When, if ever, is it appropriate to add the tag to a question that has not been acknowledged by the OP to be homework?

My position is that it is never appropriate, full stop. It’s fine to ask whether a question is homework and suggest the tag; after all, a newcomer may well not be aware that it exists. But to add it without asking is to claim the ability to read the OP’s mind, and that seems to me highly improper, even when it’s perfectly obvious that the question is very likely to be homework. (And I have been fooled a few times.) I have routinely rejected proposed tag edits of this type for just this reason and not thought twice about it. However, I recently got into an editing war with another experienced user who sees adding the tag to probable homework as being on all fours with adding a subject tag and wanted the homework filter to catch that question, since it very likely is homework. I see a very significant difference: I know whether a question is , say, but without a statement by the OP I cannot be certain that a question is homework.

I didn’t see it at the time, but not quite a year ago the same question was asked in a slightly different context, and the only answer said in part:

I and many other users consider it inappropriate to tag another user's question as homework unless they have explicitly said it is homework.

It received $10$ upvotes and one downvote. This answer to a much older question is in the same vein; it has $26$ upvotes and two downvotes.

I’m now curious as to whether sentiment has changed. I’ve no real reason to think so, but the discussion got sufficiently heated that it seemed a good idea to open it up here.

(Oh, and feel free to add or change tags: I don’t follow Meta closely enough to be very familiar with the tags.)

  • 5
    $\begingroup$ You can use [tag:tag-name] to embed the tag link and whatnot. $\endgroup$ – Asaf Karagila Mod Feb 28 '13 at 8:45
  • 1
    $\begingroup$ @Asaf: Thanks! I didn’t know that one. $\endgroup$ – Brian M. Scott Feb 28 '13 at 8:48
  • 3
    $\begingroup$ Well now you know. "And knowing is half the battle!" $\endgroup$ – Asaf Karagila Mod Feb 28 '13 at 8:50
  • $\begingroup$ @Asaf: :-)${}{}$ $\endgroup$ – Brian M. Scott Feb 28 '13 at 8:51
  • 11
    $\begingroup$ There is one situation in which I would add the "homework" tag to a question that has not been acknowledged by the OP to be homework: when OP is a student in a class I'm teaching, and I recognize the problem as one I have assigned. $\endgroup$ – Gerry Myerson Feb 28 '13 at 12:06
  • $\begingroup$ A related question that hadn't yet been linked here: meta.math.stackexchange.com/questions/3198 $\endgroup$ – Jonas Meyer Feb 28 '13 at 16:43
  • 1
    $\begingroup$ @GerryMyerson In a case like that, you should also include a comment saying that you are the professor who assigned the homework question. Otherwise nobody knows that that is the case. $\endgroup$ – Joe Z. Mar 5 '13 at 15:14
  • 2
    $\begingroup$ @Joe, that might have the effect of publicly embarrassing the student, which I might not want to do. $\endgroup$ – Gerry Myerson Mar 5 '13 at 23:35

It seemed relevant to look up other sites' definition of the homework tag: see below. The practice differs: in Physics and Chemistry, the tag is not restricted to actual assignments, and Chemistry (apparently) applies it more liberally than others. Mathematica and Statistics explicitly ask for this tag to not be used other than by the OP.

  • Physics: Applies to questions of primarily educational value - not only questions that arise from actual homework assignments, but any question where it is preferable to guide the asker to the answer rather than giving it away outright.
  • Statistics: A routine question from a textbook, course, or test used for a class or self-study. This community's policy is to "provide helpful hints." [ ... ] "Don't edit a question to add the homework tag. If there's any room for doubt at all, it's best to leave it as is. Instead, add a comment first requesting that the asker clarify the situation."
  • Mathematica: Homework questions are welcome, as long as they are asked honestly, explain the problem, and show sufficient effort. Please do not use this as the only tag for a question. Please do not add this tag to questions by other people unless they explicitly say that their question is part of their homework.
  • Chemistry: If your question is part of your homework, please use this tag. It signifies that instead of a direct answer, care should be taken to explaining the steps to the solution, preferably sticking to just giving helpful, conceptual hints. This tag may be applied to questions which are not homework but are similar to a homework problem.
  • Biology: This tag indicates that the question targets some basic concept or information that might seem trivial to professionals and can be seen as a home assignment in biology-related university courses. Please see the homework policy on meta before asking homework questions.
  • Computational Science: The purpose of homework exercises is to solidify understanding and develop skills. It is in this spirit that homework designated questions will/should be answered.
  • StackOverflow: This tag is OBSOLETE and is in the process of being removed. Please do NOT add this tag to questions. But don't remove it without looking at the question to see if it needs cleanup.

