From: Han de Bruijn on
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.

Han de Bruijn

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

> Han de Bruijn wrote:
>
>>Jesse F. Hughes wrote:
>>
>>>*Fortunately* mathematics is not an experimental science.
>>
>>Not true. Part of 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!
>
> Wouldn't that be computer science?

Is a simple calculation by hand (like 1 + 1 = 2) "computer science" ?
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.)

Han de Bruijn

From: schoenfeld.one on

Han de Bruijn wrote:
> schoenfeld.one(a)gmail.com wrote:
>
> > Han de Bruijn wrote:
> >
> >>Jesse F. Hughes wrote:
> >>
> >>>*Fortunately* mathematics is not an experimental science.
> >>
> >>Not true. Part of 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!
> >
> > Wouldn't that be computer science?
>
> Is a simple calculation by hand (like 1 + 1 = 2) "computer science" ?

No.

> 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.

> Han de Bruijn

From: Jeremy Boden on
On Wed, 2006-08-30 at 09:15 +0200, Han de Bruijn wrote:
> Jesse F. Hughes wrote:
>
> > *Fortunately* mathematics is not an experimental science.
>
> Not true. Part of 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.

--
Jeremy Boden


From: Proginoskes on

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).

--- Christopher Heckman

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