Gmail Agenda Documenten Reader Het internet meer »
Onlangs bekeken groepen | Help | Aanmelden
Google Discussiegroepen Startpagina
Bericht van Godel's proof, truth, reality, self-awareness, and all that jazz-discussie
De groep waarnaar je een bericht verzendt, is een Usenet-groep. Berichten die je in deze groep verzendt, zijn zichtbaar voor iedereen op het Internet
Je antwoord is niet verzonden.
Uw bericht is geplaatst
 
Van:
Aan:
Cc:
Reactie op:
Cc toevoegen | Reactie toevoegen | Onderwerp bewerken
Onderwerp:
Validatie:
Typ ter verificatie de tekens uit de onderstaande afbeelding of de getallen die je hoort wanneer je klikt op het pictogram voor toegankelijkheid. Luister en typ de nummers die je hoort
 
cbrown@cbrownsystems.com  
Profiel weergeven   Naar het vertalen Vertaald (origineel weergeven)
 Meer opties 26 sep 2007, 22:52
Nieuwsgroepen: sci.math
Van: "cbr...@cbrownsystems.com" <cbr...@cbrownsystems.com>
Datum: Wed, 26 Sep 2007 13:52:10 -0700
Lokaal: wo 26 sep 2007 22:52
Onderwerp: Re: Godel's proof, truth, reality, self-awareness, and all that jazz
On Sep 26, 12:48 am, Han de Bruijn <Han.deBru...@DTO.TUDelft.NL>
wrote:

> cbr...@cbrownsystems.com wrote:
> > See my previous comments on this system:

> >http://groups.google.com/group/sci.logic/browse_frm/thread/e02b54f8cd...

> > It is /not/ a system of heriditary finite sets, it is a system
> > consisting of "is a set"s (/sets/ with a single element) and "are
> > sets"s (which are finite /sequences/ of two or more "is a set"s).

> > The main problem is that his system does not obey "A = B iff (x in A
> > iff x in B)" as it stands: the encoding for {{}, {{}}} is not the same
> > as the encoding for {{{}}, {}}, and equality is defined in terms of
> > equality of encoding. Thus we cannot make sense of the statement "A =
> > {x : P(x)}" in his system.

> My "axiomtization" of the system is far from perfect, admittedly. But in
> order to avoid further misunderstanding, the following snippet of source
> code might explain why it doesn't work the way you've understood (though
> formally you may be right).

Oh, crap! Yum, yum, a delicious bowl of crow for me to eat! :)

Yes, I misread your description of a set as formalized by a sequence
of '0's, '1's, and '2's as your encoding upon which you based your
equality. As it is based on this other equality, there /is/ a
bijection such that AA = A, AB = BA, etc.

A recursive version of your encoding f : HF -> N is described by:

f({}) = 0
f(A) = sum{x in A} 2^f(x).

(I write "sum", but it would be more appropriate to use bitwise or).

f is well-defined for HF sets; since HF sets are well-founded (no
infinite descending chains of membership). This is what I assume your
rule (3) is trying to say.

Note that if x in A, then f(x) < f(A).

If we write "v" for bitwise or, "&" for bitwise and, "!=" for not
equals, and consider 2^x as a single bit shift, then we can prove that

x in A iff ((2^f(x)) & f(A) = (2^f(x)))

>From this it easily follows that

A = B iff f(A) = f(B)

so f is an injection.

The inverse encoding g : N -> HF is defined recursively as:

g(0) = {}
g(n) = { g(m) : 2^m & n = 2^m}

g is well defined, since if 2^m & n != 0, then m < n.
Again, g is an injection. And

g(f({})) = {}
g(f(A)) = {g(f(x)) : 2^f(x) & f(A) = 2^f(x)}
        = {g(f(x)) : x in A}

By induction staring with g(0), and noting that x in A -> f(x) < f(A),
we can then deduce that g(f(A)) = A; so since g and f are injections,
g is the inverse of f, and f is a bijection.

As regards your production rules, the naturals do provide a model
using f:

Define:
"A is a set" iff A = 0 or A = 2^n for some n.
"A are sets" iff not (A is a set)
"{}" = 0.

"{} is a set."
0 is a set.

"If A is a set or are sets, then {A} is a set."
Define {A} = 2^A; then {A} is a set.

"If A is a set and B is a set then AB are sets"
In this case, we define AB = 2^A v 2^B; so AB are sets.

"If A is a set and B are sets then AB are sets"
In this case, we define AB = 2^A v B; again AB are sets.

"There are no other ways of forming sets."
This is your vaguest axiom; it would be better to say that every set/
sets /can/ be formed in this way through a finite sequence of
applications of rules 0, 1, and 2 (or state it as the contrapositive).
We can take it as saying that every natural A is either 0, or is the
sum (bitwise or) of one or more powers of 2, each of those powers
being less than A; and we can satisfy the requirement via induction.

Because of the function g, the HF sets are also a model; "A is a set"
means |A| = 0 or 1, "A are sets" means |A| > 1, {A} is defined as
usual, and AB = {A} union {B} or {A} union B depending on whether B is
a set or are sets.

>From this interpretation, AA = A, and AB = BA; so my earlier comments

were wrong, wrong, wrong.

Cheers - Chas


    Doorsturen  
Je moet je aanmelden voordat je berichten kunt plaatsen.
Als je een bericht wilt verzenden, moet je eerst deelnemen aan deze discussiegroep.
Werk je bijnaam bij op de pagina met abonnementsinstellingen voordat je een bericht plaatst.
Je hebt geen toestemming om berichten te plaatsen.

Discussiegroep maken - Google Discussiegroepen - Google Startpagina - Servicevoorwaarden - Privacybeleid
©2010 Google