Not signed in (Sign In)

Vanilla 1.1.9 is a product of Lussumo. More Information: Documentation, Community Support.

    • CommentAuthorAgCl_
    • CommentTimeAug 21st 2010 edited
     
    Being a low-rep and not very active user, I don't know if I should comment on matters like these, but I believe that the policy of closing questions should be consistent and well defined. That's because it is an act against the efforts of the questioner, the answerers and possibly the to-be answerers, who are thinking about how to write down a good answer, while the question is being closed. I don't say questions should not be closed when needed, but claim that closing should be well-justified, and reasons should not be subjective-and-argumentative with respect to a majority of the reasonable users.

    There reason I'm writing this now is because I think two questions by Kelsey L are closed by totally wrong reasons:

    1) http://mathoverflow.net/questions/36224/what-jobs-are-there-in-pure-math-besides-being-a-professor-closed
    *closed as not a real question*

    My comment: This is a real question. There are many examples of this sort of questions. It should be closed as "exact duplicate"

    2) http://mathoverflow.net/questions/36256/noise-filtering-closed
    *closed as spam*

    My comment: This is not spam. The other acts of the user do not justify closing a question as spam. It could be closed as off-topic, or too localized.

    Again, I don't say this particular user is not a troll. I just say that, a consistent and well-defined policy will not upset the more respected questioners in the future if their questions get closed. This way, it will be taken less personally. Maybe a catalog of typical good examples could be listed under each of the reasons like off-topic, too localized, etc. This could be included in the faq, or it could pop up when a user presses the 'close' button.

    In summary, closing questions should not be arbitrary or based on personal tastes, and this message should be conveyed to the users wherever possible.
    • CommentAuthorHarry Gindi
    • CommentTimeAug 21st 2010 edited
     

    -1 for not reading the thread on meta.MO titled Kelsey L. MO has no obligation to be polite to obvious trolls, and no reason to try to close questions in good faith that were clearly asked in bad faith.. Don't be a concern troll on meta.

    In summary, closing questions should not be arbitrary or based on personal tastes, and this message should be conveyed to the users wherever possible.

    Then I guess we should stop using the SE engine...

  1.  
    The phrase "concern troll" is totally out of line, Harry. I agree that Kelsey L's questions needed to be closed, but I think it's fine for someone to ask about our policies.

    And we shouldn't close questions for "arbitrary" reasons or for reasons "based on personal tastes". If you feel like that describes how you close questions, then you should stop doing closing them...
  2.  
    Kelsey L was asking a lot of questions in rapid succession none of which were good. I stand by closing them as spam (not commercial spam, but still someone sending out a bunch of junk posts). But I would not have called the question spam without it being in this particular context. (In fact, for Kelsey's first few questions we closed them but not as spam and people gave explanatory comments.)
    • CommentAuthorHarry Gindi
    • CommentTimeAug 21st 2010 edited
     

    @Andy: Let me clarify. I was implying that I think that AgCl and Kelsey L are either the same person or somehow closely related. Who cries a river on meta about the rights of trolls other than the troll itself?

    As far as closing questions, it is always based on personal taste, as evinced by the large number of discussions we've had on meta (and the number of times we've simply agreed to disagree).

  3.  
    @Harry : Before accusing someone of something like that, the evidence has to be overwhelming. In this case, it isn't. I'd apologize if I were you.
  4.  

    @AgCl: If you're not the troll or a friend of the troll, then I apologize for accusing you. Could you please explain why you would bother writing up this thread about the rights of a troll?

    • CommentAuthorAgCl_
    • CommentTimeAug 21st 2010 edited
     
    IMHO, it would be better if reasons for closing could be user and time invariant, and solely be based on the question itself. That's why, I believe spam is not a good reason for that question. I clicked on the noise filtering question, and although the question was not easy to understand, and possibly off-topic, the "closed as spam" notice did not make sense to me.

