# Categorizing homework questions with textbook information in a way that makes the site act like a solutions manual

I could imagine a system of categorizing the questions that would work alongside the current tagging system. If you select the "homework" tag (or some special tag or option), it would give you the option to specify which textbook problem your question pertains to in terms of title/chapter/section/problem number. Maybe the site could present a list of textbooks that have already had one solution entered and the user could navigate through a hierarchical tree of problems.

Now the users would be able to search or browse by textbook title, go to the chapter/section/problem and see if there is a solution present from someone else who has asked a question about it before.

This feature would make the site a lot more organized and eliminate redundant questions, or at least make it possible to find all the questions related to a specific problem in a textbook by providing explicit links. Math Stackexchange could eventually turn into the authoritative source of solutions for textbook problems. I think it would make the site a little less intimidating, too, if it was easier to use without having to ask questions all the time. Searching for mathematical symbols is pretty hard to do.

• This is a questions and answer site... It is weird that asking questions be what makes it hard to use, because using it is, more or less, asking questions! May 15 '12 at 0:50
• Well, it's a question and answer archive. May 15 '12 at 1:38
• It's not exactly what you were looking for, but searching for author -(tags) can yield good results. May 15 '12 at 2:16
• I'm extremely leery of this site becoming known as a place to read answers to homework problems. I understand that this is largely possibly now, but there is some (perhaps illusionary?) difference between allowing a thing and encouraging it. May 15 '12 at 3:30
• When someone posts a problem from the textbooks he is using, it is very probable that the same problem appears in 20 (or more) other textbooks, too. May 15 '12 at 6:28
• Another question that comes to mind - if we do something like this, why should it be restricted to homework questions only. I think that justification to include metadata with the information about the book would be the same for questions stemming from self-study; questions asking for clarification of a line in a proof of theorem from some book, etc. May 15 '12 at 7:42
• It's not that it should be restricted to homework questions, but homework questions already have a way of being indexed, so why not expand the software to use it? It's just an idea; I never suggested or implied that anything should be restricted. May 15 '12 at 13:26
• I think it would help make the site a place where people could go to be shown how to do math. If you're shown how to do it, then you eventually learn how to do it on your own. What benefit is there in students spending time searching for a teacher or another student to get help on something? That's a waste of time when the technology exists to speed up the process. As far as cheaters goes, who cares? Why should we live our lives around protecting people who are intent on harming themselves? May 15 '12 at 14:02
• Not to mention punishing or at the very least denying help to those who are honest. May 15 '12 at 14:09
• Dear Matt, Could you explain what you mean by, "those who are honest"? [I hope that my earlier comment didn't come off as overly critical.] Cheers, May 15 '12 at 15:29
• I'm in a rush, so apologies if this comes over as brusque, but I'd like to suggest (for others reading, not nec. the OP) that seeing answers is not sufficient for learning how to do something. Exercises are called exercises for a reason.
– user16299
May 15 '12 at 16:46
• On the other side, I don't think the licenses here prevent you from compiling an index of Math SE questions relating to a certain book or topic (I think there was a question about the latter on meta at some point) on a wiki somewhere. There are lots of free hosts you could use. May 15 '12 at 17:49
• By "those who are honest" I mean people who want to learn math as opposed to people who are just doing it for a grade. May 15 '12 at 17:56
• Cheating aside, I'm not quite sure I understand why the above comments indicate such negative feelings towards answers to textbook problems being contained on the website. Am I really the only mathematics student that has found it useful and instructive to look up solutions to textbook problems online? May 29 '12 at 4:45
• I'm not perfect. I can't always solve every problem I'm given, especially when there's a deadline and other classes to worry about. Realistically, if I don't look up the answer, I might never come back to the question. If I look up the answer, at the very least I now know how to solve the problem and have learned something in the process, albeit, maybe not as much as if I had solved the entire thing myself. Also, I'm just really damn curious and not knowing the answer would bug the hell out of me. May 29 '12 at 4:46

I upvoted because I don't see why we shouldn't allow copies of common problems (and index them by metadata). I think the critical issue here is "cheating", which nobody supports, but there is a lack of consensus as to what extent we should facilitate or tolerate it.

I am in favor of such an index, because it would be an aid to using the site, but I don't pretend to fully understand the concerns of academic users.

