On 20 Dec 2013, at 01:01, LizR wrote:

On 20 December 2013 11:40, meekerdb <meeke...@verizon.net> wrote:
On 12/19/2013 1:30 PM, Jesse Mazer wrote:
To me it seems like "thinking something is true" is much more of a fuzzy category that "asserting something is true"

Maybe. But note that Bruno's MGA is couched in terms of a dream, just to avoid any input/output. That seems like a suspicious move to me; one that may lead intuition astray.

I seem to recall that Bruno claimed this is a "legal" move because any possible input/output can be encoded as data within the computation (or something along those lines.

Yes. Eventually it comes to decide what is your "generalized" brain. If you need the entire physical universe, with 10^100 decimals, that will change nothing in the reasoning, because in step seven, your state will still be accessed.

Of course, the entire physical universe also has no input nor output (by definition of "entire").

For the six first steps, it is easier to assume some high substitution (neuronal) for the thought experiment. Then in step 7, this "high level" assumption is eliminated.

No doubt Bruno will be able to explain much better than me).

I have tried to talk in English. Now the fact that we can put the input in the code is a fundamental theorem for the universal system, know as the SMN theorems. In terms of the phi_i it means that there is one function S of two arguments with

phi_i(x) = phi_S(x, 4)()     (S10)


phi_i (4, y, z) = phi_S(x, 4) (y, z)   (S32)

The meta-program "S" take the input (4), and put it in the code, and suppress one variable.

For example S(4, "READ x, READ y, output x + y") = "Read Y, output 4 + y".

S is really a substitution.

S is a program, so it exists a number s such that S = phi_s. You can use this to see that we can write the SMN theorems with quantifying only on numbers.

The whole of recursion theory can be based axiomatically on the two axioms:

- SMN "theorem" (here an axiom, "provable" for all reasonable programming languages, or universal system) - It exists u such that phi_i(x) = phi_u(i, x) (existence of a universal number) (again provable for each individual programming language). The universal function u computes phi_i(x), for any program i and any data x.

But I guess that here, I do not explain better than you, as I use "notation", which frighten the beginners or the non mathematicians.

Yet, we need the SMN theorem to explain the Dx = "xx" method (to define self-reference in arithmetic) in terms of the phi_i and the w_i (which I promised to do for you!)

But we might need to revise a bit those phi_i and w_i perhaps, but then I don't want to annoy you with too much technic either. What do you think? Also we started this on the FOAR list, would you like to continue this, and on which list? Take it easy. I know we are in an end of the year feast period :)



You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to