    But to be honest, I wanted to discuss this issue before, and this Kelsey L case become just a trigger for me. One of my questions were closed here quite a while ago, which was about Godel's theorem, and I believe it was closed mainly because it was boring for many of the expert users, especially on the pure side, because they have thought over the subject already in depth. But it was something that was really puzzling for me at that time, and I was lucky that it didn't get closed before I got some excellent answers. Another instance was on Math.SE site, and I was a little angry because my question was closed for a reason that didn't make sense at all to me. In both cases, I had made a good effort in writing and editing the questions before they got closed, and I wouldn't feel this way if closing was well justified and given reasons were convincing for me.
    • CommentAuthorAgCl_
    • CommentTimeAug 21st 2010
     
    @Harry Gindi: I am in no way related to Kelsey L. I saw your accusation after posting my previous post, but as I explained there, this case was just a trigger for me to write about something which I was planning to write about before.
    • CommentAuthorHarry Gindi
    • CommentTimeAug 21st 2010 edited
     

    @AgCl: The list of reasons for closure is not editable by the moderators or administrators. Most of the time, none of the reasons fit well at all, so it comes down to a judgement call. This is a problem with the SE software. Also, if you're referring to a problem that happened in the past, it is a really bad idea to use the posts of a troll as your examples.

    Edit: And regarding your post about Goedel's theorem, read the comments, which explain why we voted to close as subjective and argumentative.

    • CommentAuthorAgCl_
    • CommentTimeAug 21st 2010 edited
     
    @Harry Gindi: In my question about Godel's theorem, only two of the close votes commented, and I think they did not give solid reasons to close. I didn't come to meta about it since I had already got my answers, and I was a novice user, but I must admit it was somewhat upsetting.

    I think what I would like to see is possible with this SE software. I propose that a small catalog be written, so that people would not have hesitation in choosing the reason. It would not be perfectly precise, but I think it would be an improvement over the current practice. Like this:

    Too localized:
    Case 1: Question that could be assigned as a homework problem in one of the first and second year mass mathematics courses, like calculus or linear algebra.
    Case 2: ...

    Off topic:
    Case 1: Questions that are not related to mathematics, or the relation is not clearly indicated
    Case 2:...

    Spam:
    Case1: Questions with no relevant content but only trying to promote a product or a website
  5.  
    Harry cut it out. Accusing someone of being a sock puppet with no evidence is not ok.

    I think it's a pretty common phenomenon here that someone is asking a bunch of bad questions and they all get closed, and then someone who missed most of them is confused about why one of them was closed. I don't think there's any reason to speculate about people's motives in that situation, it happens to all of us, and the answer is just "well, in context it was the right decision, even if it looks a little funny out of context."
  6.  
    I made some spam votes for precisely the same reason as Noah. If there was one reasonably good question in a sea of spam I'd've voted differently. But none of them were reasonably good. Regarding the math jobs question, it would have been trivial to see that similar questions had been asked and answered before. That plus context equaled spam.
    • CommentAuthorYemon Choi
    • CommentTimeAug 21st 2010
     

    I agree that Harry should refrain from accusations, especially when the chain of inference is so weak. I also agree with Noah's last comment: I started off giving Kelsey L's questions the benefit of the doubt, but taken as a whole they seemed disingenuous, with no indication that the responses given by others were being read.

    Sadly, the options to close list does not presently include "taking the mickey" otherwise I would have selected that. However, "not a real question" seemed a reasonable approximation, because asking questions without due care is not a habit I wish to see encouraged on MO, even if it is done in good faith. (Which, in the present case, I now doubt.)

    • CommentAuthorHarry Gindi
    • CommentTimeAug 21st 2010 edited
     

    @AgCl: I will continue to choose reasons for closure at my own discretion. No list of closure reasons can be comprehensive, and I'm willing to make that judgement when it is appropriate for me to do so.

    • CommentAuthorAgCl_
    • CommentTimeAug 21st 2010 edited
     
    Thanks for all the responses. Maybe I chose bad examples to bring this up, as Harry said, but I just wanted to say that a bit more precision in closing questions, based on a consistent policy, might reduce complaints about closed questions in the future.
    • CommentAuthorVP
    • CommentTimeAug 21st 2010
     

    AgCl_: When 3K+ user votes to close, there is a small menu of reasons for closing, which cannot be changed. After 5 votes are cast, the reason with the largest number of votes is posted (so this isn't necessarily the reason chosen by all voters).

