7
$\begingroup$

Presently, it may well seem to users who are not yet familiar with MathOverflow that flooding the site with questions is desirable and encouraged behavior -- at least this is what the point system appears to suggest:

  • an upvote on a question adds more than twice as many points as a downvote takes away, which can make even questions with negative score yield points,

  • one can get an unlimited number of points for questions, also if one doesn't answer any question of someone else, and

  • neither questions being closed as off-topic nor posts deleted for reasons of moderation nor flags "cost" any points (except for spam flags, but these are not applicable).

Do people see a problem here which should be addressed in some way?

Added later: Some related data from the StackExchange Data Explorer.

$\endgroup$
  • $\begingroup$ Stefan, thank you for the edit (I'm removing my comments as no longer relevant). $\endgroup$ – Michal R. Przybylek Dec 6 '13 at 15:05
  • 1
    $\begingroup$ My apologies; I hadn't seen the comment activity before posting my answer (which in any case makes oblique and hopefully tame reference to an earlier version of the question). $\endgroup$ – Todd Trimble Dec 6 '13 at 16:04
  • 2
    $\begingroup$ For what it's worth, I ask a ton of questions on MO, and I really don't care about the point system. I ask questions because I want to talk about math, and I want to hear what the experts think on certain subjects, or if certain subjects are well studied, among other reasons. I'm pretty sure nobody is asking questions to get more points. Usually they just want answers. $\endgroup$ – Jonathan Beardsley Dec 18 '13 at 21:21
  • 2
    $\begingroup$ @JonBeardsley: I'd say, 58 questions in 3 years is rather not a big ton of questions ... . $\endgroup$ – Stefan Kohl Dec 18 '13 at 21:31
24
$\begingroup$

I don't see a problem with the system. It seems to me that any abuse of the system is by far the exception, not the rule.

We need to be as encouraging as possible of the asking of good questions. A case has been made (see here for example) that many people, for example graduate students or assistant professors, find the atmosphere at MathOverflow intimidating as it is, and thus withhold what might be nice questions. Some people even feel that awarding 5 points per upvote isn't enough (it used to be 10 points in MO.1).

On the other side, it has occasionally been recommended (without a view to enforcement) that people try first to get a sense of the culture by answering a few questions before asking a question themselves. This came up in discussion here: http://mathoverflow.tqft.net/discussion/1438/mochizuki-proof-of-abc/. (I think this is too high a barrier to set as a policy, but the sentiment that people should look around a little first to get a sense of the place is, I think, a sound one.)

Overall, it's pretty hard to get away with asking many bad questions (bad for MO, I mean). There are not a few users (including the OP, Stefan Kohl) who invest a good deal of time maintaining quality control. The occasional controversies that erupt are not currently, in my opinion, cause for serious alarm (in fact, they usually have a silver lining, in that those fine mathematicians who line up on one or the other side of the controversy show they really care about fostering the health of MO).

I don't think the questions of any one user has quite approached flood level. We do see now and then some (arguably) overly energetic people who eventually become a nuisance to others; moderation here can be tricky and calls for patience while the evidence accumulates. I am sure that the official moderators move too slowly for some people's taste and at the same time too quickly for others, but that's the way it sometimes goes!

$\endgroup$
  • $\begingroup$ Thanks! -- Since on a first glance your second paragraph and my question may be seen to represent opposite opinions, I'd just like to point out that this is actually not the case. -- In particular I think that within reason one can well consider rewarding good questions more than now. -- Just rather not unlimited, in particular if someone asks only questions. $\endgroup$ – Stefan Kohl Dec 6 '13 at 18:37
  • $\begingroup$ Stefan: If you have a concrete proposal along these lines, please feel free to share it (the MO.3 discussion that I implicitly linked to -- see the second link -- might be a good place). $\endgroup$ – Todd Trimble Dec 6 '13 at 19:02
  • 6
    $\begingroup$ Todd: There are really many possibilities, and I don't really have a favorite proposal. -- I think it should just be anything along the lines of: "Think of MO as a person who greatly values being asked a good question from time to time, but who gets annoyed by not well-thought questions and who sometimes also has a question him-/herself and is deeply grateful to you for a good answer then. -- Don't think of MO as an answering machine." $\endgroup$ – Stefan Kohl Dec 6 '13 at 22:49
17
$\begingroup$

By and large, the users who have created problems by flooding the site with questions have created equally many problems by also flooding the site with comments. This suggests that their behavior is driven by something other than the point system.