• First comes the issue that Martin points out, many books have similar problems. The second, and main, issue here is that it becomes a bold tool for "input book name and problem number, get a solution" sort of site. At least make them write a question, have it closed as duplicate, etc.
– Asaf Karagila Mod
May 15 '12 at 6:47
• Why should it be better for someone to write a question and have it closed as a duplicate than to simply find the answer they were looking for? I have read Dr. Strangedupe (and I agree with it substantially) but it doesn't seem to carry as much moral weight as "information wants to be free". Whatever to do about "cheating"? May 15 '12 at 7:18
• Oh, I am not at all a fan of closing duplicated questions (often it is easier just to answer them again). The point, however, is that we do not want to attract the crowd of "gimme solution, I don't really care about the math" kind of people. Yes, MSE should be open for everyone, but much like we don't like cranks and trolls we don't want to associate ourselves with those who do not want to study. We should not produce tools which are mainly for them, because in my view - such index is exactly the tool for people who do not wish to learn mathematics and want to get their homework done.
– Asaf Karagila Mod
May 15 '12 at 7:20
• It's also exactly the tool for those who do wish to learn mathematics. Getting their homework done is a side-effect. At least that is my perspective as someone who has not been to school in a long time. I do not often work from textbooks but perhaps I would if better online resources were available. May 15 '12 at 7:22
• I understand that trolling is wrong, but what exactly do we have against being a little cranky? Also, since we don't want people asking questions who are not interested in studying, isn't having the answer readily available a good way to dissuade them from asking? I imagine that most people who ask duplicates ask without having seen the duplicate; if they can find it in an index, then we save the effort of having to close (although Dr. Strangedupe is always hungry for more brains). May 15 '12 at 8:27
• I didn't say "a little cranky" I meant people which verge on trolling, e.g. those which post questions like "It is impossible to prove Godel's theorem because ..." and such. As for the second, yes it is a good way to prevent them from asking, but it is enough to attract them to the site. Personally, I would be happy if they would know MSE as a place where people ask and answer questions. Not as a place to get solutions to problems. Just for the sake of this kind of reputation a solution-index must not be created.
– Asaf Karagila Mod
May 15 '12 at 10:17
• I can't agree without invoking normative academic standards (which I am admittedly unfamiliar with and feel ambiguously towards). What else would be the conflict between a "place to ask and answer questions" and a "place to get [or propose] solutions to problems", other than the possibility of "cheating"? Good SEO (as well as good user discovery) is one thing but I am saying that the C word is a more important issue (although I haven't proposed any policies to resolve it). May 15 '12 at 10:33

This might be closer to a longer comment than to an answer - but I hope it's ok to post it here anyway.

I have been more than once in the situation that I stumbled upon a problem when reading a mathematical book. Usually I thought that I am either missing an important point in some proof or that the authors have there an omission or a mistake. I would love to be able to find out whether some reader of that book was in a similar situation before. Hence it would be very nice to have somewhere something as list of discussions concerning the various books organized by the book. (The questions/discussions would be probably very similar to the questions appearing here at MSE and sometimes at MO: Why this part of proof holds? Could you explain this step in the proof? Is this a mistake in this book?)

Related MO thread: Errata database?

What I write above is to some extent similar to OP's suggestion, although he proposes to restrict this just to homework questions.

Having said that I find idea of something like this very tempting, I do not think that MSE is the best place for creating and maintaining this type of database. Nevertheless, I believe that MSE can serve the same purpose even without additional metadata and sorting questions according to books from which their originated. It is perhaps a slightly more difficult to find the question related to the particular book. Adding metadata would made this easier, but I think amount of work which would go into it is not proportional to the profit it would bring the users.

• Please consider that whatever work that goes into such a feature pays dividends forever (by which I mean as long as MSE or its archives remain online). As mathematics is eternal, there is no simple net-present-value calculation. You say that MSE can already serve the same purpose; but is that purpose really being served as well as it would be with the proposed feature? May 15 '12 at 9:06
• @DanBrumleve: $\sum_{i=0}^\infty \delta^i W=W/(1-\delta)$, so one can certainly calculate net-present-values for eternal projects. May 15 '12 at 11:07
• @Michael, point taken, and I hope so will be mine. :) May 15 '12 at 11:28

I am curious to know if the anti-homework bloc that has downvoted this question objects to askers of exercises stating the book, problem number, and complete original wording of the problem.

Search engines would then locate the problem and solution for anyone who wants it. MSE's page ranking will put the question/answer at the top of the results.

• As for me, I consider it bad form to post a question from a book without giving a reference that can be traced to the page where the exercise appears. This is a minimum courtesy to anybody interested in the material, and it allows correction of mistakes in case the problem was misunderstood or copied incorrectly.
– zyx
Mar 10 '13 at 20:28
• How should I interpret up/down votes of this answer? (Down meaning "we object to citing exercises from a book," or down meaning "we don't object to the above"?) Mar 10 '13 at 21:18
• I was hoping for comments from the pro/anti HW blocs. There are 3 votes for my comment and 3 against the question (no upvotes), so if those are expressing the same opinion, the question will become invisible to the extent that it is popular. Maybe better if I repost later as a question. (@anorton)
– zyx
Mar 10 '13 at 21:24
• I am from the pro-HW bloc and the "solutions are bad answers"-bloc (and I confess I haven't seen an anti-HW bloc). I have just now seen the original proposal, and I downvoted due to the suggestion that MSE should be a solution repository, rather than containing answers that explain solutions, or better yet, explain and demonstrate how to go about solving a problem.
– user14972
Apr 20 '13 at 12:43
• (and I have downvoted this answer due to a reasonably strong belief that you are using a phrase "anti-homework bloc" in a derogatory fashion to refer to people who are very much pro-homework, but disagree with you on a specific point regarding homework)
– user14972
Apr 20 '13 at 12:46
• @Hurkyl, call it the anti-CPHQ bloc, the anti-homework-cheating, the anti-cheating-on-homework-or-exams bloc, or whatever other combination of the words you think best expresses the position. It has been clear for a long time that there is a left and a right wing on all the issues related to homework, exams, question closings, and a few other matters (with L and R affiliation being consistent across issues). When I wrote "pro-HW bloc" to include those who think that answering CPHQ is ethically virtuous and fine on MSE, is that also derogatory in your view? I'm in that set, if it matters.
– zyx
Apr 20 '13 at 18:45