    You did indeed choose very poor examples to illustrate your point. I completely agree with the need for a well-defined procedure for closing; but no procedure, no matter how well thought out, will be effective in the extreme cases. As far as I am concerned, any user with history of poor questions who hasn't demonstrated even basic understanding of the rules, as in the case under discussion, should have been suspended as a matter of policy. That would eliminate the necessity to determine whether any subsequent question is "spam", "not a real question", etc by vote.

    • CommentAuthoralex_o
    • CommentTimeAug 21st 2010
     
    1. Kelsey L asked a lot of bad questions. However not all of them were worthy of closure, IMHO. I liked the question on the relationship between research and
    performance in math competitions, which seems to me as valid as any other question about mathematical pedagogy.

    2. I really hate it when questions are closed as "not a real question." This begs the question: what in the world is a "real" question and how does it differ from a (presumably)
    "fake" one? If this category was narrowly applied (i.e. to rants which do not ask any coherent questions) I would be happy with it, but it seemed to have morphed into a way
    to avoid spelling out precisely what is wrong with a particular question.
    • CommentAuthorHarry Gindi
    • CommentTimeAug 21st 2010 edited
     

    raises the question

    I assume that you don't mean begging the question.

    • CommentAuthoralex_o
    • CommentTimeAug 21st 2010
     
    I also find it generally troublesome that the closure of a question can be justified by reference to previous closures and previous poor questions asked by user. This information is irrelevant; each question should be
    considered on its own merits.
    • CommentAuthoralex_o
    • CommentTimeAug 21st 2010
     
    Harry, go to the link you posted and scroll down to "modern usage."
    • CommentAuthorHarry Gindi
    • CommentTimeAug 21st 2010 edited
     

    @alex: That is generally true, unless there is evidence that the user is abusing the system.

    Edit: From the modern usage section:

    Using the term in this way is considered incorrect by most usage commentators.

    • CommentAuthoralex_o
    • CommentTimeAug 21st 2010
     
    Harry, I really don't care what "most usage commentators" say about the standard constructions of modern English. Obligatory Language Log quote:

    "For example, if we search the NYT index for recent uses of "beg the question", we find that out of the first 20 hits, 15 use "beg the question" to mean "raise the question" — and of the five that don't, four are usage articles berating people for misusing the phrase!"
    (source: http://languagelog.ldc.upenn.edu/nll/?p=2290 )

    Please, no more on this topic.
    • CommentAuthorHarry Gindi
    • CommentTimeAug 21st 2010 edited
     

    From the Language Log:

    In my opinion, those are both bad choices. If you use the phrase to mean "raise the question", some pedants will silently dismiss you as a dunce, while others will complain loudly, thus distracting everyone else from whatever you wanted to say. If you complain about others' "misuse", you come across as an annoying pedant. And if you use the phrase to mean "assume the conclusion", almost no one will understand you.

    I only take issue with the last part, which is not true if we only use the phrase formally in the company of mathematicians or logicians..

  7.  

    Harry cut it out. Accusing someone of being a sock puppet with no evidence is not ok.

    +1

    each question should be considered on its own merits.

    I agree with a version of this, but not the strict interpretation. This user asked nine questions, four of which have been deleted as spam, one of which had the body text "I have been anorexic for the past 5 months and it has been affecting my studies. What should I do?" The remaining five are responsible for 12 of the 13 current spam/offensive flags. At this point I think people are perfectly entitled to treat all the users question as suspect. This user would have to ask a pretty good question with little room for interpretation to convince me that s/he's not just fishing around for something that sounds just legitimate enough to not be shut down.

    ... seemed to have morphed into a way to avoid spelling out precisely what is wrong with a particular question.

    That's too bad. Unless the reason for closure is crystal clear, I wish people would always leave (or vote up) a comment explaining in some detail what is wrong, and perhaps how it could be fixed. Unfortunately, different people have different opinions about what is crystal clear and most people are kind of lazy. I think that discussing individual questions on meta is a decent solution. In this case, AgCl started this thread, which had the effect of holding up to the light the actions of people who closed these questions.

    One of my questions were closed here quite a while ago, which was about Godel's theorem, and I believe it was closed mainly because ...

