Should we rename the homework policy?

Question

The homework policy is a constant source of confusion for new (and sometimes established) users. We see this confusion, for example, when users respond to closures based on the homework policy by defending their post with "This is not a homework problem", or similar. Some users have even been conditioned to claim a non-homework basis for their posts preemptively. It seems the word "homework" throws people off.

Should we modify the help center, associated close reason, and our own language to make clear that whether or not a question comes from homework isn't the real issue?

This post is very similar to a previous one. However, I am not only asking about the tag. I'm asking about the help center, close reason, and language used by users as well.

Answer 1

Yes, we should.

"Homework" isn't the issue

I think we all agree that whether or not a post comes from a homework assignment is irrelevant, as illustrated by this example question:

Consider a transmon qubit with capacitance C and Josephson junction critical current I_c. What is the matrix element <1|Q|0> for this system?

I might be motivated to ask this question in my research job. It's not homework, but it is also not a question we would like to support. We prefer if I would try the calculation myself, and then ask a specific conceptual question about whatever part has me stuck.

The help center's entry on homework-like questions gets that part right when it says

If you have a [homework-like question], narrow it down to the specific concept that is giving you trouble and ask about that.

Unfortunately, while the help center correctly tells the user what to do, it does not correctly tell the user when to do it. In the help center, what I transcribed as [homework-like question] in the previous quote is actually written as

question about a homework problem, or any problem of an educational nature

This, incorrectly, tells the user that only questions which come from homework assignments or other educational origins are subject to the requirement that they ask something specific and conceptual. This is obviously not what we intend.

The accepted and much up-voted answer to the canonical meta post on this subject also defines homework-like in a (different!) confusing way

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.

This classification fails to reject the example question posed earlier in this post. It also omits the idea that posts should have a conceptual core. The notion that questions should be conceptual is mentioned later in the answer, but readers only get to that after they've already been mislead once by the help center and then again by the first few paragraphs of the answer. Also, the emphasis on the question's "value" as a pedagogical tool is subjective, thus offering further confusion.

Of all the resources related to the homework policy, the close reason is the only one which unambiguously gets at the part that matters. It is nice that it refers to "homework-like" questions to emphasize the fact that pedagogical settings have nothing to do with the appropriateness of a post.

Get rid of the word "homework" (mostly)

We should just clear state that we only support questions with a conceptual core. If a post just asks how to solve any specific problem, it's off topic. Problem solving questions should, as the close reason says, ask about a specific physics concept and show some effort. Surely, questions derived from homework will tend to fail this criterion more than other questions [a], and that should be noted in the help center and associated close reason, but it should not be the fundamental criterion itself.

Here is a list of the specific (small!) changes I think we should make (I write "calculation request" as a stand-in label for what we currently call the "homework" policy, but I'm sure we can think of a better name):

• Modify the help center to say

  Calculation requests, e.g. "A 4kg ball is traveling at 8m/s in the x direction, how do I find...", or requests to have your work checked, e.g. "Here's my calculation, but the answer is wrong, where did I mess up?"

  Physics - Stack Exchange is not a calculation service or a work checking service. If you need help with a particular calculation, narrow it down to the specific concept that is giving you trouble and ask about that. Make sure your question cannot be answered with a small amount of research and show that you've made effort to resolve the issue yourself.

• Modify the close reason to say

  Questions about particular calculation, whether a request for the calculation itself or a request to have your work checked, must ask about a specific physics concept and show some effort to work through the problem. We want our questions to be useful to the broader community, and to future users. See our meta site for more guidance on how to edit your question to make it better

• We can probably leave the tag as it is, since it says "homework-like". Changing it to "problem solving" would be better but may not be necessary. I do not know how hard it is to change a tag.

• (Harder) Stop referring to this whole thing as the "homework policy" and call it the "problem solving policy".

Some objections and rebuttals

• "We have to call it the homework policy, otherwise new users won't understand."

Saying that problem-solving questions must center on a specific concept is no less clear than referring to homework. In fact, as argued above, I think "problem-solving" is considerably less vague and subjective.

• "We'll be flooded with homework questions from users who don't know the rules."

If folks are really afraid we can do a few week trial period. Stack Exchange gives us awesome SQL tools so that we can measure the effects objectively.

• "This isn't worth the change."

