What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 128 Stack Exchange communities.

Recently there was a question asking about generating all subsets of a set using a stack and a queue, which was closed (and now deleted it seems) as not a real question for no good reason, since it didn't fit into any of these conditions:

It's difficult to tell what is being asked here.

No, it was clear what was being asked.

This question is ambiguous, vague, incomplete, or rhetorical and cannot be reasonably answered in its current form.

Not ambiguous, not vague, not incomplete, definitely not rhetorical and could easily be answered if one knew the solution.

Now, the exact same thing has happened with this question:

http://stackoverflow.com/questions/2791982/a-shortest-path-problem-with-superheroes-and-intergalactic-journeys

I am interested in hearing a logical argument for why that question is either ambiguous, vague, incomplete, rhetorical or cannot reasonably be answered in its current form.

It seems that (the same bunch of) people like to close questions that they think are homework questions, especially when they think people want to be served the solution on a platter, which is also not the case:

Any suggestions or ideas of how this problem might be solved would be most welcomed.

Most of the time the people asking these questions are very reasonable and appreciate even the most vague idea, yet their question is closed.

Let's go further and assume that it IS a homework problem. So what? When I registered here I didn't see any rule that said not to post homework problems, nor do I see such a rule now. What is wrong with posting homework problems that makes people hunt them down with a passion to close them without even reading the entire question body?

This site is full of questions asked by people who get paid to know the things they are asking, yet their questions are considered fine. How is solving someone's homework problem worse? In some places (like where I live), computer science is a mandatory high school subject, and not everyone is interested in it. How is helping at least those people worse than doing someone's JOB?

Not answering homework questions is fine and it's everyone's choice, but I consider closing them to be an act of power abuse, selfishness, and an insult to the fellow community members who are also interested in a solution or want feedback on their proposed solution.

So my questions are:

  • Why do questions like the above get closed for reasons that do not apply? Why do you close them? Why don't you?
  • Why doesn't a vote to reopen a question reopen it automatically? Needing 5 votes for a reopen takes too long, and it's not fair because one reopen vote basically cancels out a close vote, making it 4 close votes (or 5 to 1, which is the same as only 4 people wanting to close the question), which isn't enough to close the question. I think a question should only be closed when CloseVotes - ReopenVotes >= 5.

I'm hoping this will stay up, but I realize it probably won't. In either case, I think this is worth saying and discussing, since it IS community-related.

share|improve this question

migrated from stackoverflow.com May 8 '10 at 13:35

This question came from our site for professional and enthusiast programmers.

1  
Hear hear, word, QFT –  Moak May 8 '10 at 14:35
6  
-1 because there is no such thing as a question that gets closed for no reason. Some questions may be closed for bad reasons, but that's why the reopen mechanism exists. In the particular case you linked to, I agree with Bill and Lasse. –  Pops May 8 '10 at 15:58
    
@Moak: what is QFT? –  Pops May 8 '10 at 15:59
    
@Popular Demand: QFT = Quoted For Truth, ie ,+1 –  Sathya May 10 '10 at 6:24
    
See this feature request: meta.stackexchange.com/questions/125/… –  Lance Roberts May 10 '10 at 16:55

8 Answers 8

The question you link to has a problem, it is a question simply because Stack Overflow has questions, which then have answers.

Let me explain.

Just because you have a problem, or a challenge, does not mean you have a question. The distinction can sometimes be subtle.

Here's an example:

  • I need to build a car. How do I do that?

For sure, this is a question in the grammar sense. The question can be rewritten to a single sentence so that it reads "How do I build a car?"

However, for a site like Stack Overflow, or any of the sites in the series, this is not a good question. It isn't one that can easily be answered.

For instance, the actual answer to such a question depends on:

  • What kind of car do you want to build?
  • How much knowledge about building cars do you have?
  • How much technical sense do you have?
  • What kind of tools do you have available?
  • Have you actually tried building the car?
  • How far did you get?
  • Did you have any problems while doing so?

As such, such a question, even if SO was about cars, would most likely be closed, probably for the same reasons the question you linked to was closed.

However, let me show you a different question, that could very well be asked by the same person, that would have a much higher chance of both not getting closed, as well as actually getting good answers.

