# Why don't we use a homework policy… like physics.SE?

I was reading Proposal: ban verbatim homework questions which have no accompanying text and I came upon some startling old statistics.

Since 2013, the number of questions on this site has multiplied by more than 6 times.

Likewise, the number of answers has multiplied by more than 5 times.

We are clearly the second largest site on the network, with almost triple the number of new questions per day than the third leading site in Q's/day.

Reading over A Consolidated Homework Policy, it appears that there is no clear consensus on how the community wishes to handle homework-like questions. One of the main problems is we aren't very clear on what we mean by "homework-like", as shown in What is our policy on questions that are quite clearly homework?.

Going back to the first link, however, we may look upon some interesting examples of homework policies from other sites. Taken from physics:

They have a closing option for homework:

And the following policy is found from the help center:

"Do my homework"-type physics questions

"A 4kg ball is traveling at 8m/s in the x direction, how do I find..." Physics - Stack Exchange is not a homework help site. If you have a question about a homework problem, or any problem of an educational nature, narrow it down to the specific concept that is giving you trouble and ask about that. You can find more information about acceptable homework questions on our meta site.

Why don't we have a similar policy? Something along the lines of a copy/paste, changing the first line.

Likely, the first question that will be posed against this policy might be...

How do I know its a homework problem? (or any problem of an educational nature) - paraphrased from zyx

(I presume 'educational nature' may refer to questions involving self-taught topics)

We may take, once again, from Physics.SE:

A "homework question" is any question whose value lies in helping you understand the method by which the question can be solved, rather than getting the answer itself. This includes not just questions from actual homework assignments, but also self-study problems, puzzles, etc.

On the other hand, questions that come up in the course of doing a homework problem, but are separate from the main point of the problem, might not be considered homework questions. There's a bit of a judgment call to be made, depending on the context of the problem. If you're not sure, it's probably safer to treat your question as a homework question and later find out that it isn't, than the other way around.

I think that the first paragraph would not apply well here, but that the second paragraph would. I'd rather change it to:

A "homework question" is any question whose value lies in obtaining an answer to an isolated question and provides no significant value to other significantly substantial problems. For example, asking "An equilateral triangle has sides of length 8m, how do I find its area?" is considered a homework question because its only goal is to find an answer to a specific question that cannot be used to help solve other significantly substantial problems. On the other hand, asking "A triangle has some given side lengths a, b, c. How can I find its area in general in terms of a, b, and c?" is less likely to be a homework question since it asks a more general question which may be used to solve other related problems.

On the other hand, questions that come up in the course of doing a homework problem, but are separate from the main point of the problem, might not be considered homework questions. There's a bit of a judgment call to be made, depending on the context of the problem. If you're not sure, it's probably safer to treat your question as a homework question and later find out that it isn't, than the other way around.

Note that this policy also allows some leeway. There is no black and white, as the second paragraph points out, and certainly the first paragraph can't encompass every case. However, providing a clear and precise enough policy on what homework questions are and how we should deal with them is, in my opinion, a good first step.

Other such questions may be asked, but I certainly can't cover all of them.

Here's a lollipop for reading all of this, so please provide some constructive feedback on my thoughts.

