allowed to go into public beta, we had to answer what our site was about. As Paul Graham points out, the entire field has been dominated by "the exploration of knowledge that has no practical use." (As it happens, I disagree with most of that essay, but it does seem true that as soon as a bit of philosophical knowledge becomes practical, philosophers loose interest in it.) So how do you test if an asker is sincerely asking a question and if the answers will actually matter to them? Rather than psychoanalyzing the user, we decided to focus on "domain knowledge".

The question of motivation is pretty well addressed a little later anyway. ("If your motivation for asking the question is...") What makes a Q&A site useful is a group of "experts" who are willing to watch the questions and provide reliable answers. Asking impractical questions on SO risks driving off the folks who answer questions there because SO is about practical programming. People answering questions on philosophy are ambivalent about practicality, but care deeply that questions somehow engage the existing literature. In both cases, questions that don't demand the sort of skills and knowledge that experts have obtained are likely to be a distraction to the core contributors of a site.


Get expert answers to detailed questions

Focus on questions you've seriously pondered. Include details about what you have already learned and what you are trying to find out. Putting more thought into questions before you ask will help you get better answers.

The about page is focused more on guiding new users, so instead of focusing on why the policy protects the site, we should try to persuade them to ask good questions for their own good. I picked the reason that is perhaps the most general. But there are plenty of reasons for asking well-specified questions:

  • You might find your own answer and not need to ask. (Or better: self-answer.)

  • Understanding your own question helps you understand the answers.

  • Upvotes.

But hopefully, the proposed text will help askers reorient themselves toward asking questions for the sake of learning from the answers.


Rather than make the requirement contingent on the asker, point out that people shouldn't ask questions that are boring to an expert.


So long as your question isn't hypothetical in a meaningless "I'm bored, entertain me" way -- that is, it is interesting to other experienced users, that's probably fine.

                                                — Jeff Atwood

I tend to agree. The ultimate motivation between the question lives in the asker's mind, and it cannot be known without telepathy. Let's not worry about something we don't know and have no way to find out. Evaluate the question on its own merit.

Is the sock question practical? Not applied to socks, no. But applied to something else? Why not? Socks are just the way the question is presented. Concrete examples help in understanding a problem and imagining a solution. So yes, the sock question is eminently practical. And it's evidently answerable: it has received good answers. There is a problem to be solved. Not all problems resolve to “solving this will bring on world peace” or “I can make money from this”. “I don't understand this and I want to” is a problem to be solved, and it's one that many other people often face.

Strictly-speaking, problems like that fail the "practical, answerable questions based on actual problems that you face" test. And yet, I would keep that question simply because I learned something useful from the answers

You don't need to compromise your principles or abandon your litmus test. The test works. You just need to recalibrate your actual-problem-o-meter.


Objections to "Practical" Questions

The objection to requiring practical questions faced by the poster seems to boil down to the idea that it would render everything on theoretical, faith-based, or genre sites off-topic. This is a misapprehension of the Q&A format and the focus of these sites.

Theoretical Science is Still About "Practical" Questions

The scientific method is about forming and testing a hypothesis, making it a practical problem to be solved in the sense that it is seeking a practical approach to testing a hypothesis or proving a theorem. Idle thought experiments without any interest in testing the hypothesis are just intellectual onanism, and not suitable for the Q&A format.

Philosophy, Religion, and Fiction

The rest of the objections to "practical" questions seems to lie in assuming that questions about soft subjects requires a lack of rigor. If that were true, philosophy students wouldn't be required to take logic courses, and religious studies wouldn't require research skills or bibliographies on term papers.

While I think that we can all agree that asking questions about Cylons is not "practical" in the same way as an applied sorting algorithm expressed in working code, it is at least a question can be answered based on specific expertise, accepted canon, or research within the available literature/media.

Note that I'm not saying this is a great question; it isn't, for a number of reasons, including the complete lack of a working hypothesis by the OP. However, it is certainly on-topic for the site, answerable by those with knowledge of the domain, and allows for the possibility of a canonical answer. That should really be a baseline threshold for almost any Q&A topic I can imagine.

