Why do we not delete answers to closed questions which smell like homework?

Why do we not do this? Could we consider making the answers invisible to everyone but moderators once it received, say, 3 votes for 'off topic due to lack of context'?

Also at the moment, there are nothing discouraging people answering 'homework looking questions with no attempts'.

Here is an example of what I meant:

https://math.stackexchange.com/questions/613323/solve-an-equation-3-frac-1x-1-2/613326#613326

• Please be aware that some professors encourage their students to collaborate on homework, including posting it here, e.g. see this famous, highly upvoted answer by JDH. I confess that I am shocked by some of the recent suggestions on meta, which propose to strongly inhibit the spirit of teaching and sharing mathematical knowledge. – Bill Dubuque Feb 27 '14 at 1:17
• @BillDubuque agreed, but i think this site's policy is generally against such questions with no attempts? – Lost1 Feb 27 '14 at 1:19
• Have you never been completely lost, completely stumped when attempting to solve a problem? Should you be denied help in that case? You would not get very far in your mathematical studies were such a strange policy uniformly enforced by everyone in the mathematical community. – Bill Dubuque Feb 27 '14 at 2:09
• Indeed, suppose that all of your math ebooks refused to display their proofs until you "show some effort". How would you feel about that? Do you think you could learn mathematics efficiently that way? You'd essentially have to reprove every mathematical result from scratch. It would take many, many lifetimes to learn mathematics that way. – Bill Dubuque Feb 27 '14 at 2:14
• @BillDubuque in which case some hints/clarification are fine, but still, imo, not an answer. In the example i posted, i would think something like 'if you log both side, what do you get' is helpful,but the full answers as given there, imo, are not. – Lost1 Feb 27 '14 at 2:23
• @Bill: I never asked my teachers to help do a problem before attempting it, nor when stumped have I ever tried to simply drop my homework on my teacher's desk to await explanation, so I'm not sure such a "strange policy" would have affected me in any way. Those who are actually completely lost, IMO, would benefit much more from actual interactive help rather than the MSE format anyways. – user14972 Feb 27 '14 at 3:00
• @Lost1: As a practical matter, I doubt there is enough political will to attempt a reform of the community opinions on answers. But to be fair, I was quite surprised there was enough will to push through the reform of the community opinions on questions to get to the current situation where we close such questions. – user14972 Feb 27 '14 at 3:04
• Some seem unhappy with the notion of a student learning math from others at all, preferring that he must derive the entire discipline from the ground up, pulling the result from the void with raw brain magic. In practice, this doesn't appear to happen very often. – Alexander Gruber Feb 27 '14 at 3:10
• @BillDubuque I think they call that the Moore method. :) – Alexander Gruber Feb 27 '14 at 3:16
• Here is a related question from meta.stackoverflow that you may be interested in. – user14972 Feb 27 '14 at 3:24
• @Alex Rumor has it that the new Moore ebooks are almost to market. Using cutting-edge biotech, they have sensors that will not allow a reader to turn the page until sufficient blood, sweat and tears have been detected struggling on the current page. I wonder how well they will sell... – Bill Dubuque Feb 27 '14 at 3:44
• "Could we consider making the answers invisible to everyone but moderators once it received, say, 3 votes for 'off topic due to lack of context'?" But then the moderators could still cheat on their homework. "Also at the moment, there are nothing discouraging people answering 'homework looking questions with no attempts'." Well, there is something --- you can always leave a comment on the answer, suggesting there's a community consensus against answering such questions. I have done that. – Gerry Myerson Feb 27 '14 at 6:20
• @BillDubuque Please be aware that some professors encourage their students to collaborate on homework, including posting it here This "some" is almost certainly a very small number. I would venture to say that the large majority of teachers encouraging collaborative work intended students to exchange their thoughts and work on the problem. I doubt they ever intended to encourage PSQ-style posting, even the ones that encouraged internet posting. – rschwieb Feb 27 '14 at 13:36
• @Gerry Where did you get the strange idea that there is a community consensus on homework matters? This has always been one of the most contentious issues in the community. – Bill Dubuque Feb 27 '14 at 14:01
• @Bill, I didn't say that there is a community consensus --- I said OP could suggest that there is a community consensus. In any event, while you are certainly correct in writing that homework is a most contentious issue, my impression of the discussions on meta is that those who condone posting the type of answer that can be copy-pasted directly into a student's paper and handed in as the student's own work are greatly outnumbered by those who don't. – Gerry Myerson Feb 27 '14 at 21:49

Deleting or downvoting answers to questions which "smell like homework" seems to me generally counterproductive.

