Not signed in (Sign In)

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

    • CommentAuthorvoloch
    • CommentTimeOct 1st 2010
     
    From Aaronson's blog,
    http://scottaaronson.com/blog/?p=466

    "Look: if I really needed to know what (say) the best-regarded PhD programs in computer science were, I could post my question to a site like MathOverflow—and in the half hour before the question was closed for being off-topic, I’d get vastly more reliable answers than the ones the NRC took fifteen years and more than four million dollars to generate."
  1.  

    He may be on to something...

  2.  

    I think this is a good thing, and a bad thing. On the one hand, people recognize MO as a great source of knowledge for mathematics. This is great. On the other hand he says "I could post my question to a site like MathOverflow—and in the half hour before the question was closed for being off-topic, I’d get vastly more reliable answers than the ones the NRC took fifteen years and more than four million dollars to generate" meaning that perhaps people don't respect the editorial policies of MO enough, because someone will answer off topic questions before they are closed. This is a bad thing.

  3.  
    The site

    http://graduate-school.phds.org/rankings

    allows you to set your own weights to the various categories that were considered. If you want to rate the quality of the faculty and research, with an emphasis on current research, I suggest

    NRC Quality measure

    Regression 5

    Research Productivity

    Publications 3 (5 is possible, but some fudging in this category is possible)
    Grants 5 (NO fudging in this category)

    That's all--ZERO for all other categories. Do this for computer science and the rankings look quite reasonable (MIT, where Scott is, comes in 2nd). Do it for math, and the results are more surprising, but not bizarre. Notice that this weighting downrates esteemed departments with very senior (over-the-hill?) faculty. As a member of the senior (OTH?) gang, I am not sure that this is bad.

    Now if a very good prospective grad student who is not sure of direction but is trying to narrow down the list of possible schools wants to rank schools, he or she should add other categories. Average support, average time to degree, etc. are irrelevant. The student will learn from the offer letter what is his or her support, and who cares about the AVERAGE time to degree? What is most relevant is the availability of strong faculty. I suggest adding

    Program Size

    Student-faculty ratio (low option) 5

    Even if you do not agree with the resulting ranking, it gives you pause for thought.
  4.  
    @Grétar Amazeen: I do not agree. This is NOT a topic for MO, even if it is fine to discuss it on Meta. MO is (appropriately, IMO) reserved for research level questions in mathematics.
  5.  

    @Bill: I think you misread what Grétar wrote.

  6.  

    Moreover, Scott Aaronson is well aware of what is an isn't appropriate at MathOverflow, and was making tongue-in-cheek fun of our stodginess (which he, I think, implicitly approves of).

  7.  

    @Bill. I think I wasn't clear enough. I agree, that this is not a topic for MO, my point was that MO can't get a reputation for being a site where people can ask questions that are not research level, and get good answers before they are closed.

    @Scott. I know that Scott Aaronson was being tongue-in-cheek, but it might reflect upon how other people view the site. I mean, we are still getting obvious homework problems as questions, and they are still getting answered before they are closed.

  8.  
    @Grétar Amazeen: Sorry; I misunderstood.

    I do wish that people would not answer HW level questions. When someone I can identify as being a serious person does, I sometimes down vote the answer and make a comment.
  9.  
    This conversation has drifted a little from its original intent---but if there is a moderator reading, can I ask whether the following is appropriate: if any experienced user sees "Q: do my homework" "A: Ok here is the answer" then they can hit the "flag" option, and flag the answer for moderator attention. The moderator then comes along, *deletes the answer*, and closes the question. This way, even though the OP may well get the answer anyway (depending on how quickly they check for an answer), it _looks_ as if they didn't get it to anyone who cares to look later on, thus giving the _impression_ to newer serious users that they shouldn't be answering these questions.
    • CommentAuthorWillieWong
    • CommentTimeOct 4th 2010 edited
     

    But unless we make clear to those who do answer homework questions, many of whom presumeably don't read Meta, that this will be the policy, I can imagine a whole bunch of whining about "what happened to my answer?"

    If we were to clearly post this policy somewhere obvious to see, I fear that it would partially defeat the purpose: some new serious users may be apt to think that this is not a community policy, rather a strict enforcement by a minority of moderators who are tight a%%es. (As in, the purpose of "lead by example and not by iron fist" will be compromised.)

  10.  

    Since I am a moderator and reading this, I thought I would just chime in. At the moment, I don't feel (and I haven't gotten any strong sense from the other moderators about this) that the the amount of HW being posted is problematic, or that it requires a more active response than we are taking right now. On a given day, close to 100 questions are active (asked/answered/edited) and about 3 or 4 are closed. I still kind of wonder who the people asking these HW questions are, but on the internet, there are alway a certain number of people who can't get with the program; it comes with the territory.

  11.  

    Kevin, I see no problem with flagging, for those who are willing to put in the effort to click and type a bit. I have been deleting obviously spurious answers whenever I encounter them, so this would not be a substantial change. Those of you with 10k+ points can check out (tools) -> (links) -> (recently deleted) if you're really interested. I usually leave a short comment explaining why I am deleting the answer.