My opinion is that if a tag is a tag, it should be used as a tag: to organize the data. Conversely: if a tag cannot be used as a tag, then it is not a tag. The way is used makes it akin to a message from the OP that is entered in the "tags" field. It does not belong there. In my humble opinion, tag should be deleted deprecated/blacklisted, as it is on StackOverflow.

In fact, I find myself in complete agreement with the StackOverflow tag wiki:

In cases where users do not wish to be given the entire solution, they should state this fact within the question body itself, or phrase the question so that it is restricted to only those points which require clarification.

Users who feel the need to know if a question is homework or not should simply judge by the content of the question and act accordingly. There is no need to spend time debating whether or not the question is homework, or to what extent the question should be answered.

If a question is of low quality, such as those directly copied from a workbook, simply vote to close the question, downvote if appropriate, and move on.

  • $\begingroup$ But you do take this "message from OP" into account by putting it on ignore, right? $\endgroup$ – Michael Greinecker Mod Feb 28 '13 at 17:41
  • $\begingroup$ @MichaelGreinecker Yes, I adapt to the way in which the tag is used at present. This does not mean it's used properly. // As the site continues to grow, hiding by tags will become less efficient. You can only get 50 questions per page from the site, and if 45 of them are in your ignored tags, only 5 remain on the page. At some point one has to give up browsing and use various search methods to find, rather than exclude, questions with which to engage. StackOverflow passed this point already, and they no longer care for the homework tag. Math.SE is moving in that direction too. $\endgroup$ – user53153 Feb 28 '13 at 17:53
  • 2
    $\begingroup$ While the quote you've given seems to be the "official" word on the matter on the Statistics site, my experience is that high-rep users routinely add the homework tag and this is, at least tacitly, supported, if not encouraged, by the mods there. The tag is also used much more generally there, i.e., for any question that could potentially be perceived as homework. Being a smaller and somewhat more tight-knit site, the treatment of homework in answers is also much more uniform there. $\endgroup$ – cardinal Mar 1 '13 at 2:00
  • $\begingroup$ @cardinal Thanks for giving an insider's perspective. $\endgroup$ – user53153 Mar 1 '13 at 3:08
  • 1
    $\begingroup$ [[it is preferable to guide the asker to the answer rather than giving it away outright]] Any experience on whether this really works? Or do users "give it away" despite the tag (as we see here)? $\endgroup$ – GEdgar Mar 1 '13 at 18:38

I thought, but I guess I'm completely wrong, that you add the homework tag if you only want hints, instead of full answers. At least, that is the way I use it, and it works very well in that way.

Many question I ask with the homework tag are not homework at all, but I like the way users treat those homework questions at MSE.

I would not find it a bet idea to change the "homework" tag, in something like "hints-only" tag. If the OP want to cheat the school system, the OP will find a way anyway. If the OP wants to really learn something from his time MSE, the OP can add the "hints-only" tag.