$\endgroup$
  • $\begingroup$ I think this is a good point. -- Though it is hard to guess people's motivation, or what would change if the 'rules' would be different in this or that respect. $\endgroup$ – Stefan Kohl Dec 6 '13 at 17:36
  • 2
    $\begingroup$ The two users that in my opinion were the worst two as regards "flooding the site with questions" (that were IMO questionable but not universally perceived as bad, which I agree with OP of q is worse than clearly off-topic things) did not do that much beyond that. And, I am sure they were sure they contributed something positive. (I do not want to name the two; not those that got mentioned in the thread; you, in the sense of OP of answer, likely can guess one of them if you recall some discussion about one of my 'favorite' questions). $\endgroup$ – user9072 Dec 7 '13 at 13:32
  • 5
    $\begingroup$ To clarify what I mean let me add that I agree that there are certainly reasons other than the point-system why some might post too much content of some form, yet I still think that fact that one gets point plays an important role. Even if the behavior is not (initially) motivated by points, I still think that it is rewarded by points justifies the behavior for the users. These things were even made explict sometimes. Somebody complains to some such user about too much posting, and they reply something like: what are you complaining about obviously I contribute something positive. $\endgroup$ – user9072 Dec 7 '13 at 13:43
15
$\begingroup$

On my opinion, the site is flooded with bad questions (homework, trivial questions, poorly edited/illiterate questions etc.) People who ask them usually do not earn many points. On the other hand, people with high reputation usually ask good questions, so the reputation system fulfills its purpose to some extent.

On the other hand, so-called "soft questions" usually bring much more points than serious mathematical questions, for the simple reason that more people are likely to read a soft question, because everyone understands it. But I think we can live with this; earning points is not the main purpose of our activity after all.

$\endgroup$
  • 11
    $\begingroup$ More to the point though, I don't think those questions are being encouraged by the point system. They are encouraged by the hope of getting quick answers. (I also think "flood" is hyperbole, but that's a separate question.) $\endgroup$ – Todd Trimble Dec 8 '13 at 4:11
13
$\begingroup$

From my own experience, the worst questions on the site are usually homework problems, and these sure are not being asked for the sake of reputation.

I sympathize with rep points for asking questions. These reinforce the concept that questions themselves are content and can have quality (or not). I feel that without this concept, people would write their questions in a more throw-away fashion (this can be witnessed on the AoPS forums: questions often lack LaTeX and fail basic grammar, on occasion also forgetting to define notation or state important assumptions; no matter how well-written the answer is, the readability of the thread is seriously impaired).

$\endgroup$
  • 5
    $\begingroup$ Sure -- the worst questions as such are usually homework problems. -- But such questions are usually quickly closed / removed, and I cannot remember a case where a student has posted really a lot of them. Also, homework questions don't bring anyone in the position to take much influence on the scope of the site. As long as this is so, I think they are hardly more than a little nuisance ... . $\endgroup$ – Stefan Kohl Dec 7 '13 at 11:45
  • 10
    $\begingroup$ @StefanKohl yes for the moment, but I think it is a good idea for MathOverflow to warily watch these developments. For example on Physics SE, which had a decent but not exclusively research level once, the number of popular, very basic, and bad homework questions ramped up after the last election a year ago. This brought too many people, who are not knowledgeable about advanced topics but SE politically very active, into positions to determine the scope of the site (> 3000 rep). It is them who are dominating for example the close/reopen queues at present, which is not good to say the least ... $\endgroup$ – Dilaton Dec 7 '13 at 12:41

You must log in to answer this question.

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