I'm building my own car, and I have a problem. I've managed to get the engine running, but whenever I hit the gas-pedal to try to accelerate it, it sounds like it dying or choking, and seems to struggle to operate. If I let go of the pedal, it goes back to running as normal. What should I look at or look for?

Do you see the distinction about the original "How do I build a car?" question and this one?

The point is, you should strive to build a specific question, not just any question. Additionally, it helps to give as much information as possible, not just about the problem at hand, but also about what you tried.

For instance, the engine problem above could stand to be accompanied by information like:

I've looked at the carburator, it seems to be operating properly, is clean, and has no obstructions. Also, all the plugs are firing (I verified this by unhooking one by one and noticing that the engine ran imperfect for each), I think I've gotten the right gas/air ratio because if I adjust this, again the engine runs worse than it does now.

This should explain why most homework questions has such a low chance of getting answers. Homework questions are typically worded like a problem, but not like a specific question. Saying "Work out a way to build all subsets of a set using only a stack and a queue" is not a question, that's a problem. Showing us sample code and asking why it seems to miss a few of the subsets, now that is a question.

But as I said, the distinction can be subtle, which is why there isn't a single vote needed to close any question, there has to be a consensus by at least 5 people to do it.

share|improve this answer
    
But there are problems for which you have no idea where to start. Whats wrong with asking "I want to build a car. Where do I start?". It might be of interest to others and can be answered reasonably, and it might also give those that like building things a good challenge. –  IVlad May 8 '10 at 14:18
8  
@IVlad: It can't be answered reasonably. The only reasonable answer is "go read a book on it or something." Of course, asking for a list of getting-started books would be a reasonable question, and hey, look at that - we have them on Stack Overflow! –  Aarobot May 8 '10 at 16:00
2  
@Lasse: excellent example and discussions. I vote that this is edited and added to the SOFU FAQ. –  IAbstract May 8 '10 at 16:30
    
@Aarobot: the metaphor is a little off in this case. That question basically asks "How do I build a car that runs on less fuel?", which can be answered by assuming the OP knows how to build a classical car and telling him how to use that knowledge for another type of car. In any case, dropping the silly metaphors, he only needed references for a few graph theory concepts, which he got, so I don't see what the problem is, since even you say questions that ask for references are fine. That's what he asked for: ideas. –  IVlad May 8 '10 at 17:26
1  
@IVlad: Your "better analogy" isn't much better. How do you build a car that runs on less fuel than what? How much less? What kind of car? What kind of fuel? What parts are available? And what equipment? What background does the author have in building cars? What will the car be used for, and how often? What is the budget? The question is nowhere near specific enough for the answers to be of general use or interest. If the OP "only needed references for a few graph theory concepts", then he could have asked for those references specifically; that would have been a valid question. –  Aarobot May 8 '10 at 17:50
    
@Aarobot: no metaphor is good, precisely for the reasons you stated. There's no need to use metaphors. It's just an online judge problem, as you should be able to tell if you've ever been to one. The question is clearly specified, and the poster asked for suggestions. There's no need for him to post his CV, as few other question askers post their qualifications and that's never a problem. If they understand the solution great, if not they'll ask follow-up questions. The question is definitely specific enough and of interest, hence the +5 votes. Are you saying online judges are of no interest? –  IVlad May 8 '10 at 18:00
2  
@IVlad: The question is not clearly specified. There was no question. There was a list of specifications for an assignment with a line at the end saying "help me." That is not a question. The +5 was not for the question, it was for one of the answers which more or less said "solve the problem yourself" - people voted him up for telling the guy to do his own homework. –  Aarobot May 8 '10 at 19:14
    
@Aarobot: the question was definitely there. Saying you're looking for pointers is as good as asking for them. And you're not paying attention yet again, the question itself also got +5. –  IVlad May 9 '10 at 5:59
2  
@IVlad: Saying you're looking for pointers without saying what you're looking for pointers on is not a question. "I have this assignment. What should I do?" is no better than "I'm hungry. What should I eat?" or "I'm bored. Any ideas?" And I don't know what you're on when you say the question got +5, I clearly see a zero there. –  Aarobot May 9 '10 at 13:33
    
