15
$\begingroup$

In reading the endless discussions of the acceptability of postings by particular users (e.g., are six postings per day too much for one person), or by users in particular situations (e.g., should people be "allowed" to post questions only if they have not been assigned those questions as schoolwork), I find the following slogan helpful:

-- "Math.SE is not a social networking site."

That is, the focused Q & A, low-discussion framework of SE means that although people are users of the system, structurally the site is not set up to directly create relationships between users, but only between questions and answers. For example, the software does not provide a facility for users to send each other messages.

Corollaries of this principle include:

  1. imagining the questioner as an AI is a good habit for reducing the incentive to flamewars.

  2. whether a posting is homework (the answer recipient will be paid in grades), job-work (the recipient will be paid in money), or research work (the recipient will be paid in status), or even math.SE-work (the recipient is trolling for reputation points) doesn't matter. Certainly a questioner can try to make it more attractive to others to spend time answering, by explaining why he wants an answer, what he has done, relevant references and sources, etc. But if the asker does not provide this, it seems to me to be out of bounds to interrogate aggressively for such information or to close a question. The correct solution to the "problem" of homework or other seemingly exploitative postings is to improve the rating and filtering mechanism for questions in ways that make it irrelevant to determine whether anything is homework or not.

Though the above is less a question than an unsolicited answer, and math.SE is not a discussion forum, the beta-test version of meta.math.SE certainly is a discussion forum as it concerns questions of policy and FAQ content, so I assume it is OK to post it here.

$\endgroup$
  • 1
    $\begingroup$ @T: You post is certainly acceptable, these kinds of posts are made on meta all the time $\endgroup$ – Casebash Aug 14 '10 at 7:13
  • $\begingroup$ kind of related: meta.math.stackexchange.com/questions/588/… $\endgroup$ – Grigory M Aug 14 '10 at 9:08
  • 5
    $\begingroup$ +1 for "not a social networking site". This is a very important to keep in mind. Another +1 for "imagining the questioner as an AI is a good habit for reducing the incentive to flamewars". (Here I can't actually vote; I am an unregistered user). $\endgroup$ – user1119 Aug 14 '10 at 10:08
  • 2
    $\begingroup$ I completely agree with everything in your question except this: "Certainly a questioner can try to make it more attractive to others to spend time answering, by explaining why he wants an answer, what he has done, relevant references and sources, etc. But if the asker does not provide this, it seems to me to be out of bounds to interrogate aggressively for such information or to close a question." We (at least I) may "interrogate" or vote to close such questions not because it's asked by a human, but because incomplete questions are simply poor questions. Nothing to do with a person asking. $\endgroup$ – ShreevatsaR Aug 14 '10 at 16:38
  • 1
    $\begingroup$ I should edit to make it clearer, but there is a world of difference between asking for information about the question (references, any known partial solution or special case, "motivation" in the mathematical sense of related subject matter) and information about the person posting: is it homework, why do they want an answer, what work have they done so far, etc. Even snide opinions seem fine as long as they discuss only the question, such as "I would be more inclined to spend time on this given more background, partial progress, or the problem source". $\endgroup$ – T.. Aug 14 '10 at 16:53
  • 2
    $\begingroup$ @T..: I still don't understand what's wrong with expecting every question to contain motivation as in why they want an answer, what work they have done so far, etc. At any rate, this is not a corollary of your "not a social networking site" statement (with which I agree); imagining that every question is posted by a different user would give the same outcome. "Not a social networking site" does not imply questions posted by a AI, nor do we want questions posed by AIs (or textbooks, problem sets). We want "real" questions. $\endgroup$ – ShreevatsaR Aug 15 '10 at 6:37
8
$\begingroup$

I largely agree, except that, from my position as a professional teacher, I'm not opposed to an atmosphere that discourages blatant homework cheating. On the other hand, I see T..'s point that interrogations on this point, as on any other, have the possibility of igniting flame wars.

My hope is that, through polite commenting and appropriate (i.e. non-accusatory/non-speculative) use of the homework tag, a polite atmosphere encouraging honesty about homework can be created. At the same time, people should err on the side of civility rather than judgmentalism, otherwise flame wars may well erupt. I'm open to the suggestion that I'm being too idealistic; if convinced, I would fall into even stronger agreement with T..'s position.

