Re: Unconventional partial halt decider and grounding to a truthmaker

On 5/18/24 1:11 PM, olcott wrote:

On 5/18/2024 3:23 AM, Mikko wrote:

On 2024-05-17 17:01:23 +0000, olcott said:
>

On 5/17/2024 5:45 AM, Mikko wrote:

On 2024-05-16 14:48:21 +0000, olcott said:
>

On 5/16/2024 5:42 AM, Mikko wrote:

On 2024-05-15 15:06:26 +0000, olcott said:
>

On 5/15/2024 3:06 AM, Mikko wrote:

On 2024-05-14 14:32:26 +0000, olcott said:
>

On 5/14/2024 4:44 AM, Mikko wrote:

On 2024-05-12 15:58:02 +0000, olcott said:
>

On 5/12/2024 10:21 AM, Mikko wrote:

On 2024-05-12 11:34:17 +0000, Richard Damon said:
>

On 5/12/24 5:19 AM, Mikko wrote:

On 2024-05-11 16:26:30 +0000, olcott said:
>

I am working on providing an academic quality definition of this
term.

>
The definition in Wikipedia is good enough.
>

>
I think he means, he is working on a definition that redefines the field to allow him to claim what he wants.

>
Here one can claim whatever one wants anysay.
In if one wants to present ones claims on some significant forum then
it is better to stick to usual definitions as much as possible.
>

Sort of like his new definition of H as an "unconventional" machine that some how both returns an answer but also keeps on running.

>
There are systems where that is possible but unsolvable problems are
unsolvable even in those systems.
>

>
When Ĥ is applied to ⟨Ĥ⟩
Ĥ.q0 ⟨Ĥ⟩ ⊢* embedded_H ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥ.qy ∞
Ĥ.q0 ⟨Ĥ⟩ ⊢* embedded_H ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥ.qn

>
This notation does not work with machines that can, or have parts
that can, return a value without (or before) termination.

>
00 int H(ptr x, ptr x)  // ptr is pointer to int function
01 int D(ptr x)
02 {
03   int Halt_Status = H(x, x);
04   if (Halt_Status)
05     HERE: goto HERE;
06   return Halt_Status;
07 }
08
09 int main()
10 {
11   H(D,D);
12 }

>
That notation is not any better for the purpose.
>

>
I refer to transitioning through a specific state to indicate
a specific halt status value, for Turing Machines.

>
That does not satisfy the usual definition of "halt decider".

>
Yet it <is> an incremental improvement over both YES and NO are
the wrong answer for input D. YES <is> the correct answer and H
can not SAY this answer in the conventional way.

>
For every computation "yes" is the correct answer if and only if one can
construct a finite sequence of configurations so that the first one is the
initial configuration, each other one follows from the previous one by a
transition rule, and no possible configuration follows from the last one
by any transition rule. If "yes" is not the correct answer then "no" is.
Therefore there is no D where neither "yes" and "no" is wrong for the
same input.
>

>
You are correct and I merely had a typo, I mean "NO" is the correct
answer if the above is not met, otherwise YES is the correct answer.

>
That obviously implies that there is no case where both "yes" and "no"
are right and there is no case where both "yes" and "no" are wrong.
>

What everyone gets confused about is that they disagree that:
a partial halt decider must determine its correct halt status decision on the basis of the actual behavior that its input actually specifies.

>
Who, other than you, has ever said otherwise?
>

 You are in the wrong forum, I am changing it to comp.theory
 Everyone besides me says that H must report on the behavior
of the computation that it is contained within this is both
impossible and incorrect.
 

No, H must report on the behavior of the program that its input represents, EVEN IF that program calls (a copy of) the decider its self.
The fact that this means that H needs to determine what it will do before it can decide what it will do is what gives it problems and make the job impossible, but that is what it is required to do.