12
$\begingroup$

Very recently the following "question" was posted on the main site

The purpose of this "question" is to create a solutions manual for a freely available set of notes in algebraic geometry. In particular, the OP requests that each "answer" consist of the statement and solution of one exercise from these notes.

In my opinion, at the very least such a question should be closed since it is not a question. At another level, such a question may be construed as a way to avoid rate limits associated with asking questions. All in all, to me this feels like a misuse (or possibly abuse) of the math.SE system, and a grand departure from its purpose so far.

There have been previous meta-questions asked about posting and categorizing multiple questions from single sources:

but the question under consideration appears to be a different type of animal that I cannot recall seeing before.

So I ask the following:

Should we allow individual questions which ask for solutions to multiple unspecified problems from an individual source for the purpose of creating at least a partial solutions manual?

$\endgroup$
  • 2
    $\begingroup$ I think this should be closed. $\endgroup$ – Asaf Karagila Dec 8 '13 at 16:05
  • 4
    $\begingroup$ @Asaf Your bald statement of opinion is of very limited interest. I would be interested to hear your explanation of why you hold this opinion. $\endgroup$ – MJD Dec 9 '13 at 14:49
  • 2
    $\begingroup$ @MJD: Wasn't this topic discussed before? I don't think the site exists in order to construct solution manuals. If someone notices that there are plenty of questions from a particular book that appeared on the site, and wishes to compile an external list of links for questions and answers, that's great. But those aren't really an intrinsic part of the site's goal. Moreover in a single question to try and accumulate a solution manual for an entire book, or even a chapter for that matter, seems preposterously outrageous and by definition "Too broad". $\endgroup$ – Asaf Karagila Dec 9 '13 at 15:16
  • 1
    $\begingroup$ I agree, the format of the site is not well-suited for writing a solution manual. The solutions would be disorganized and hard to search, and displayed in a random order to the readers. The post which contained the solution manual would constantly be bumped to the front of the active list, displacing more deserving and urgent questions. So the project would not effectively achieve its aim, and it would disrupt the site's regular workings while doing so. $\endgroup$ – MJD Dec 9 '13 at 15:50
  • 1
    $\begingroup$ It should be closed. It is violating the spirit of this site. $\endgroup$ – copper.hat Dec 10 '13 at 21:09
  • 1
    $\begingroup$ Arthur, the question and its title do not match. Do you object to change the title to something like "requesting unlimited number of answers to one question about a book, so as to create a solution manual"? This is very different from the listed topic, "using MSE to create solution manuals", which would mean doing so one question at a time in standard SE style. There is likely to be a wide spectrum of opinion on the general idea in the title and an extremely narrow, basically unanimous set of opinions about the actual question you posed, and that PNB answered. $\endgroup$ – zyx Dec 12 '13 at 1:53
  • $\begingroup$ @zyx: I've attempted to improve the title, and I think it matches the question much better now. If you still feel it is lacking, feel free to improve it further. $\endgroup$ – user642796 Dec 12 '13 at 6:08
  • $\begingroup$ Perfect, and much better than my version would have been. Thanks. @ArthurFischer $\endgroup$ – zyx Dec 12 '13 at 6:19
22
$\begingroup$

This question can be understood as

  1. Not a question, but a call for contributions to a project. Should be closed as "not a real question" (no longer a standard reason, but can be a custom off-topic reason).
  2. Is a many-in-one question: how do I solve $n$th problem in the book, for $n$ from $1$ to $N$? Should be closed as overly broad.

A solution manual is a book. If a question is such that answering it requires writing a book, that question is not appropriate for any SE site. SE is not a book publishing platform, and attempts to use it as such never led to anything good. A long-term book-writing project is not something we want bumped to front page with every update, displacing actual questions waiting for an answer.

There are nice publishing/crowdsourcing platforms available, such as MediaWiki. ExWiki is built precisely for the purpose stated in the question, and its author should be directed there after the question is closed.

The aim of ExWiki is to collect exercises from various fields of mathematics and computer science. Each exercise is equipped with hints, solutions and respective references to books in which it appears as an exercise. The goal of ExWiki is to serve as a database of exercises and also as a solution manual.

Duplicating ExWiki here is as pointless as duplicating Wikipedia.

$\endgroup$
  • $\begingroup$ The HUGE problem with custom off-topic reason is that they only appear as a comment, and not in the actual closing reason. I'm not even sure if the comment survives the closing process, or gets deleted like the case of a duplicate suggestion. $\endgroup$ – Asaf Karagila Dec 8 '13 at 16:05
  • 8
    $\begingroup$ ExWiki has all its content posted by one person at an average rate of one solution every 3 days (70 solutions in the last 7 months from page caches, or check exwiki.org/mw/… ). The list of exercises was last updated in May 2012 ( exwiki.org/mw/index.php?title=Category:Exercise ). Technically it is extremely limited compared to SE. Are you seriously suggesting that MSE is "duplicating effort" of that web site or that anyone wanting to discuss solutions to a book exercise should prefer to go there instead of posting here? $\endgroup$ – zyx Dec 8 '13 at 20:12
  • $\begingroup$ I have noticed that ProofWiki quite often has listed several sources where the particular results, see for example the three sources at the end of this article. But I do not know whether they include only results proven in those book, or whether it is intended for exercises, too. $\endgroup$ – Martin Sleziak Dec 13 '13 at 7:30
  • $\begingroup$ @MartinSleziak, yes, ProofWiki is another good thing for this. It will go through some of the exercises in a text and skip others. Unfortunately, there are barriers to participation on that site that I will not discuss in detail here. $\endgroup$ – dfeuer Dec 13 '13 at 20:33

You must log in to answer this question.

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