• I am not sure what you are looking for. A question without any effort will very likely get closed in MSE. What's more do you want? – user99914 Oct 7 '17 at 3:17
• @JohnMa "A question without any effort will very likely get closed in MSE" Are you basing this on some statistics? – Did Oct 7 '17 at 4:22
• No, not at all. @Did – user99914 Oct 7 '17 at 4:22
• @JohnMa OK. FYI, my impression is exactly the opposite to yours (and I have no statistics either). – Did Oct 7 '17 at 4:24
• This time I agree with trump – user99914 Oct 7 '17 at 5:04
• @JohnMa After having spent time in CRUDE, I've seen many many questions that are older than a week that show no effort and have not been closed. Likewise, I've seen many questions that show no effort be reopened. I'm asking the community if such a policy would be accepted, and if not, why? – Simply Beautiful Art Oct 7 '17 at 11:30
• Making this an official policy would have numerous effects. Hopefully, users will be able to close questions as homework questions rather than closing them as "missing context", which most agree to be the wrong 'reason' to vote to close a homework question. Note that this is also relatively consistent with How to ask a good question, since homework questions that show effort will likely fall into "questions that come up in the course of doing a homework problem", and questions which include other context such as links are less likely to be isolated problems – Simply Beautiful Art Oct 7 '17 at 11:52
• I suspect it's a by-product of the fighting that went on when the sentiment that we needed a homework policy finally grew to the point that it could no longer be shouted down by those who were certain we should adopt no such thing. It was not a time conducive to a nuanced discussion about how to implement such a policy, since it would get hijacked by the arguments why we shouldn't have one. Also I think there was skew towards trying to offer a compromise to those who thought we shouldn't have a policy, such as the focus on trying to "rescue" bad posts. (e.g. the "show your work" craze) – Excluded and Offended Oct 7 '17 at 13:01
• I am the one to vote to close as too board, @SimplyBeautifulArt . You are probably aware than there are very extensive discussion in MSE. however, I do not see why you bring up this question again (The only reason you mention is that MSE has bigger popilation). Your question is basically the collection of all questions you linked, decorated as a board question "why don't we ... like physics SE?" To be honest, I don't even know where to start the discussion. – user99914 Oct 7 '17 at 14:03
• @Hurkyl well, as I'm not an old enough user to have participated in such discussions myself, but I do have to wonder: has the last conclusion benefited the site? Looking at the numbers, the amount of posts in the review queue recently has been larger than 20% of the new questions per day, and has exceeded 40% a few times. Not all of them are homework questions according to the definition I provide above, but a good majority are. We're the second largest site on stack exchange, and I feel we should do something about this. – Simply Beautiful Art Oct 7 '17 at 15:31
• @JohnMa as per why I'm bringing this up, the number of questions multiplied by 6 since the last major discussion. We've grown a lot, and so has the number of homework-like questions. I think there is greater motivation to have such a policy as opposed to in 2013. – Simply Beautiful Art Oct 7 '17 at 15:34
• But what do you propose to do specifically? Differently, what do you mean by an official policy? I do not think that adopting a new custom close reason and adding a few lines of text to the help center will change much anything. Especially as it is mostly redundant with "lack of context." To mention the word "homework" is in my opinion not a good idea as it adds an angle to it that creates emotions and conflict yet is ultimately a distraction. As you mention yourself not few of those questions in fact do get closed with the current reason. So what do you propose? – quid Oct 7 '17 at 16:05
• If this is the point "isolated and don't contribute much to the site" I'd agree that it is substantially different to what we have. I am not sure it is feasible to get a consensus what would qualify though, but that's a separate concern. – quid Oct 7 '17 at 16:57
• Perhaps after having conversed about it so long, that may be the issue I want to hit after all @quid – Simply Beautiful Art Oct 7 '17 at 17:01
• Well.. if you just look at here, then you will see that Shog had created a homework closure option on 25/6/13 but had to remove it on 18/11/13. I dont know the reason why it was removed but this was just to show you that such an option had earlier been considered and then thrown away. – SchrodingersCat Oct 9 '17 at 8:27

### Context and history

While we are making comparisons with Physics, let's observe that there is some dissatisfaction on the other side, too:

If you were just curious about why Physics and Math sites evolved in different ways, two of the reasons are

• The ratio of (teaching or learning) / (research) activities is greater in Mathematics than in Physics.
• The existence of MathOverflow means that those users who really do not want to coexist with homework questions can turn elsewhere. (Emilio Pisanty addresses this aspect.)

### Sorting the grains of sand

But perhaps there is more than idle curiousity, and you are proposing a change.

these questions are "isolated and don't contribute much to the site"

That is true for a vast majority of questions. A generic observation is Sturgeon's law. In the context of SE, it's Optimizing For Pearls, Not Sand. Most questions, homework or not, are sand.

If the problem is a large number of boring, repetitive homework questions, then closing is a wrong tool for the job. Trivial questions are almost immune to closure: it takes much less time for someone to type: "the area is half of base times height, so (formula)" than for five 3000-rep users to look at the question and vote to close it. The insistence on closing such question has a side effect: the answers that we manage to prevent are those that take time to conceive and write out. These correlate with questions that, with whatever flaws of presentation, have more mathematical content than others.

Off topic questions have to be cleared out of the way, but NOT via closure. By the time I have read the 1000th stars-and-bars question, I have already lost, and casting a close vote won't bring back my time. To avoid losing, one has to avoid reading them.

### If not closure, then what?

Stack Exchange is seemingly trying to help:

We're going to explore some ideas for better filtering of the questions that you see indicated by information that you give us. If you really only enjoy solving medium to difficult problems, that's what we need to prioritize showing you.

Yes, this was written (by a SE employee) about a year ago. But there's still hope that the company will take a little break from maximizing clicks on job ads on Stack Overflow and do something about the experience of expert users. For now, at least they gave us "interesting" sort on the homepage as a hand-me-down from SO.

On this site, I read only questions with nonnegative score, my favorite tags, no upvoted/accepted answer, and posted by users with $\ge 100$ reputation. (The last part is made possible by a userscript). When a boring question somehow gets through, I downvote it and it drops from the filter because of having negative score. It works for me better than custom close reasons or policy debates on meta.

