Not signed in (Sign In)

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


    I think MO is a great tool for the mathematicians, but I think also that it's time to go into high gear.
    The research in mathematics cannot be reduced to: ask a "good" question and respond a "good" answer.
    In my opinion, the research needs to be less rigid than that, and much more open.

    What's research? It's dealing with the unknown.
    Do you think it's a good strategy to deal with the unknown only with questions which are well written regarding to the known part of the knowledge?
    You lock yourself in the existing concepts...

    Of course, I don't suggest being less rigorous, what I say is that we can authorize yourself a new step on the mathematics activity:
    a visionary step, before the rigorous step.

    So ok, the research activity not reduce to MO and most of you have (of course) an intuitive, visionary or fantasy level in their research activity.
    My point is that MO is a social network, so it's very sad that we can't interact at this level.
    I think it's possible: you interact at this level with yourself, why can't you do the same thing with others?

    So ok, I can already hear the responses saying that if we allow this, it's going to become rubbish.
    It's the reason why I propose to update MO, in the sense that you divide the questions section into two parts:
    a rigorous one and a visionary one, with some restrictive conditions, for example:
    - To post a visionary question, a user need to have more than 250 points of reputation.
    - Every visionary question need to be attached to a rigorous one.

    Also, remove the up and down votes for a visionary question (because it's nonsense to judge the subjectivity), but keep the stars and the view counting.
    And of course, the rigorous (resp. visionary) answers are not restrict to rigorous (resp. visionary) questions: these two aspects can interact.

    Maybe any of you will cite many forums where such interactions already exist.
    I respond that if we are looking for such interactions at a professional level, there is much less.
    And also that the important advantage is that the two aspects (rigorous and visionary) are possible in a unique place, to interact between them.

    You can say to me that what I propose is technically impossible because the StackExchange 1.0 software engine not allow such modifications.
    I respond: update the software too!

    Finally, you can think that it's nonsense to do what propose an "outsider", a young beginner...
    I respond: << out of the mouths of babes and sucklings comes the truth >> ... :)

    So deal with your own consciousness.

    • CommentAuthorWill Jagy
    • CommentTimeJun 10th 2013
    Oh, dear.
    • CommentAuthorYemon Choi
    • CommentTimeJun 10th 2013
    I am in a rush right now as I have meetings to get to, but just wish to express my profound disagreement with the OP. MO does not have the primary function of being a social network, and people using it as a blog are just contributing to noise as far as I am concerned.
    Dear Sebastien,

    If you want something done, you need to do it yourself.

    In this case, that "something" is writing software and convincing people to use it.

    Good luck.
    • CommentAuthorquid
    • CommentTimeJun 10th 2013 edited

    To add to Alexander Woo: nobody here can change the software. Once MO moves to SE2.0 you will however be able to make "feature requests", which you might consider doing then (though I think the chance of success are small).

    And for my personal opinion: +1 to Yemon Choi


    Sébastien, I suggest that you also see the section entitled "original answer" in the post

    The "money in the bank" analogy captures fairly well how these things work in practice. Similarly, it makes a lot of sense to me that a "visionary" question or answer should be well-written and backed up by a thorough knowledge of relevant technical details. And for what it's worth, in my experience, "visionary" questions that are well-written and supported by a strong technical background tend to be fairly well-received.


    We've had many, many discussions on meta on what works and what doesn't work, given the functionalities of MO. You should probably become familiar yourself with those discussions before tossing out ideas. Just like in science: nothing is written in stone, and you are free to question the great Einstein etc., but to enter the discussion, you have to be familiar with what people have said and argued and why the consensus is where it is, before you jumped in the game.

    According to current understanding, which seems pretty robust, MO is designed to work for precise questions that come up in professional research that you don't know how to address, but feel that surely someone out there would. Maybe you have such questions that you're sort of itching to know the answer to? I suggest you try using MO in that way first, and get to know a bit more about its strengths and weaknesses, before attempting to change how things work around here.


    Sebastien, I am very happy that your suggestion has no chance of being implemented. But if you do create a separate social network that emphasizes visionary approaches, I will be glad to take a look at it.

    MathOverflow works quite well as a place for well-formulated mathematical questions that have definite answers. Visionary speculation usually doesn't have a well-defined answer, so it doesn't fit well with the underlying software.

    If you need to share more speculative mathematical ideas, you can post papers on the ArXiv or Vixra, where they will be timestamped and shared to the rest of the world. Obviously, there is no guarantee that anyone will read your ideas there, but you don't have a natural right to force people to read what you want them to read.

    Other visionary ideas can be sent here:

    Can we close this now?