  • 4
    While it's possible to come up with an interpretation of the word "practical" that includes anything you want an objective answer to, I think that its most obvious meaning excludes anything purely theoretical or intellectual, and that it's therefore misleading in a rule of thumb presented in a FAQ. Commented Jan 21, 2013 at 19:03
  • 2
    As I already mentioned, the scientific method only comes into play when you talk about creating new science, not using existing established science. The science sites do the latter. While some "idle thought experiments" are crappy (eg "What would happen if the moon hit the Earth"), and are too broad to be answered, others bring up valid theoretical concepts (I'm talking about thought experiments that lead to apparent paradoxi, etc, like physics.stackexchange.com/q/23824/7433). Don't generalize. Commented Jan 22, 2013 at 5:40

I agree with Manishearth. It would be helpful if we had a few examples of questions that should not be closed and this FAQ item is the only reason for closing them (too-localized, etc. don't apply to them).

I might be wrong but as I understand this FAQ item tells me that we should ask a question only if we really care about knowing the answer to the question. Every day we can think of several out of curiosity questions but we don't really care about the answer for many of them. We don't seek an answer to them. And when we don't care we don't spend time thinking about them, we don't try to find the answer to the question by ourselves. It is likely that if we ask a question we haven't spent time thinking about we end up with an unanswerable question because we don't know what we want to know and we don't really care much about the question so we would leave it there.


To play off of Jeff's answer, maybe there's some middle ground, focused learning-by-accident. What I mean by that is while the SO way has been a solid line directed at a goal (vector):


and the all out forums are more like big circles around the goal:


maybe what is optimum is a small slice of pie, a bounded area that keeps close to the straight line, but gives a little flexibility.

Circle with Pie Slice

So you're not just learning anything by accident, you're learning stuff directly related to your field. I've found many ideas in my life by accident that I was able to use in my coding and other computer work because it's the idea that counts. Once you get that idea, your specific implementation might be just what you needed to improve how you code. Heck, I have gotten a lot of good stuff like that off of Jeff and Joel's blogs.

Currently, SO actually does use a slice of pie, because there is no perfect way to stay on the line, I just think that it can handle a slightly larger slice.

I don't plan on ever sorting my socks that way, but the next time I have a large sorting task, I'm planning on rereading the answers to see if I can do better than I've done before.

  • 1
    Then it should stay on blogs, no? Commented Jan 21, 2013 at 8:12
  • 1
    @JeffAtwood, yes, the blogs are a better format when just one person has a great idea(s). The benefit of the format here is that you can get multiple good answers consolidated on one question/topic, and immediate feedback from interested others. Commented Jan 21, 2013 at 14:25
  • @JeffAtwood, maybe someone needs to create a blog environment modeled after Stack Overflow, allowing good long answers and voting. Commented Jan 21, 2013 at 16:44
  • 2
    @Lance: meta.stackexchange.com/questions/110998/… - I still think it's a good idea, but a big part of that is pure hatred for WordPress.
    – Shog9
    Commented Jan 21, 2013 at 17:49
  • @Shog9, thanks, great link. Commented Jan 21, 2013 at 17:58

Are we justifying the existence of many sites that primarily ruminate over thought experiments? As Q&A sites go, they're a great way to fill the page and time as questions focussing on real, actual problems are shut down for being too localised or not ethereal enough.

If more than a third of a site's questions are "what if?" then you either remove the rule completely and surrender, calling the forums like it is, with smooth voting UI, or actually look at the bubbling overthrow.

Questions missing the right sense of humour and failing to track on social networks are closed justly as the non real questions they are. Not every question will find a home on a Stack Exchange site. Unless it's showered in upvotes and long, long answers. Then it's more equal than others.

When a hypothetical hits the jackpot it gets elevated and projected into the echelons of questions you search for for their own sake and stick around for that reason alone. They are referenced more for their novelty and notoriety than the actual solutions presented. You can learn from them, but is what they present the right fit or is it just a mindshare grab at work?

As more of these are welcomed and branded under the Stack Exchange umbrella the more you become a Ponder & Debate network, not a Question & Answer one.

