Re: HHH(DDD) is correct to reject its input as non-halting --- EVIDENCE THAT I AM CORRECT

Op 30.jun.2025 om 19:08 schreef olcott:

On 6/30/2025 2:40 AM, Fred. Zwarts wrote:

Op 29.jun.2025 om 15:46 schreef olcott:

On 6/29/2025 5:38 AM, Fred. Zwarts wrote:

Op 28.jun.2025 om 15:02 schreef olcott:

On 6/28/2025 3:50 AM, Fred. Zwarts wrote:

Op 28.jun.2025 om 01:30 schreef olcott:

On 6/26/2025 4:16 AM, Fred. Zwarts wrote:

Op 25.jun.2025 om 16:09 schreef olcott:

On 6/25/2025 2:59 AM, Fred. Zwarts wrote:

Op 24.jun.2025 om 16:06 schreef olcott:

>
None of the code in HHH can possibly cause DDD correctly
simulated by HHH to reach its own simulated "return" statement.

>
Yes, exactly, that is the bug.
>

>
Recursive emulation is only a tiny bit more complicated
than recursion yet no one here seems to have a clue.
Do you know what recursion is?
(If you don't that would explain a lot)

As usual irrelevant claims without evidence. No rebuttal.

>
Ah so you don't know what recursion is.

>
As usual a false claim without evidence.
>

>

HHH has a bug that makes that it does not recognise the halting behaviour of the program specified in the input.

>
If you don't even know what recursion is then
you are totally unqualified to review these things.

>
And since the condition in the 'if' fails, the conclusion is not true.
>

>

Even a beginner can see that the input is a pointer to code, including the code to abort and halt. But HHH is programmed to ignore the conditional branch instructions, when simulating itself, so it thinks that there is an infinite loop when there are only a finite number of recursions.
But Olcott does not understand that not all recursions are infinite.

>
When the measure is whether or not DDD correctly
simulated by HHH can possibly reach its own "return"
instruction final halt state nothing inside HHH can
possibly have any effect on this.
>
That you don't know this proves that you are unqualified
to review my work.

The failure of HHH is an incorrect measure for the halting behaviour specified in the input.
That you do not understand this explains your invalid claims.
The halting behaviour of the input can be analysed by several other methods and they show that HHH is incorrect in its analysis.
>
>

>
No Turing machine can ever report on the behavior of
any directly executing Turing Machine because no TM
can ever take another directly executing Turing Machine
as its input.

There is no need to report about another Turing Machine.

 The conventional halting problem proof incorrectly requires this.
 

As usual repeated claims without evidence.

void DDD()
{
   HHH(DDD);
   return;
}
 int main()
{
   HHH(DDD);
   DDD();
}
 When the input to HHH(DDD) is correctly simulated
by HHH then HHH correctly rejects this input as
specifying non-halting behavior.

No, it *incorrectly* does that. The input is DDD calling an aborting HHH, so the input specifies a halting program.

 The directly executed DDD() halts yet is not and cannot
be an input to HHH, thus it outside of the domain of HHH.

Irrelevant. It is only relevant that the input specifies a halting program.

 

 It only needs to report about its input. In this case the input includes the abort code and in this way specifies a halting program.

 You keep getting confused about the program under test here.
If HHH was the program under test the the internals of HHH
would be relevant.

You keep confused by combining the program under test with the input. Not the program under test is under test, but the code of the program under test is part of the input. The whole input is under test. This code cannot be replaced by other hypothetical code with different behaviour.

 Since DDD is the program under test and HHH remains a pure
simulator of DDD until HHH correctly determines that DDD
correctly simulated by HHH cannot possibly reach its own
simulated "return" statement final halt state, the internals
of HHH cease to be relevant.
 

And this behaviour to abort and halt, is part of the code given as input to the simulator.
That HHH has bugs to ignore that part of the code, does not change the fact that that part of the code is also under test, because it is used by DDD.