@Aarobot, click on the zero (if you have 1K on that site) and you'll see +5/-5. –  Lance Roberts May 10 '10 at 16:50
    
@Lance: That's nice, but does the total number of upvotes really matter? We all know the phenomenon of pity upvotes / cretinous random upvotes. –  Aarobot May 10 '10 at 21:53

I think the following comment from the question you linked to is good advice:

You would get much better response if you distilled this question into a technical question, not some rambling storyline with a problem hidden in it... – John Weldon

It's hard to talk about this sort of thing in the abstract. Specific examples need to be dealt with on a case-by-case basis, which is what the community does when it votes to close. I think the one example you gave was a good close.

share|improve this answer
    
Never said the question was well-worded, but you could easily understand what was required. This might be a good reason to downvote, but not to close. –  IVlad May 8 '10 at 13:42
4  
@IVlad: If it's a simple matter of poor wording, I'm the first person to jump in and edit. You can't expect people to read through several paragraphs of text to get to the point, though. Most of us are busy (despite appearances). Questions need to be distilled down to their technical essence. –  Bill the Lizard May 8 '10 at 13:46
    
I agree that there is filler in that question, but I think it's kept to a minimum. Writing the problem rigurously using graph theory terms would result in a similar amount of text, so I don't think one would waste more than a few seconds reading through it. –  IVlad May 8 '10 at 13:53
    
thanks for the kind comments :) –  John Weldon May 11 '10 at 17:24

"Not a real question" tends to get used a lot for "There might be a real question in here, but we're not sure where and don't want to swim through the sea of text to find it", which I suspect was the case there; that guy obviously copy/pasted the question directly from his homework assignment (or from a coding challenge site; that looks like the sort of thing Google Codejam does). People are willing to answer homework questions to a certain point, but people need to at least find the question and post that.

As for:

Why doesn't a vote to reopen a question reopen it automatically? Needing 5 votes for a reopen takes too long, and it's not fair because one reopen vote basically cancels out a close vote, making it 4 close votes (or 5 to 1, which is the same as only 4 people wanting to close the question), which isn't enough to close the question. I think a question should only be closed when CloseVotes - ReopenVotes >= 5

Right now only 5 people can vote to close, so that's the max close votes something can get. You look at the post and think "aww, just one reopen vote would push it under 5", but that's because there's a hard limit at 5 -- if that didn't exist, I'm sure more than 5 people would've voted to keep it closed. I look at it as there's a motion to close the post, and 5 people voted yes, so it's closed. That motion is over -- if you want a separate motion to reopen, you can, but that's unrelated to the close motion

share|improve this answer
1  
A motion allows you to say no while the motion is still going. This is more like "I think this should be closed, if anyone is with me raise your hand, and if you're not with me I don't care, you'll get your say later". Maybe 10 people would have voted to close if there was no limit, but maybe 20 would have opposed it. I think that question was of interest to other programmers out there, since it posed an interesting variation on a classical problem. If people don't like homework questions that's fine, but as long as there are no rules against them, closing HW questions is going a bit too far. –  IVlad May 8 '10 at 14:14
3  
@IVlad, but it wasn't closed as a homework question, but as "not a real question for a Question & Answer-like site". –  Arjan May 8 '10 at 15:10

So my questions are:

  • Why do questions like the above get closed for reasons that do not apply?

Because the community thinks that they do apply. Stack Overflow is not a court of law, it is a community-run and community-moderated site, and if people want to stretch the definition a little because they think a particular question is terrible or simply not in the spirit of Stack Overflow, that's their prerogative. That's democracy. Tyranny of the majority and all that.

Meta is your "court of appeals." If you feel that a specific question should be reopened, it's fine to discuss it here and explain your reasoning, but please, don't turn it into a general complaint about closing, there are already far too many of those.

Why do you close them? Why don't you?

I tend to stick to the more literal definitions and mostly close as Exact Duplicate or S&A. "Not a Real Question", for me, is mainly reserved for questions that actually aren't questions, they're meanderings or rants occasionally thinly-disguised as questions.

