XPost: comp.theory, sci.logic, sci.math
On 3/22/2022 6:18 PM, Richard Damon wrote:
On 3/22/22 10:53 AM, olcott wrote:
On 3/22/2022 9:39 AM, Ben Bacarisse wrote:
olcott <NoOne@NoWhere.com> writes:
Yet you cannot point out a single error in my halting problem
refutation.
The mistakes have been pointed out so many times that it's reasonable to >>> assume you can't see them or are simply ignoring them. The latest
monster error is that if (as you claim) this is true of your Ĥ:
"Ĥ.qx maps ⟨Ĥ⟩ ⟨Ĥ⟩ to Ĥ.qn"
but
"H maps ⟨Ĥ⟩ ⟨Ĥ⟩ to H.qy"
then Ĥ is not even a Truing machine, let alone the specific Ĥ in Linz's >>> proof.
If the halt deciding criteria compares the finite strings of Turing
machine descriptions as its halt deciding basis then it will find that
H and the copy of H embedded at Ĥ.qx are not the identical (embedded_H
is a longer finite string) thus providing the basis for H to see that
embedded_H will transition to Ĥ.qn and halt.
Except that inside embedded_H IS an exact copy of H, which should match
the 'string compare' if you could possible built it.
Yes this is correct.
Ben's point was based on a lack of understanding the x86 machine
architecture where otherwise identical copies of machine code will have different machines addresses that can be used as halt deciding criteria.
Since H doesn't have a unique, or even a finite number of
representations, it makes it hard to try to 'string compare' to find
copies.
The Linz proof has identical copies, alternative proofs don't even
specify any copies.
--
Copyright 2021 Pete Olcott
Talent hits a target no one else can hit;
Genius hits a target no one else can see.
Arthur Schopenhauer
--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)