• The OP has likely already seen the answer, so you aren't preventing them from cheating, in the case that it is homework.

• The answerer will likely be annoyed at the downvotes, and may be discouraged from providing answers in the future. (I once downvoted a few answers for giving full solutions to questions which I felt should be given only hints. As you might expect, the users who posted these answers were annoyed.)

• The number of votes on the answer will not reflect whether it is a good answer or not.

• Perhaps most importantly, deleting or downvoting answers allows a user to cover their tracks, which is exactly what we don't want. It is an issue that users delete their questions once they are answered, and the best solution to this is upvoting the good answers, because a question with upvoted answers can't be deleted. Deleting and downvoting answers to these questions works directly against the goal.

Many users agree that lazy homework questions should be closed (put on hold) and discouraged. But actively working to close or downvote the answers is the wrong approach. This will at best annoy the answerer and at worst hide the tracks of a cheater.

• I would upvote were it not for the rude paragraph that begins "most of us...". Please keep in mind that many teachers, including most of the best I've known, do encourage collaboration on homework. So asking for help here need neither be "cheating" nor "lazy". – Bill Dubuque Feb 27 '14 at 13:50
• @BillDubuque I am absolutely aware of that and was merely stating what I have perceived to be the majority opinion on the matter. As of now, the linked post has 58 upvotes and 8 downvotes. – 6005 Feb 27 '14 at 14:09
• And an opposite opinion by a professor has 62 upvotes and 8 downvotes. But those meaningless numbers are not what matters. Rather, what matters is that these issues are contentious, have always been so, and will always be so. Forcing one viewpoint on the entire community will alienate some of the most valued members of the community. Alas, this has already happened. MSE has already lost some of its most prolific, knowledgeable members due to inability to reach good compromises on contentious matters. Let's not lose more. – Bill Dubuque Feb 27 '14 at 14:18
• If the opinion is from a professor, it must be valid. Also, that post is from 2011 and the general sentiment has changed quite a bit since then with the rising numbers of lazy homework questions. Nevertheless, I take your point. I've changed it from "most of us" to "many users". – 6005 Feb 27 '14 at 14:28
• @BillDubuque I think .MSE has also lost prolific members due to the drop in the level of the quality of the questions? I personally think this is compounded by the number of duplicate homework smelling questions drowning good questions. Do you think my belief is wrong? – Lost1 Feb 27 '14 at 17:07
• @Lost1 Dealing with the increasing volume of questions is better dealt with by lobbying SE to provide better tools to filter questions. This will also allow users to easily ignore questions that they don't like for whatever reason. The volume will only keep increasing, so that will have to be addressed eventually. – Bill Dubuque Feb 27 '14 at 17:13
• @Goos fyi, I mentioned that the author was a professor not as any sort of appeal to authority (or experience) but, rather, to help dispel the common misconception that teachers always discourage collaboration on homework. – Bill Dubuque Feb 27 '14 at 18:15
• @BillDubuque Please stop trying to frame the issue as "pro-collaboration" vs. "anti-collaboration". Or, if you must, then you might also accept that most people here are looking for more collaboration between asker and answerer (in the form of an exchange of thoughts/ideas, rather than the simple giving of an answer) rather than less. – Scott H. Feb 27 '14 at 19:12
• @ScottH. That was never my intent. Rather, my point is to make better known the many diverse views in the community, so that hopefully better compromises can be reached, so that the community does not lose further significant contributors. – Bill Dubuque Feb 27 '14 at 19:18

The obvious thing to do would be to downvote answers at sight, without a comment. If done often enough, this is already a deterrent to answering.

Then there are a few options:

1. 20K rep users can vote to delete negatively scored answers. (Which is another reason why downvotes are helpful). However, the number of 20K+ users is not huge, and not many of them will want to do this.
2. 10K+ rep users can vote to delete the question (which of course deletes any answers to it).
3. 125+ rep users can downvote both question and answer enough for automatic deletion to happen. Since the question has at least one answer, auto-deletion requires several things, in addition to question being closed (full details here):

• Question score $\le 0$
• No answers with a score $> 0$