
kr
Senior Member

Posts: 1887
Joined: Sep 2002
|
Tue Oct 28, 03 03:10 PM
|
|

Quote
get to the heart of the matter, extract the essentials, and generalize that
I don't mean to be argumentative, but this is just easier said than done. At least in regards to Hilbert's list of problems, the challenge there is that one isn't always sure exactly what the essentials are. For instance, Hardy and his gang used to think that an elementary proof of the prime number theorem (i.e. one without the use of complex numbers) was impossible. Then Bombieri figured out how to do it. So, Hardy misjudged what the essence of the problem really was. There are other problems that are like this - different ways to reach the same result, completely different methods, etc. Or, the solution to the problem may require a little pinch of this, a tablespoon of that - take a look at the Fermat's Last Theorem proof. There are attempts at generalization, but it was really specialization of general theories that solved the problem.
Particularly in analytic number theory, there has been a press in the last fifty years to solve some generalized versions of the specific techniques used to solve easier problems, in order to get better results. I would say the results of this form of attack have been rather poor, with researchers winding up in the doldrums of their generalizations.
Anyhow, this just shows my bias, and the approach of number theorists is not much like other branches of math. I'm also partial to Hilbert's idea of identifying the great problems as a means of focusing and directing the field. But it's been a hundred years, and we need a new list now. Maybe the whole idea is out of date... I don't know how I'd argue that one.
-------------------------
Fools! I'm Nonius!!!!
|
|