From: Virgil on
In article <1150876556.902172.6800(a)i40g2000cwc.googlegroups.com>,
mueckenh(a)rz.fh-augsburg.de wrote:

> Rupert schrieb:
>
> > Are you familiar with existential quantifiers and universal
> > quantifiers? You are making a statement of the form "There does not
> > exist an f such that...", or, alternatively "For all f it is not the
> > case that..." Then later on you use f to refer to a specific function.
> > You should use different letters on these two occasions.
>
> No. "For all f it is not the case" means the same as "there does not
> exist any f for which it is the case". A f (f =/= surj) <==> ~E f (f =
> sur).
>
> And I let f be any function you like.

Let it be any function which does not have K_f as a value.
> >
> > "Let f be an arbitrary function. There does not exist a bijective
> > mapping g:N->M, where M is the set consisting of all the finite subsets
> > of N together with the set K={k e N: k /e f(k)}."
>
> That is what I said.

Then f is not arbitrary, since you insist on restricting the function f
to have some n in N such that f(n) = {x in n: x not in f(x)}.

With that restriction, no such f can exist, absent that restriction,
lots of them do.
From: Virgil on
In article <1150897288.785127.87000(a)g10g2000cwb.googlegroups.com>,
mueckenh(a)rz.fh-augsburg.de wrote:

> Rupert schrieb:
>
>
> > > Except that if the function is require to have K as a value, the f and K
> > > and M cannot exist:
> > > if f(n) = K = {k e N: k /e f(k)} is n a member of K or not?
> >
> > This is a perfectly good proof that K cannot be in the range of f. But
> > as far as I can tell, there would still be a bijection g:N->M, and K
> > would be in the range of g. g would of course be different from f. No?
>
> 2) 0.12324389
> 3) 0.23123123
> 4) 0.85348714
> ..
>
> 1) 0.244...
>
> This is a perfectly good proof, that a surjective mapping g: |N -->
> [0,1] does exist, isn't it?

No. And it totally ignores the fact that Muecken's alleged example of a
n uncountable countable set is fatally flawed, as is shown above.
From: Daryl McCullough on
In article <1150876556.902172.6800(a)i40g2000cwc.googlegroups.com>,
mueckenh(a)rz.fh-augsburg.de says...
>
>
>Rupert schrieb:
>
>> Are you familiar with existential quantifiers and universal
>> quantifiers? You are making a statement of the form "There does not
>> exist an f such that...", or, alternatively "For all f it is not the
>> case that..." Then later on you use f to refer to a specific function.
>> You should use different letters on these two occasions.
>
>No. "For all f it is not the case" means the same as "there does not
>exist any f for which it is the case". A f (f =/= surj) <==> ~E f (f =
>sur).
>
>And I let f be any function you like.
>>
>> "Let f be an arbitrary function. There does not exist a bijective
>> mapping g:N->M, where M is the set consisting of all the finite subsets
>> of N together with the set K={k e N: k /e f(k)}."
>
>That is what I said.
>>
>> This is false. There always will exist such a bijection g. It will of
>> course be different to f.
>
>The same is true is you take Hessenberg's mapping of f:N --> P(N). It
>is not surjective because of K(f). But if you define g(n+1) = f(n) and
>g(1) = K(f), the proof is no longer a proof.
>
>Regards, WM
>

From: Daryl McCullough on
mueckenh(a)rz.fh-augsburg.de says...

>Rupert schrieb:

>> "Let f be an arbitrary function. There does not exist a bijective
>> mapping g:N->M, where M is the set consisting of all the finite subsets
>> of N together with the set K={k e N: k /e f(k)}."
>
>That is what I said.
>>
>> This is false. There always will exist such a bijection g. It will of
>> course be different to f.
>
>The same is true is you take Hessenberg's mapping of f:N --> P(N). It
>is not surjective because of K(f). But if you define g(n+1) = f(n) and
>g(1) = K(f), the proof is no longer a proof.

Sure it is. f is not a surjection from N -> P(N), because
K(f) is not in the image of f. Similarly, g is not a surjection
from N to P(N) because K(g) is not in the image of g.

Let's go through this one more time.

Let f be any function from N to P(N).
Let image(f) =

{ A in P(N) | exists x in N: f(x) = A }

Let K(f) =

{ x in N | x is not in f(x) }

Let M(f) =

{ A in P(N) | A is in image(f) or A = K(f) }

Finally, let g be defined by

g(0) = K(f)
g(x+1) = f(x)

Okay, so we have the following facts:

1. f is not a surjection from N to M(f)
2. f is not a surjection from N to P(N)
3. g is a surjection from N to M(f).
4. g is not a surjection from N to M(g).
5. g is not a surjection from N to P(N).

So, M(f) is countable, since there is a surjection g from
N to M(f). But there is no surjection from N to P(N). So
P(N) is *not* countable.

--
Daryl McCullough
Ithaca, NY

From: Rupert on

mueckenh(a)rz.fh-augsburg.de wrote:
> Rupert schrieb:
>
>
> > > Except that if the function is require to have K as a value, the f and K
> > > and M cannot exist:
> > > if f(n) = K = {k e N: k /e f(k)} is n a member of K or not?
> >
> > This is a perfectly good proof that K cannot be in the range of f. But
> > as far as I can tell, there would still be a bijection g:N->M, and K
> > would be in the range of g. g would of course be different from f. No?
>
> 2) 0.12324389
> 3) 0.23123123
> 4) 0.85348714
> ..
>
> 1) 0.244...
>
> This is a perfectly good proof, that a surjective mapping g: |N -->
> [0,1] does exist, isn't it?
>

No, that's absolute nonsense.

> Regards, WM

First  |  Prev  |  Next  |  Last
Pages: 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Prev: integral problem
Next: Prime numbers