  • 7
    $\begingroup$ The homework tag can’t be used reliably as a hints-only tag: some users quite routinely give full solutions even if the homework tag is present. On occasion I do myself, if I can’t think of any useful hint, or if I think that the OP is sufficiently lost to benefit more from a full solution this time. $\endgroup$ – Brian M. Scott Mar 1 '13 at 16:25
  • $\begingroup$ Maybe it's just me, but that's not how I interpret the hw tag at all. I find it simply satisfying to see OP be willing to have full disclosure. As to hints-only, there has been at least one such suggestion since o started perusing this site. Maybe you cam find it and see how the community reacted (it's a pain on my phone). $\endgroup$ – gnometorule Mar 1 '13 at 16:29
  • 7
    $\begingroup$ Here's an outlandish idea: if the OP wants to receive only hints and not a full solution, they should say so. $\endgroup$ – user53153 Mar 1 '13 at 17:27
  • 1
    $\begingroup$ @5pm, I have seen problems where OP did ask for hints only, and got complete answers anyway. There's no foolproof way to get what you want on m.se. $\endgroup$ – Gerry Myerson Mar 5 '13 at 23:38
  • $\begingroup$ @GerryMyerson Very true, any statement by OP may be ignored or unnoticed. But I guess that "I need only a hint, not a solution" would have a better chance of having the desired effect than any tag. (Especially not a tag the interpretation of which is still subject to debate.) $\endgroup$ – user53153 Mar 5 '13 at 23:49

I think that is broken as it is. The main problem with it is that we have no objective source of information we could use to retag a question besides the OP own words. Of course, humans as we are, some just don't know about it, and others want to plainly cheat. I guess we should decide on

Who takes the responsibility for giving away answers to problems that possibly will be graded later?

Two extreme options are:

  • If people want to cheat, let them do as they like, it is their responsibility.
  • The community assumes the responsibility and takes extra care to avoid being cheated.

Either way, the does not help. The first case suggest that the tag should be named, for example, (as suggested by Kasper) and be applied by the OP only. The second extreme implies name like or and should be applied by anyone who feels like it.

Some ideas:

  • I guess that would be useful, while the should be deprecated as in SO. The system should warn users that "The OP explicitly asked for hints only." in some bar like the "10 new answer" or "you've earned XYZ badge" notifications.
  • I think that it should be the system who decides whether the question is a possible homework or not (based on wording or question containing only a picture, etc.) and displays a bar saying "This question is homework-like, please do this and refrain from doing that". It would apply for all users with reputation $< n$, and users with reputation $> m$ could remove the "possible-homework" flag (this should not be a tag) if the OP described the context or for some other reason.

My personal opinion is that the purpose of this site is to answer questions, not raise kids; I would prefer to give the OPs the benefit of doubt. However, I do not trust users with reputation $1$, and huge volume of low-quality posts (they do not even care to search...) deteriorates my experience with math.SE. The conclusion is, that I believe that there is some trust level above which we could let the OPs decide for themselves, while below we hammer-down strict policies.

What do you think?

  • 1
    $\begingroup$ I think these suggestions are unlikely to be implemented, as they require changes to the code that can only be performed by the StackExchange admins and not by any members of the community. Also, they amount to formalizing a policy of what should or should not be done for homework-like questions, and there has been substantial opposition to doing so in the past: see the answers to the proposed consolidated homework policy. $\endgroup$ – user856 Mar 11 '13 at 11:11
  • 1
    $\begingroup$ Just doing away with the homework tag sounds like a good idea to me though, and I'm not aware of any arguments against it. $\endgroup$ – user856 Mar 11 '13 at 11:16
  • $\begingroup$ @RahulNarain It does require changes to the code, however, this is not a problem limited to math.SE only. With some serious proposal and backing from more than one *.SE sites maybe we could get enough leverage to make it done :-) $\endgroup$ – dtldarek Mar 11 '13 at 11:21

There seem to be a number of users who simply do not want to deal with homework-like questions, as exemplified for example in Qiaochu's answer. If the impetus of these people is not sabotaging the educational value of homework-assignments, there is little we can do (apart from radical changes in MSE policy). But if users simply do not like to have to see these kinds of questions, we can simply introduce a new tag that marks such questions. For example, we could have an tag for those kind of questions that usually appear as textbook problems, on comprehension exams etc. This seems to be rather harmless and adding such a tag is less accusatory.

Of course, this smells a bit of meta-tagging, but many users apparently read as a meta-tag. While the number of such users may be small, they include extremely active users with rare skills.

Edit: I turned this into community wiki because I'm not decided on the issue myself.


It might help for me to describe more thoroughly where I'm coming from.