Nevertheless, there are exceptions to the rule, and this was definitely one of them. Stack Overflow is not rentacoder (or in this case borrowacoder). There's an issue of precedent here; if we start allowing people to post questions that effectively say "do my homework/project for me", then the site will quickly become flooded with them. So we have to draw a line in the sand somewhere; it may be a fuzzy line that gets covered up and redrawn from time to time, but it's there and it's necessary.

  • Why doesn't a vote to reopen a question reopen it automatically? Needing 5 votes for a reopen takes too long, and it's not fair because one reopen vote basically cancels out a close vote, making it 4 close votes (or 5 to 1, which is the same as only 4 people wanting to close the question), which isn't enough to close the question. I think a question should only be closed when CloseVotes - ReopenVotes >= 5.

If I didn't know any better I'd think you were trolling; this is frankly one of the craziest suggestions I've ever heard.

Essentially you are telling us that every time the counter dips below 5, it should be reopened. We already have problems with close/reopen yo-yo questions, to the point where it's been suggested (and I agree) that we make this more difficult.

Your version of this would result in questions being closed and reopened within seconds of each other, make it nearly impossible for the author to properly edit the question while it's closed, or for anyone else to properly answer it while it's reopened. This is a bad idea. A very, very bad idea.

share|improve this answer
    
While this is a very bad idea, he has a point. He identified a problem, and proposed a bad solution that actually made it worse. But still it is a problem. There might be a better solution. –  forget it May 8 '10 at 16:17
    
Closed questions can be edited just fine. This basic proposal of a running close/reopen tally has been made before to reduce the yo-yo effect (e.g. 15 people vote close, 10 vote keep-/re-open, it stays closed continuously instead of yo-yoing 5 times (c-r-c-r-c)), and I think, if done right, it would be better (but there's other, much better posts on meta about it). For example, only changing the status 15 minutes after the most recent vote (based on the current tally at that time) gives plenty of time for people to have their say with the only yo-yo happening over hours or days. –  Gnome May 8 '10 at 16:28
    
@Lo'oris: I disagree that it is a problem. I think I explained my rationale very clearly. –  Aarobot May 8 '10 at 17:01
    
@Gnome: The difference between the proposals you refer to and this proposal is that the running tally would have to go all the way back down to 0, or all the way back up to 5. I think that's a good solution. Reopening the question as soon as one single person votes to reopen is a terrible idea that would make it nearly impossible to keep a question closed. And yes, closed questions can technically be edited just fine, but it's going to be very difficult and annoying if keeps getting closed and re-opened while you edit - you'll keep getting that irritating warning message. –  Aarobot May 8 '10 at 17:04
    
Democracy is anything but the tyranny of the majority. The question poster didn't ask anyone to do his homework for him, as I keep saying, but apparently everyone is ignoring that part. He just posted an interesting problem and asked for suggestions and ideas on how to approach it. Please get your facts straight. And I never said that proposal is the best thing ever, but I think it's a stap in a better direction. Obviously it can be improved. It wouldn't cause yo-yo-ing as much as you think. If the question deserved to be closed, a lot more votes would be casted to close it and it wouldn't y-y –  IVlad May 8 '10 at 17:21
    
@IVlad: That's your interpretation. What I see is a bunch of specs, possibly for a real project, possibly for homework/Project Euler/etc., and a line at the bottom saying "any suggestions." That's not a question. And you seem to be the only one who thinks that your proposal is a step in a "better direction." –  Aarobot May 8 '10 at 17:42
    
@Aarobot: so you interpret "Any suggestions or ideas of how this problem might be solved would be most welcomed." as "do my homework!"?. I am the only one thinking that? because I did get some upvotes on this post, and the question I linked to got 5 up / 5 down. Too bad the upvotes were probably given by the little people who can't vote to reopen yet. I haven't been been following meta and I don't plan to either way, so I'll just take your word for it however. –  IVlad May 8 '10 at 17:49
    
@Aarobot, actually the idea of each re-open vote cancelling a close vote has come up many times on meta, it's a feature-request with many votes, see this link: meta.stackexchange.com/questions/125/… –  Lance Roberts May 10 '10 at 16:53
    