$\endgroup$
  • 3
    $\begingroup$ Since the meta site here is based on the stackexchange software rather than a traditional discussion forum, flame wars are less likely by design itself. Moreover good and helpful posts would get voted up and rude and nasty posts will get voted down. Upvotes are possible even for comments. Thus there will be immediate feedback. For all these reasons I like the software backend a lot. Still you do have a point; humans are humans and we better cultivate a pleasant atmosphere. $\endgroup$ – user1119 Aug 14 '10 at 10:11
  • $\begingroup$ @George S.: Because the meta site here is an SE site, it's cumbersome and annoying to use... $\endgroup$ – 97832123 Aug 14 '10 at 14:36
  • 1
    $\begingroup$ @97832123 : Here we can even follow up our comments by looking at "responses", which is great. If what is bothering you is the lack of capability for discussion, maybe the stackoverflow chat announcement will warm your heart: blog.stackoverflow.com/2010/08/chat-now-in-public-beta ... I believe every stackexchange site will obtain it eventually. $\endgroup$ – user1119 Aug 14 '10 at 15:39
  • $\begingroup$ @George: Actually, math.se already has a chat room, it was simply used once and nobody ever came back. $\endgroup$ – Larry Wang Aug 14 '10 at 15:44
  • $\begingroup$ @Kaestur Hakarl: I think the chat room will take off after there is enough traffic on the site. Right now there are simply not enough active users. $\endgroup$ – user1119 Aug 14 '10 at 15:56
2
$\begingroup$

I agree with Matt that we should be less accusatory and speculative about questions being homework. We only need to ask about whether a question is homework when it really looks like it. It is certainly not irrelevant if a question is homework - then we know to just gives hints. Often we won't be able to tell if the question is homework - then we will just have to trust the asker. For other specific questions that are unmotivated, we should simply ask about motivation instead and what they have tried.

With regard to questions focusing on individuals, those are also general questions as well. The general questions are valid, even if the question focused a bit too much on individuals.

$\endgroup$
1
$\begingroup$

The correct solution to the "problem" of homework or other seemingly exploitative postings is to improve the rating and filtering mechanism for questions in ways that make it irrelevant to determine whether anything is homework or not.

So, the idea here is to ignore the problem (not to close such questions etc) and wait for better rating system?

Well, probably, [waiting for] the elixir of life is the correct way to heal any disease... just not too practical, one might say.


Nevertheless, I agree, that it's

out of bounds to interrogate aggressively for such (whether it's a homework, what's the motivation for asking etc —G.M.) information.

The obvious suggestion is

  • if the question is bad — just downvote it (see numerous discussions of what questions are good/bad elsewhere; I certainly don't want to start a new one here);

  • if "it's difficult to tell what is being asked here; this question is ambiguous, vague, incomplete, or rhetorical and cannot be reasonably answered in its current form" — it's a valid reason to close, one don't have to "interrogate" the OP for missing details;

  • if it's not clear, whether the question is homework, it's OP's problem — we should just rate the question as it's stated (or close as "too localized", if it is); but if giving complete answer doesn't feel right — just post some useful hint.

$\endgroup$
  • 2
    $\begingroup$ It is possible to ask if a question is homework without "interrogating". Maybe sometimes it'd be better to assume it is homework and only give a partial answer or hint $\endgroup$ – Casebash Aug 14 '10 at 10:30
  • $\begingroup$ @Casebash agreed; incorporated this idea in the answer $\endgroup$ – Grigory M Aug 14 '10 at 10:33
  • $\begingroup$ "ignore the problem ... wait for elixir of life" is not the proposed algorithm. The phrase "improve rating and filtering mechanisms" includes things that can be done right now such as adding a [task] tag to objectively mark one of the undesirable features of many homework (and non-homework) postings. People who don't like [task] (or [unsourced], or [5000+ char], or other identifiable types of question) can filter them out if they are tagged. Rating users by answered-to-asked ratios can be done now from the monthly data dumps. No elixirs or interrogations are required. $\endgroup$ – T.. Dec 20 '10 at 6:34

You must log in to answer this question.

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