    • CommentAuthorWill Jagy
    • CommentTimeJun 10th 2013 edited
    Scott, in The Phantom Tollbooth, the kingdom of Dictionopolis had a half bakery.

    I have reading all your comment with interest, and I understand that my speculative issue is not (yet?) adapted to MO.
    Maybe I will defend my point of view on metaMO about evolutions in this direction (as @quid said in his comment)
    But for now, it's ok, please delete this issue:
    @will I don't understand your last comment, maybe it contains puns that a french can't understand.
    Ok, I understand: "half-baked idea" is an ironic idiom that designates an idea which is not finished to be thought, so that it appears uncouth.
    Of course, there is a difference between speculative ideas and half-baked ideas, and assimilate my deleted issue into an "half-baked issue" is mockery...

    Anyway, dealing only with speculative ideas is not my point. My point is that there is a continuum in our mind between speculative ideas and accomplished ideas.
    The fact that these two aspects, and all the continuum between them, could find their place in a common platform, and interact, would be able to help to accomplish
    (in a mathematics research level) many new ideas that would otherwise remain suspended in the speculative level.

    I'm not at all an expert in website and software engineering, I do not think I will develop such an application, for now.
    Maybe it would be easy for Anton Geraschenko to do that very quickly with the same quality as MO.
    I would have rather preferred MO develops further in this direction (in a very proper manner, of course).

    The intermediate manner would be to create an adjunct of the main site, as meta.MO is. This could be called spec.MO:
    On spec.MO, each issue could be classified in several level of speculativeness (for example: from 1 (almost accomplished) to 7 (very speculative))
    controlled by people with great reputation, and so that each issue can progress in time (thanks to interactions with other people),
    and finally reach the level for being posted on the main site.
    • CommentAuthorquid
    • CommentTimeJun 11th 2013

    Some more remarks on technicalities:

    Not to take anything away from Anton Geraschenko, to whom I am grateful and who impressed me on numerus occassion in relation to running this site, but he did not write the software running this site and he cannot change it in any fundamental way. (There were some costumizations made to the site, eg regarding supporting math-typesetting, by him and some others, likely Scott Morisson deserves special mention, though these thing happen before I was around so I am not firm on the details.)

    The main compotent of the site comes from stackexchnage (technically this might be incorrect as the name of the company then I believe was different but anyway it comes from somebody else entirely). They first made for programming then some other sites on similar subjects superuser and serverfault and also decided the could offer the infrastructure as a service to anybody who had an idea for a Q&A site on whatever subject and various sites were created on various subjects, one of them on math, this one.

    Then for some reason the idea of offering the infrastructure as a service to anybody was discuntinued and the Stackexchange Network was created a network of numerous Q&A sites for all kinds of subjects, from IT rlated things to science (there is also a math site there, not this one) to science fiction to gardening, cooking, languages, ...

    Yet, the service for existing sites was continued and this is the current situation of MO, a site on a legacy version of this software. However, MO is in the process of joining this network and updating to the newer version of the software which is however similar.

    So, now what does this mean:

    1. at the moment the MO software essentially cannot be changed at all.

    2. once this migration happens it is in principle possible to make suggstions (to stackexchange) for changing the software (that then would typically affect the full network): for such a major change (perhaps not concerning the work it would take but changing the character of the site) this is unlikely to work out in my opinion.

    3. since you seem to imagine somehow a site as a complement to MO, this is technically simple. On the stackexchange network one can suggest sites on a new topic, see here So you could formualte a proposal and if you find enough people interested in it, the site will be launched; there is no programming involved; you "only" need to find enough people interested in it.

    As said, personally, I am not interested in such a thing. But I can imagine others might be.

    Thank you @quid for this complementary information and for your candidness.
    I did not know that it could be so simple to open a new social network.

    In my opinion, it’s unfortunate that mathematics social network multiply: : for people studying math at any level and professionals : for professional mathematicians

    and maybe in the future: : to interact between “speculative ideas” and “concrete ideas”, for professional mathematicians. : to interact on “near-math topics”, for professional mathematicians.

    because there is a solution to unify all!!

    (This spirit of separation and partitioning afraid me)

    The solution is the following. As I said in my previous comment:
    - There is no separation between “concrete ideas” and “speculative ideas” , but a continuum between them.
    - There is no separation between “maths topics” and “near-maths topic”, but a continuum between them.

    - well-formulated-issues / bad-formulated-issues
    - useful-issues / useless issues
    - research level topics / student level topics
    ... etc.... all the parameters we want...

    So, it suffices to consider all these parameters p, and for each issue, the writer assigns honestly a number n between 1 and 10. Beforehand, the f.a.q. explains in details what’s a parameter p at level n. If the assignation is not correct, some readers with high reputation can suggest (or vote) a corrected assignation. Of course, the spirit is that each issue can progress thanks to the interaction with other people. So, after a progress, the writer is free to honestly modify the assignation of its issue.