Questions are being posted to math.SE much more quickly than they used to be, and average question quality also subjectively seems to be decreasing. As a result, math.SE is, at least for me, less interesting and more overwhelming to browse unmodified than it used to be. Accordingly, I have started ignoring certain tags, one of which is the tag, to improve my own browsing experience, and I suggest that other users do the same (not necessarily including the tag).

Both users who have favorited a given tag and users who are ignoring a given tag benefit from questions being properly tagged, and as filtering the ever-growing stream of questions becomes a more pressing need, proper tagging becomes more important for proper filtering. This is as true of the tag as it is of any other tag.

If I or someone else tags a question which is probably homework as homework and it turns out we're wrong, then the tag can easily be removed by the OP. I don't see this as a big deal. I'm not accusing anyone of anything. I just think we should recognize that math.SE has gotten bigger and adapt accordingly.

  • 13
    $\begingroup$ The homework tag is a big deal. Many users change the way they answer questions if it's homework. Some ignore the tag altogether (I started doing so sometime ago). Some only give hints. Some even downvote full answers to homework questions. User behavior around this tag is different from all other tags, and this is why I'm with Brian on this; I don't think we should add the tag ourselves. $\endgroup$ – Ayman Hourieh Feb 28 '13 at 13:26
  • $\begingroup$ @Ayman: what is a probability $p$ such that if I estimate the probability of a question being homework as being at least $p$, you think it would be reasonable for me to tag it as homework? $\endgroup$ – Qiaochu Yuan Feb 28 '13 at 18:07
  • 3
    $\begingroup$ The problem is, there is no way to objectively assess the probability of a question being homework. We need a general guideline that applies to everyone regardless of their experience with homework, teaching, etc. I also think that the potential damage of mistakenly applying the tag is greater than letting an uninteresting question slip through. $\endgroup$ – Ayman Hourieh Feb 28 '13 at 19:29
  • $\begingroup$ @Ayman: sure, in the same way that there is no way to objectively assess the probability of a question being algebraic topology or complex analysis, but humans seem pretty good at doing this in practice anyway. How much greater is the potential damage of mistakenly applying the homework tag? Is it great enough that you would reject $p = 0.95$ (nineteen times greater)? What about $p = 0.99$ (ninety-nine times greater)? $\endgroup$ – Qiaochu Yuan Feb 28 '13 at 20:13
  • 5
    $\begingroup$ I disagree. The exact same question can be asked in homework and non-homework contexts (self-study, contest, exam prep, etc), and there is no way to tell unless one gets more information from the OP. For complex-analysis on the other hand, the question itself is usually enough to make a judgement. To decide whether a question falls under a certain mathematical topic, one can refer to books and classification schemes. This is not true for homework. As for probabilities, not everyone is good at assessing homework probability; we're looking for a general guideline here. $\endgroup$ – Ayman Hourieh Feb 28 '13 at 20:53
  • 3
    $\begingroup$ @Ayman: Thank you for your comments; however, I would disagree somewhat with the assertion that it's not easy to tell between actual homework questions and self-study. I would posit that very often it's quite easy to tell the difference based on style and initial effort on the part of the OP alone. :-) $\endgroup$ – cardinal Mar 1 '13 at 1:53
  • 2
    $\begingroup$ @cardinal Self-study questions are usually more motivated than homework, I agree. But self-study can still generate routine questions that many are trying to ignore by tagging with homework. :) $\endgroup$ – Ayman Hourieh Mar 1 '13 at 15:07
  • $\begingroup$ Dear @Ayman: I am actually in favor of a more liberal definition of "homework", similar to how it is used on stats.SE. My point was only that differentiating between an actual homework question and a self-study question is often not as difficult as I thought your comment seemed to imply. I'd like a more uniform treatment of homework-like questions limited to pedagogical hints and suggestions rather than outright answers, but I know that's a pipe dream (on this particular site). $\endgroup$ – cardinal Mar 2 '13 at 1:01
  • $\begingroup$ @cardinal Understood. Thanks for your comments. $\endgroup$ – Ayman Hourieh Mar 2 '13 at 1:16

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