From: John Schutkeker on
Jeremy Boden <jeremy(a)jboden.demon.co.uk> wrote in
news:1156925658.5449.8.camel(a)localhost.localdomain:

> On Wed, 2006-08-30 at 09:15 +0200, Han de Bruijn wrote:
>> Jesse F. Hughes wrote:
>>
>> Modern mathematics _is_ an experimental science.
>> In very much the same way as physics is: theory as well as
>> experiment. Or would you like to say that structural (numerical)
>> analysis is not a form of mathematics? And how about computing a
>> large prime number?
>>
>> Welcome to the 21th century!
>
> Do you mean that there are theories of numerical analysis which are
> accepted until found to be wrong due to practical experiments?
>
> Actually, I don't think anyone has discovered a practical method of
> *computing* a large prime number - although there are methods of
> performing primality checking on a given number. The faster methods
> are only probabilistic - but slower methods will give a definite
> answer.

All of which is just quibbling over details. You are describing
numerical experiments.
From: John Schutkeker on
Han de Bruijn <Han.deBruijn(a)DTO.TUDelft.NL> wrote in
news:20b9$44f53d00$82a1e228$14726(a)news1.tudelft.nl:

> Gerry Myerson wrote:
>
>> Somewhere along the way you should check to see whether what you have
>> done (or what you are planning to do) has already been done. [ ... ]
>
> Precisely! And there are a lot of things out there which have NOT been
> done. To mention an example. Take a couple of (3-D) bricks and find
> out how to make a discretization of the equations describing
> incompressible and irrotational 3-D flow with the least squares finite
> element method.
>
> But ah .. nobody will appreciate (or even comprehend) if you have done
> such an uncommon thing.

There are people all over the place specializing in fluid simulation,
who would appreciate such work. If you feel that such a simulation
would answer an important question (and you have the talent to write it)
it is unsconscionable not to do so, even if you have to write a fresh
grant application first. :P
From: schoenfeld.one on

Proginoskes wrote:
> schoenfeld.one(a)gmail.com wrote:
> > Han de Bruijn wrote:
> > [...]
> > > Or is it just mathematics? In the latter case, computing a large prime
> > > is also mathematics, because it could be done - in principle - by hand.
> > > (What else does computer science add except more speed and more space.)
> >
> > Then there is no experiementation. Mathematics is not an experimental
> > science, it is not even a science. The principle of falsifiability does
> > not apply.
>
> Written by someone who has not done any math research.
>
> One of many examples: Try dividing 2^n by n and keeping track of the
> remainders. You won't get 1; you get 2 a lot, but you never seem to get
> a 3. So you conjecture:
>
> CONJECTURE: The remainder of 2^n divided by n is never 3.
>
> However, this conjecture is false; in particular, the remainder of 2^n
> divided by n is 3 if n = 4,700,063,497 (but for no smaller n's).

Hello Crackpot.

> --- Christopher Heckman

From: Han de Bruijn on
schoenfeld.one(a)gmail.com wrote:

> Then there is no experiementation. Mathematics is not an experimental
> science, it is not even a science. The principle of falsifiability does
> not apply.

Any even number > 2 is the sum of two prime numbers. Now suppose that I
find just _one_ huge number for which this (well-known) conjecture does
_not_ hold. By mere number crunching. Isn't that an application of the
"principle of falsifiability" to mathematics?

Han de Bruijn

From: Han de Bruijn on
schoenfeld.one(a)gmail.com wrote:

> Proginoskes wrote:
>>
>>CONJECTURE: The remainder of 2^n divided by n is never 3.
>>
>>However, this conjecture is false; in particular, the remainder of 2^n
>>divided by n is 3 if n = 4,700,063,497 (but for no smaller n's).
>
> Hello Crackpot.

Hey, hey! Wash your mouth! If *that* is crackpottery, I'll eat my hat!

Han de Bruijn

First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Prev: Any coordinate system in GR?
Next: Euclidean Spaces