    @AgCl: When you see a question that has been closed (or looks like it's going to be closed) with insufficient explanation, regardless of whether it's your question, always consider the option of starting a thread on meta and linking to it in the comment thread of the question. Even if you then completely disappear, the action of starting a thread draws eyes and brains to the question. I think such threads are extremely healthy for MO. They usually generate an informed consensus quickly, and they keep high rep users accountable for their actions. For MO to run smoothly, it's important that many people have a say, but it's also important to exert some pressure for them to behave responsibly. It's very hard to exert that pressure in an automated way ... human action is needed.

    ... should have been suspended as a matter of policy

    I agree, but there is no benefit to suspending a user with no rep. Setting up a new user is trivial (as it should be), and blatant trolls will have no qualms about creating many users, so immediately suspending an obvious troll just means that we'll gave a less persistent identity to deal with.

    • CommentAuthoralex_o
    • CommentTimeAug 21st 2010 edited
     
    Since what I wrote above is a bit negative, here is a (positive) suggestion. "This question is not well-written" should be a possible reason for closure. I get the sense this is what people are _really_ objecting to in Kelsey L's questions, which are all quite obviously "real" questions. As an additional data point, compare

    http://mathoverflow.net/questions/36165/work-habits-of-professional-mathematicians-closed

    which is "off topic," with

    http://mathoverflow.net/questions/9799/how-much-work-does-it-take-to-be-a-successful-mathematician

    which is apparently on-topic.

    I strongly agree with AgCl_. I don't want Kelsey L's questions open; I just want consistency.
    • CommentAuthorHarry Gindi
    • CommentTimeAug 21st 2010 edited
     

    With all due respect, I want them closed and don't really care all that much about consistency. The vote-to-close dialog presents a false heptachotomy.

    • CommentAuthorYemon Choi
    • CommentTimeAug 21st 2010
     

    I would go further than saying "this question is not well-written", in the cases of Kelsey L's questions, and say "this question does not pass a basic Turing Test for demonstrating due care and thought".

  8.  

    With all due respect, I want them closed and don't really care all that much about consistency. The vote-to-close dialog presents a false heptachotomy.

    With all due respect, you are nonetheless accountable for your actions. Objecting that the close dialog presents a false choice does not absolve you of your responsibility to give some indication of why you think a question should be closed. If none of the eight reasons for closure fits the bill, you can leave a comment, in which you can communicate any one of 256^600 possible messages. It is not acceptable (and should not be acceptable) to just pick a random reason because you vaguely feel that the question should be closed. If you strongly feel that the question should be closed, you should at least leave a comment giving some indication of why you feel that way. If you don't think you can write and stand behind such a comment, don't vote to close.

    • CommentAuthorHarry Gindi
    • CommentTimeAug 21st 2010 edited
     

    I am usually not the first person to vote to close (when I am, I usually make a meta thread), and I usually vote up someone else's comment. That was directed towards AgCl and alex_o, who both object to the "closure reason" used for certain questions. My point is that they are not sufficiently expressive without a comment, and there is no reason to come up with a formal "code" for which means which (this was the main suggestion that they were making).

    Edit: Oh wow, there are eight reasons? I thought I counted seven. I even went to the trouble of looking up what the corresponding word for dichotomy was for the case n=7.

    • CommentAuthorVP
    • CommentTimeAug 21st 2010
     

    alex_o's lament "What in the world is a real question", emended with reference to questions 36165 and 9719, reminds me of Louis Armstrong's apocryphal response to the question "What is jazz?":

    If you've gotta ask, you'll never know.

    Once again, the list of reasons for closure votes is immutable, but "not a real question" is among the easiest to decipher. In my interpretation, it also subsumes questions that are too broad (paradoxically, some of them also fall under the idiosyncratic "too localized"). By the way, I don't think there is an obligation to "spell out what is wrong with a question" if it is obvious from the context (the text of the question, the comments, the reason for the closure, history of similar questions where reasons were given) that the question doesn't pass the FAQ test. Furthermore, I may be misremembering, but I was under the impression that controversial closures were typically classified as "subjective and argumentative" and "off topic".

    • CommentAuthorHarry Gindi
    • CommentTimeAug 21st 2010 edited
     

    Worst Math(Overflow) joke:

    We should close all questions about fields as "too localized"!