Judging by discussions in comments and the chat room, it seems the homework policy is confusing to lots of users largely because people don't understand what qualifies as "homework". The fact that "homework" has an obvious and established meaning which conflicts with how we use it is a clear problem. As I've shown, a result of this is that the official documentation disagrees with itself on the definition.

[a] This is probably more because people who have finished school have learned how to ask good questions and use online communities, rather than anything having to do with whether or not the physics problem at hand came from homework.

Answer 2

I have the same objection to this that dmckee had last time we talked about this.

Essentially, everybody who posts a question on this site wants a problem solved. The problem may be lack of understanding, or it may be homework, or it may be just about anything else. But the very existence of a question implies there is a problem to be solved, and such a tag would then apply to everything under the sun.

I would rather see the meta tag go away entirely and we downvote bad "homework-like" questions into oblivion and close them just as fast as we are now. There's no need for a tag to point out garbage.

The "good" questions (ie. the ones that already conform to the policy) don't need a separate tag to tell us that they conform to the policy. They will be well-posed questions looking for conceptual explanations and thus no different from any other good question.

I don't think the close reason needs to say "Homework like ..." nor "Problem solving ...", it could simply say:

Questions should ask about a specific physics concept and show some effort to work through the problem. We want our questions to be useful to the broader community, and to future users. See our meta site for more guidance on how to edit your question to make it better

which then means that close reason can apply to all of the really bad questions that come in and make no sense nor effort to justify them. It's kind of a "non-mainstream" close reason for mainstream topics. We want all of our questions to be about specific physics concepts -- that's part of what makes us unique from World Building.SE. We close questions that start "Imagine if the sun instantly disappeared..." as non-physical, so this kind of close reason would work for that as well as "A ball is rolling down a 45 degree ramp..."

Regarding the help center, I would say it doesn't need changed because we do want to forbid homework questions as they are defined there. And we want to forbid bad questions as defined elsewhere in the help center.

Answer 3

Back in the good old days there was no option to close as homework. Instead we had a close reason described as too localised.

The definition of too localised was that it was of interest to too few people i.e. it did not fit with the long term aim of the site to become a repository of useful information about physics. Homework questions often (but not always) fall into this category because the answer is of interest only to the person doing the homework. I much preferred the too localised option to homework because it allowed a lot more flexibility. For example it allowed me to answer a homework question if by doing so I could explore an interesting and useful concept.

I wasn't privy to the reasons why the too localised option was removed, but I would guess that it was viewed as too vague. You could argue that an answer to a blatant homework question taken from Gravitation would be useful to more people than a answer about some complex bit of string theory. While I can see this, I don't see it as a problem. My view is that too localised means what the >3000 rep users think it means and I don't see that it would cause any more arguments that the current homework option does.

So I believe we should can the homework option and bring back too localised instead. I think this is basically equivalent to Daniel's suggestion, and I don't really care what exact words are used in the close reason as long as the spirit is the same.

Response to ACuriousMind's comment:

ACuriousMind says in a comment:

It probably was removed because people like you used it as a flexible close reason

and I freely concede this is probably absolutely correct. My argument is that this is a good thing not a bad thing and it should be encouraged.

We now have lots of >3k rep users, so lots of people can vote to close or reopen. I don't see why we need any legalistically precise prescriptions for when a question can or can't be closed. If I vote to close a question as too localised then any of the many >3k rep users can vote to keep the question open or vote to reopen it. In the end the questions that are closed will reflect the views of the majority (of >3k rep users) as to what this site should look like.

Answer 4

I'll prefix this with "I'm not an active physics.SE participant", I'm over on another P.SE - programmers. And we too have a problem with homework questions in that the vast majority of the time, they aren't good questions.

To this, end, I wrote a post titled Open letter to students with homework problems trying to explain what the problem with just copying and pasting homework problems into the text area is, and that even if we do answer the question, it may lead to problems for the student's future learning endeavors (they learn a short cut that works 90% of the time, but completely fail to learn the long way of doing it needed to understand why it fails in the 10%, or other topics). This post has since become the highest voted post on Programmers.SE's meta site.

The key thing to all this is not so much "its homework, its bad" but rather that the student needs to learn how to ask good questions. This, unfortunately, isn't often a skill taught by many institutions and learned the hard way. In the context of StackExchange, this is often learned through down votes and closed questions - which, while a necessary and critical tool for StackExchange is a less than ideal one for pedagogy.