    So what is the purpose of this system? The purpose is the following:
    If (as you said) you are exclusively interested in issues which are:
    - very concrete
    - very pure math
    - very well-formulated
    - very useful
    - very high research level

    then, you adjust all the parameters of your “personalized mathoverflow” to 1, so that you will see only what you want.
    Now let’s me enounce all the advantages of such a system:
    - It’s unified: it’s not necessary to multiply the social networks.
    - Everyone is happy: because we can adjust the parameters as we want.
    - Smooth moderators’ work: in the former system there is only acceptable/nonacceptable, it’s violence for the user, then, with this new system with assignation of parameters (from 1 to 10), it’s smoother. People are more honest than we think, and everything will be all right.
    - More contributions: in the former system, some people (with ideas on an issue) do not dare answer, because they are not expert in the subject. In the new system, we can honestly parametrize our answer, and so we dare answer.
    - Very dynamic: the issue progresses over time thanks to the interaction with other users.
    So maybe a contributor read an issue or an answer and say: “it’s not perfectly formulated, but it’s a good idea, and I know how upgrade your answer”…

    I wonder what Anton Geraschenko thinks about that; maybe I will suggest this unification to stackexange directly, instead of the opening of another new network, I don’t know…

    It is obvious to me that the research will develop more and more in this direction.
    The research will then move much faster. Here is an analogy:
    The British mathematicians William Shanks (1812-1882) computed by hand 707 digits of Pi during 20 years. Today, such a work is nonsense because of our powerful computers…
    In the future, the computers will certainly not discover and prove new theorems, but we will use them to organize the research into a unified social network, in the spirit I described above, so that in a few years, mathematicians of the next generation will say:
    “Mathematicians have needed 100 years to prove the Poincaré conjecture, but today it’s nonsense, with our powerful unified research network, we can easily prove such conjectures in 1 year!”.
    Sebastien: Some people think von Neumann algebras aren't very interesting. Mostly, they deal with this by working on something other than von Neumann algebras, not by trying to talk everyone who studies von Neumann algebras into doing something else entirely.

    Some people think the mission of MathOverflow isn't very interesting. Mostly, they deal with this by spending their time reading other blogs and websites that are more to their taste, or by starting their own blogs and websites. Let a thousand flowers bloom.
    It is possible (judging from the latest comment of the OP) that the kind of thing he should be looking at is the Polymath project (see e.g. and links therein), rather than Mathoverflow.
    Thank you Steve for your poetical comment.
    Steve, of course I'm not against diversity.
    I am just looking for a complete, collaborative and unified math-research network.
    Thank you Vladimir, I did not know the Polymath project, it's interesting.
    It's a little closer to what I want, but in reality, it is still very very far.
    In many standard (finishing, nice, functional, fun, possibilities, number of users, diversity of topics ...) mathoverflow is much higher than Polymath Project.
    Moreover, in Polymath, you can’t post a proposal or a comment without the approval of moderators: it’s too restrictive, too slow...

    In fact, both of them are largely incomplete. What I would like is to keep the presentation of mathoverflow and add this collaborative aspect,
    but replacing all the restrictions by assigning a number from 1 to 10 to the following parameters, to each issue (as I have already explained):
    1. concrete/speculative
    2. pure-math/near-math
    3. well-formulated/bad-formulated
    4. useful/useless
    5. research level/student level

    Add also the opportunity to upgrade each question, answer and comment, thanks to the interaction with other people.
    Everyone would be able to adjust the display parameters for the questions, the answers and the comments, so that we see that we want.

    We then get a real complete, collaborative and unified math-research network.

    Let me take another analogy:
    Today, solving a problem for mathematicians, it is a bit like trying to get out of a maze for inhabitants of 'flatland'...
    What I propose is to add at least 5 dimensions to this land, so that, we have for example, the opportunity to rise above the maze,
    thanks to the concrete/speculative dimension etc...
    Dear Sébastien,

    No one is preventing you from making such a website of your own. If you really believe in your vision, not only should you tell people about it (as you have), but also you should be willing to invest your own time and effort into making it happen.
    Dear Sébastien,

    I personally feel that I would want to avoid a version of Mathoverflow allowing speculative/badly-formulated questions. I suspect that many other MO users share this feeling. It would be great if you and other champions of this idea create such a website and manage it, but I do not see a single good reason to incorporate it into MO. Globalisation is not always a great thing. A very high standing of MO is, among other things, due to the fact that its creators maintained the right level of "hygienic standards" from the very beginning.