@Lance: Again, it's not the same thing. It would still take 5 reopen votes to reopen a question after it is closed. That proposal is simply asking for the ability to cast reopen votes before the question is closed, to nullify an existing close vote. I actually agree with that idea, although I can understand why it's not implemented (who do you list as the closers when it took 50 back-and-forth votes?) –  Aarobot May 10 '10 at 21:55

This site is full of questions asked by people who get paid to know the things they are asking, yet their questions are considered fine. How is solving someone's homework problem worse?

solving real world problems might be intresting.

solving homework is not.

share|improve this answer
    
Just reading through this thread, and thought I'd mention how disappointing I found this sentiment. –  Patrick87 Aug 19 '11 at 5:36

I vote to close these types of questions, not because they are homework, but because they have no programming content. SO is a programming site, not an algorithms or maths one.

share|improve this answer
5  
You can't remove algorithms from programming in my opinion. Algorithms mean nothing without a programming language to implement them in and programming doesn't mean anything without algorithms. –  IVlad May 8 '10 at 13:44
4  
@IVlad The latter is true, the former not. Algorithms existed long before programming was thought of. –  nb69307 May 8 '10 at 13:50
3  
I'm of the opinion that algorithms can be separated from programming, but I'm eager to keep algorithm questions on Stack Overflow if they have even a hint of programming applicability. It's much easier to find examples of math questions that just don't have a thing to do with programming, so they should be removed. –  Bill the Lizard May 8 '10 at 13:52
    
@Neil Butterworth: if you agree that programming doesn't mean anything without algorithms, why do you close them? The implementation matters a lot as well. I've seen plenty of good algorithms implemented in ways that made them much slower. I maintain that you cannot separate the two, not in a computer science context. –  IVlad May 8 '10 at 14:01
1  
@IVlad If a question asks about newtons method, I would vote to close. If it asked about implementing newtons method in code, I would not. And SO isn't a computer science site either. –  nb69307 May 8 '10 at 14:05
2  
@Neil: where does it say it's not a computer science site? The FAQ says anything goes as long as it's of interest to other programmers. I think algorithms are of interest to other programmers, or they would have all failed their many algorithms classes in college. –  IVlad May 8 '10 at 14:22
2  
@IVlad Lots of things are interesting to programmers - what kind of monitor to use, how to schedule jobs using cron, or how to set up a consultancy business, to give just three examples. all would be off-topic on SO. And most programmers I know didn't do CS at college. –  nb69307 May 8 '10 at 14:26
    
@Neil: I think the FAQ refers to programming interests, though my main point is that either way, the FAQ should be much clearer about a number of things. Your monitor, politics, using cron, setting up a business, gardening and others are not things that will affect your programming abilities directly. CS however is something that can and will. And anyone who majors in CS does... CS, so maybe those people had different majors. In any case, I don't think they're not interested in it. –  IVlad May 8 '10 at 14:36
2  
@IVlad: If the FAQ page tried to resolve every single conceivable question or ambiguity then it would be 58 pages long. That's why we have this faq. Read it. Learn it. –  Aarobot May 8 '10 at 15:41
    
@Aarobot: like you said, this isn't a court of law. Nobody is going to split hairs, it's just some common sense ambiguities that need to be addressed. Also, nobody who registers on stackoverflow is even going to find that unless they're looking, let alone read it all. The essentials should be put in the FAQ page, and some kind of announcement should be made that invites people to discuss its contents. I barely even knew about meta until today. –  IVlad May 8 '10 at 17:30
    
@IVlad: You're right, people won't find it unless they go looking. I don't blame the guy who asked the question for not knowing right off the bat. Eventually, people learn how the system works if and when their questions get closed. That's OK. You, on the other hand, should have had the sense to look this up and see if your issue had already been discussed before writing up a long complaint. –  Aarobot May 8 '10 at 17:46
    
@Aarobot: the very definition of a complaint is an expression of a displeasure. Why should me looking it up or not influence me writing a complaint? You're not saying if others complained before I have no right to, are you? And this wasn't JUST a complaint. –  IVlad May 8 '10 at 18:03
1  
@IVlad: No, I'm saying that your specific complaint about your specific question was valid, because these issues are handled on a case-by-case basis, but the format of this question and much of its content are almost as bad as the question it links to, in the sense that it rambles on about a non-issue (one that is very easy to find information on!) instead of attempting to solve a specific problem. If not for the link, this question would/should be closed as an exact duplicate. –  Aarobot May 8 '10 at 19:19

I have a lot of respect for what you do Jeff but I have to agree with dbasnett on this one. I don't really care for this actual topic but more of what is going on on the SO sites in general.