  • 8
    See the examples here. Point is, you generally don't face science "problems". You face curiosities, apparent paradoxi, and the likes, and these help others. "Problems" you face in science are generally engineering/homework things, and don't help others at all. Commented Jan 21, 2013 at 16:35
  • 3
    Then they're not fit for the Q&A model designed here, but elsewhere, like Quora
    – random
    Commented Jan 21, 2013 at 17:45
  • 4
    I don't see the connection. These sites are thriving with the Q&A model. Such questions generally have no problems (as in, they don't have a higher frequency of other problems when compared with "I have a pr0blem" questions) other than their conflict with the "must be a practical problem you've faced" rule. They are helpful to many others, they are interesting to read, they don't promote discussion, and the make the Internet better. Plus, they are fitting the Question+Answer form (it's not like they are twisted into questions, the question is a natural one). Commented Jan 21, 2013 at 17:49
  • 2
    My point (see my last sentence in my last comment) is that these fit naturally into the Q&A form, they don't need to be "twisted"("shoehorned") into it. Commented Jan 21, 2013 at 17:59
  • 1
    And that is why it's a Ponder & Debate setup, not the Q&A as designed
    – random
    Commented Jan 21, 2013 at 18:01
  • 7
    Are we justifying them? Not really, but if they have no justification then we're certainly wasting a lot of time and effort that could be better spent elsewhere. There is an assumption here that if enough folks are willing to ask these questions, and enough folks are willing to answer them - not discuss them endlessly, but actually attempt to resolve whatever doubt, concern, or confusion is expressed by the asker - then there is a real need for a venue such as this to house them. This assumption has not always proved correct, but it is still the one we're operating under.
    – Shog9
    Commented Jan 21, 2013 at 18:04
  • 5
    The questions @Manishearth posted were clear, intelligent, and garnered clear responses without devolving into forum garbage. If the only problem with these questions is that they fail the "Problem you face" test, then we should re-think the "problem you face" rule for these SE sites. Commented Jan 21, 2013 at 18:07
  • 6
    @random, I think you're confusing "objectively answerable" with "practical." These are largely orthogonal concerns. The former is what the SE model (not just the technology, but the whole ethos) works well for. The latter is a characteristic that's desirable, and perhaps even essential to relevance, in many contexts, but not in all. Commented Jan 21, 2013 at 18:07
  • 5
    @random: [I'm talking about Physics and Chemistry only here, I don't know enough to talk about other sites] Theoretical questions don't usually trigger pondering and debating. There generally is enough established science behind such questions so that they can be reasonably answered, without debate (everyone agrees) or pondering by anyone well-versed with the subtopic.. This is science I'm talking about. It's supposed to work like that. Commented Jan 21, 2013 at 18:07
  • 2
    The only place where this doesn't work is cutting-edge science which is being researched -- one can have some debates there. But, as long as it is a mainstream theory being researched, and as long as the theory being talked about is mentioned, it almost never causes a problem. Non-mainstream physics/chem ("I have a theory!"/or questions talking about fringe theories) are the problematic ones -- and they're off topic. Commented Jan 21, 2013 at 18:08
  • 3
    @Manishearth This is just wrong all the way around. The scientific method is about forming and testing a hypothesis, making it a practical problem to be solved in the sense that it is seeking a practical approach to testing a hypothesis or proving a theorem. Idle thought experiments without any interest in testing the hypothesis are just intellectual onanism, and not suitable for the Q&A format.
    – CodeGnome
    Commented Jan 21, 2013 at 18:32
  • 3
    @random: What's the brand? I really doubt "questions on problems you actually face" is a big part of it ("Make the Internet Better" is good enough as the "brand", on the other hand). It seems like that was just a tangential rule put there to deter Gorilla vs. Shark questions -- which generally works on SO, but there are too many false positives on other sites. A lot of these rules have been made with SOFU in mind. If you're expanding the network to other sites, you have to remember the intention behind these rules and apply the accordingly. Otherwise, don't expand the network :/ Commented Jan 21, 2013 at 18:33
  • 5
    @random, The SE brand, as it exists today, was born in May 2010. The first theoretical discipline's site to launch under that brand went into public beta in July 2010 and was launched in October 2010. I'd say that horse left the stable shortly after the stable was built. Commented Jan 21, 2013 at 18:33
  • 5
    @CodeGnome: Wha? I'm not talking about that. I'm saying that when talking about established science, there is only one answer that you can reach, and you reach it without pondering and debating. I am not talking about the scientific method. I am not supporting all "idle thought experiment" questions. Take a look at the questions I linked to.. These are questions about established science that get a good, conceptual answer. Commented Jan 21, 2013 at 18:37
  • 4
    @CodeGnome: In fact, the scientific method doesn't even feature here. The SE sites aren't trying to expand and build upon science. We're trying to explain the existing science. Commented Jan 21, 2013 at 18:38

You must log in to answer this question.

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