11
$\begingroup$

This is in relation to http://meta.math.stackexchange.com/questions/1868/list-of-generalizations-of-common-questions : I noticed that there have been quite a few questions (recently and otherwise) which to the trained eye immediately says "Apply Sylow-1, then apply Sylow-2, then Sylow-3, then chug."

It is somewhat redundant (in my opinion) to have so many questions whose answers all boil down to standard manipulations plus a computational step involving factoring some number; this question is to ask whether other users feel the same.

  • Should we create an "application of Sylow theorems" generalisation and just point users toward it in the future?

  • Another possibility would be to create a (sylow-theorem) tag so it would be easier for users to locate those questions (as a model/template for problem solving, I suppose).

Other ideas, suggestions, comments on why this is a bad idea, are welcome.

Per Alex Becker's request in the comment, here's a sample of stuff you get if you search for "group order" on the main site.

$\endgroup$
12
  • 1
    $\begingroup$ Can you link to some of these questions? I am rather bad at recognizing how to solve them, and would relish the examples. $\endgroup$
    – Alex Becker
    May 9, 2012 at 7:43
  • 2
    $\begingroup$ @Alex: considering the audience of this website, my kneejerk reaction to seeing a question asking about structure of group of order $N$ where $N$ is a reasonably small positive integer (such as existence of certain subgroup of certain order, the non-existence of normal subgroups of certain order, whether the group is simple etc.) is that they are almost certainly homework/exercise questions (based on the arbitrariness of the formulation) after a lecture on Sylow theorems have been given. There are, of course, exceptions where more elementary (Cauchy or Lagrange theorems) can be used... $\endgroup$
    – Willie Wong
    May 9, 2012 at 7:55
  • $\begingroup$ ... as well as problems that require (or can be solved more simply by) slightly higher powered machinery. But as you can see from the list I just added there is a definite trend to the "shape" of those questions. $\endgroup$
    – Willie Wong
    May 9, 2012 at 7:57
  • $\begingroup$ An example of an exception is this recent question math.stackexchange.com/questions/142827/… which needs a bit more input than just direct computation. $\endgroup$
    – Willie Wong
    May 9, 2012 at 8:00
  • $\begingroup$ Thank you, those should serve as instructive review for me. $\endgroup$
    – Alex Becker
    May 9, 2012 at 8:45
  • $\begingroup$ I think the "Applications of Sylow's Theorems" generalisation is a very good idea. If for any reason people object to that, then having a sylow-theorem tag might at least help. $\endgroup$
    – Tara B
    May 9, 2012 at 11:16
  • 8
    $\begingroup$ Dear Willie, I think that a Sylow-theorems tag would be a good idea. Regards, $\endgroup$
    – Matt E
    May 9, 2012 at 11:35
  • 2
    $\begingroup$ I've seen a website, which I can't find at the moment, where one enters a natural number (below some bound, maybe 1000) and the site automatically provides a proof using the Sylow theorems that there are no simple groups of that order, or a classification of the simple groups of that order. $\endgroup$
    – Zev Chonoles
    May 9, 2012 at 16:58
  • $\begingroup$ Do you have in mind a certain question that could be promoted to a catch-all reference? Otherwise I'm willing to try and write something in the space below — I think it would be an interesting diversion. $\endgroup$
    – Dylan Moreland
    May 9, 2012 at 20:10
  • 1
    $\begingroup$ @Dylan unfortunately no. The closest I can find is this answer. I wish I still have my undergraduate algebra notes: John Conway gave an algorithm (along with wonderful remarks on why doing it that way is the best) for doing problems like this. So your effort will be very much appreciated. $\endgroup$
    – Willie Wong
    May 10, 2012 at 7:42
  • $\begingroup$ @WillieWong I hope you can dig that up! It brings up an interesting point: without an algorithm, how do you generalize all of the examples given? I suppose the thing to do is talk about the various techniques available (the three main theorems, counting elements of certain orders, embedding into symmetric groups, semidirect products, etc.) and display them in some fundamental examples (order $pq$, $p^2q$, $p^2$, etc.) Is this what we're after? $\endgroup$
    – Dylan Moreland
    May 10, 2012 at 23:57
  • $\begingroup$ @Dylan: I don't have the notes anymore :( but from my vague recollection it is just a step by step instruction for problems of the type "classifying all finite groups of order $m$", which of course starts by finding all the $p$ subgroups, then counting conjugacy classes and ascertain the normal ones, then ruling out impossible combinations by some elementary number theory, and then writing down what the group can be. $\endgroup$
    – Willie Wong
    May 14, 2012 at 7:15

1 Answer 1

Reset to default
5
$\begingroup$

Creating a generalized question seems like the best option to me—particularly since you can ask and answer the question you think is appropriately general and see what the response to it is before adding it to the list of generalizations or closing other questions as a duplicate of it—but I see no reason why you couldn't both create the generalized question and have an appropriate tag.

$\endgroup$

You must log in to answer this question.

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