I feel that because of this so called democracy the sites feel sterile and uninviting. I have often posted somewhere else because I felt that I would get voted down to hell.

I agree with you that noise is a major issue on any site but I have all to often seen down voting for they question style, composition and Construction. I have also seen instances of snobbery, much more on this site than any other I visit. None of this matters in terms of the question asked.

Fair enough you should not have to wade through a wall of text to find what is being asked but you could simply ignore or encourage the author. There are many reasons as to why a question may not be a good question voting down because of trivial issues is a disgrace, it kills community.

You know the problem could be quite simple, maybe democracy just doesn't work...

share|improve this answer

The one thing I dislike about all of the trilogy of sites is the high school clique-ishness feel they have. Why is the power to close a question so important? Once you read it and feel it is beneath you, don't read it anymore. Someone else may not share your opinion. Sure, some of the questions are hidden or not present. Don't you think being ignored or told to clarify would suffice. Do we really need to be voted off the island?

My mother, god-rest her soul, would tell you all that if you don't have something good to say, then don't say anything. This down-voting thing also reminds me of high school, mine is bigger than yours.

"I vote to close these types of questions, not because they are homework, but because they have no programming content. SO is a programming site, not an algorithms or maths one." This has to be one of the most humorous things I have read. Is it the tool or the process that is important? Would you rather have a great carpenter that used a hand saw or an average carpenter that used a circular saw? I see plenty of examples were fundamentals (algorithms) need to be stressed. I have received down votes for pasting VB Code in C questions, even though what I posted was as good or better than the "real" answers.

I don't care what your score is. All I care about are the words coming out of your mouth at a given instant in time.

If it were truly the intent of the creators to simplify and make these sites more friendly they would get rid of the ability to get voted off the island, and down voting. But high school is important when you are there, and I haven't been in some time.

BTW - please feel free to down vote any post I make on any of the sites. I don't care.

share|improve this answer
5  
This has been discussed a million times. If you allow discussing "anything" with no boundaries you won't have a community that's worth visiting. If you disagree so strongly, there might be other sites on the internet with more of an "anything goes!" ethos that you might enjoy spending time on more. Of course, I'd argue that the reason there AREN'T sites like this worth visiting is precisely because they get utterly destroyed by exactly what you're proposing. But hey, you wanted it, right .. :) –  Jeff Atwood May 9 '10 at 10:57
2  
also, "you read it and feel it is beneath you" is an .. odd .. way of stating this. I don't FEEL anything about questions. They are either on topic and useful, or off-topic and not useful. We have a democratic voting process where experienced users can vote on whether they agree or disagree on this, and IMHO that's as good as it gets when human beings are involved. –  Jeff Atwood May 9 '10 at 11:00
    
I didn't mean anything, as in the fate of the aardvark. I meant somewhere in the ballpark of a language / algorithm question. You posted a "question" on server fault about "How much network latency is “typical” for east - west coast USA?". If theoretical questions are allowed on your part (I am assuming that your question was about the traversal of many networks that you don't control, which makes it theoretical) why not every body else. I participate in many forums, and they all have rules. Most of them are not so harsh and cruel. –  dbasnett May 9 '10 at 12:09
    
I like reading what you write, some I agree with, some I don't. I know that these sites are your babies and that you are protective. I hope in time you will see that down voting / arbitrary removal of questions as counter productive. –  dbasnett May 9 '10 at 12:17
    
"What is the "typical" latency I should expect when moving network packets from the east coast <--> west coast of the USA?" WHY wasn't this closed for lack of specificity. What route? What protocol? What TOD? How much data? I resisted the urge to post this "This question is ambiguous, vague, incomplete, or rhetorical and cannot be reasonably answered in its current form." In a democracy you can tell the king he has no clothes. –  dbasnett May 9 '10 at 13:10

You must log in to answer this question.

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