Ideally, the focus of a homework policy (named that or anything else) should be on:

• Reinforcing good question quality on Stack Exchange
• Setting the expectation of what can (and can't) be answered and timeframes (how often does one see the "urgent - need answer by 5pm" in questions?)
• Note the possible problems that a student who asks a question here may encounter (inadvertently skipping around in the syllabus without proper comprehension of previous material - its in that order for a reason).
• Encourage that someone needs to discuss a topic, that chat may be a better resource than a main site question and its comments

There's also a quote that I'm fond of from Programmers.SE, slightly tweaked to Physics:

Picture Physics like a room with a whiteboard. No computer, no books, no papers. You write your question on the whiteboard and experts walk by. Someone sees one that they can answer, they stop in and help.

Answer 5

OP's meta question is a much-needed initiative to improve/change/update the homework policy wordings of the help center, the meta question 714, the wiki discription, and close reason; to make them useful and consistent, and stress that the issue is not whether the question is actual homework, but rather whether it asks a conceptional question.

Interested policy makers and editors in these matters are encouraged to come forward and meet in the Phys.SE chat room.

Similarly, about a year ago (October 22nd, 2014), the homework tag was already changed after several and long discussions on meta and various chat sessions to the self-explanatory homework-and-exercise tag, which has for the past year mostly run smoothly and as intended. Any further change in the tag name seems to warrant a separate meta post, which could possibly be conducted after an overhaul of the above mentioned policy texts.

Finally, a reminder to not forget that that the main objective is to first-and-foremost serve the Phys.SE community of active researchers, academics and students of physics and astronomy, as it says in the intro pages. In plain English: The interests of people asking off-topic questions might not coincide with the best interests of the Phys.SE community.

Answer 6

Rather than using the term "homework-like questions" or "problem-solving questions", or the "too localized" concept, a better descriptor for our objections might be "drill-oriented or standard instructional question" or "standard drill problem."

Many of the questions that have come through in the past weeks are stock physics drill questions designed to give students practice in learning a concept or technique. These would be questions which have been asked and answered over and over for years. Many of the OPs say something like "I don't know where to begin," or "I'm lost." My first mental response is "Get off the computer and go ask your instructor this question."

If we already have a catalogue/collection of drill-problem type answered questions we could have a standard link to give to questioners. "This is a standard drill problem. Go to this link and find your related concept."

In my short association with this site, I have concluded, happily, that Physics.SE doesn't want to be the first-stop-shop of physics students seeking a drill-problem answer (like Answers.com [gag!], but a compendium of good physics thinking for conceptual questions encountered along the way.

OTOH, if a questioner gives some statement about their thoughts on a drill problem which indicates a poor concept, we could give them a nudge in the right direction, and I've noticed this happening in comments ( Wouldn't those nudges be more suited for a brief Answer?)

I think it is fair to tell questioners "This problem/question is a standard drill problem that has been asked and answered many times. Try to get some local assistance."

Answer 7

Some time ago I proposed banning homework questions altogether, which was a reasonably popular proposal. While I don't quite have the energy to keep pushing for this, I still support it. In the context of your question, I think it would be a good and workable solution. To be clear, the solution would look something like this:

1. The "homework policy" would be replaced with a new document, which states that homework questions (in a strict sense) are completely banned. No ifs, no buts: your question will be closed if it looks like you're trying to get help with your homework, even if you actually aren't. The only way you can get help with your homework is by asking a conceptual question.

2. We also have a new policy along the lines of what you suggest, which can be adjusted to taste independently of the related but distinct goal of making sure we don't supply answers to school/undergrad homework questions.

Overall I feel that this would be less confusing for the users. The current policy of "you can ask homework questions if you ask about a concept" is pretty confusing to the uninitiated and means that a judgement call has to be made every time a homework question appears in the review queue. "No homework, or anything that looks like it" is much clearer for both askers and reviewers.

The "it has to be about a concept" policy would then require a judgement call only in the minority of cases where it doesn't look like a homework problem, but is nevertheless judged not to be conceptual enough for the site.

In other words, homework is an issue on this site. It is an issue not because homework questions tend to be non-conceptual but because they are homework questions. Answering them here is bad for the students' education and bad for the quality of this site, so let's just have a clear and straightforward policy of not doing so. Non-conceptual problems are also an issue, but they are a separate one, and better addressed using a separate policy.