• These are very interesting points. Thank you for sharing! – Simply Beautiful Art Oct 7 '17 at 17:54
• Some comments. Re: "immune to closure", your description is almost tautological -- questions don't get closed because people aren't closing the question. If you look at SO, for example, you'll find that the analogous questions are very rapidly downvoted and also deleted. – Excluded and Offended Oct 8 '17 at 0:19
• Re: "To avoid losing, one has to avoid reading them." That's why it's important to make this a community effort; you need a thousand other people to have already done something to those stars-and-bars questions so that you (and everybody else) don't get inundated with them. – Excluded and Offended Oct 8 '17 at 0:23
• Re: "The existence of MathOverflow means that those users who really do not want to coexist with homework questions can turn elsewhere." That's not really a solution at all. There is a huge amount of area in-between "contributing to mathematical research" and "doing people's homework problems for them", and most of it happens at mse, not at mo. – Excluded and Offended Oct 8 '17 at 0:25
• @Hurkyl There is indeed a lot of area there, but that doesn't mean Desire's point is irrelevant. Think of the contrapositive, if you will: the lack of an effective MO equivalent has put many more people looking for MO-level content in contact with no-effort homework, and PSE's policy is an attempt to resolve that tension. You can ignore MO as much as you want if you're just focused on MSE's internals, but if you want to compare with PSE then you're just fooling yourself if you think MO is not a factor. – E.P. Oct 10 '17 at 15:47
• (Another relevant comparison post, btw, is physics.meta.stackexchange.com/questions/9822) – E.P. Oct 10 '17 at 15:47
• Did I say that it's good to see you active here once more? It's good that you're back in action. – Asaf Karagila Oct 12 '17 at 11:22

## A note from the other side

This may not be evident from a single superficial look at Physics SE's homework-policy canonical meta post, but please do not come away with the impression that it is uncontroversial or that the stated policy fully represents the way it's implemented in practice.

More to the point, there is a widespread feeling in PSE that the homework policy needs reforming, and there is an ongoing conversation about how to reform it since... well, it feels like forever. An uncountable forever raised to an uncountable power. Like, hell freezes over then thaws then burns then cools then freezes over then thaws and so on, until the expansion and contraction from the thermal cycling causes metal fatigue on the foundations of hell and it all spills over into nothingness. More to the point, a conversation that's been so exhausting, and that's gone on for so long, that there's probably no more community energy left to actually achieve a consensus.

(For an unstructured look at that conversation, see e.g. the links at the beginning of this answer.)

Now, I've got no idea how homework really works in practice here, and copying over parts of the stated PSE policy may well be a good idea, for all I know. Just please don't take the stated policy as representative of the spirit of the in-practice policy that keeps the more-homeworkers and the ban-homeworkers from leaping at each other's throat. And don't ask us for a suitable summary of what the in-practice policy really is, either, 'cuz we've been fighting that one out over the past four years and we still don't have a meaningful answer.

Good luck with your conversation, though! :)

• Thank you very much for sharing. Judging from your description it seems to me that the two sites and their user bases are very much alike :-) – Jyrki Lahtonen Oct 12 '17 at 6:52
• @JyrkiLahtonen And yet the general overall attitudes towards PSQs seems markedly different (you know, statistics has ways of differentiating clouds of points even when they overlap...).. – Did Oct 18 '17 at 7:59
• For example, answers are routinely deleted by mods for "being complete answers to homework-and-exercises questions", see physics.meta.stackexchange.com/questions/10198 and the links therein. A policy soon to be implemented on math.se? :-) – Did Oct 18 '17 at 8:11
• @Did. That could be tempting. But for it to become a policy here we would need a higher level of acceptance (IMHO). – Jyrki Lahtonen Oct 18 '17 at 11:14
• @Jyrki What do you mean by "higher level of acceptance"? Since when is "acceptance" a criteria when it comes to vulgar profanity, or double, triple posting? Just because it happens doesn't mean it's therefore accepted. Standards need to be set, and consequences need to be enforced; and in that way, they become "accepted." but it sounds like your saying that unless a person wants to be nice, no one should enforce a policy that they be, in fact nice? – Namaste Jan 13 '18 at 22:22
• ... I.e., it sounds like you're sort of copping out, by saying as long as people do answer blatant homework PSQs with no effort, no one should enforce a policy, or consequences to users who enable such repetitious homework-completion service in the name of this site? – Namaste Jan 13 '18 at 22:23
• @amWhy Profanity et cetera are not contentious issues In other words, the policy to delete such material has a very high level of acceptance. Nearly universal I would say. But we are (according to my judgement) quite far from that level of consensus re homework. May be I'm timid to use my superpowers, but the longstanding practice has been that moderators don't lead the rush to the barricades. – Jyrki Lahtonen Jan 13 '18 at 22:55
• @Jyrki, That's fair. And I did not mean to aim my comment at you personally, nor moderators in general. But there are ways users can help counter the enabling that occurs by rep hunters who answer most everything, in those occasions when they are answering blatant homework-like PSQ's. I just like for both sides of how such questions will be received, and their consequences, to be presented when talking about "approaches to handling homework. Focusing only on, "seems nothing can be done" is only half the story. There are alternatives, and there are many users using them. That's all. – Namaste Jan 13 '18 at 23:21