On 6/23/21 8:01 PM, olcott wrote:
> On 6/23/2021 6:13 PM, André G. Isaak wrote:
>> On 2021-06-23 14:07, olcott wrote:
>>> On 6/23/2021 12:29 PM, André G. Isaak wrote:
>>>> On 2021-06-22 13:17, olcott wrote:
>>>>> On 6/22/2021 1:38 PM, André G. Isaak wrote:
>>>>>> On 2021-06-22 11:05, olcott wrote:
>>>>>>> On 6/22/2021 11:47 AM, André G. Isaak wrote:
>>>>
>>>> <snip>
>>>>
>>>>> In the case of the simulation of P it is verified on the basis of
>>>>> the x86 execution trace of B, thus an established verified fact.
>>>>
>>>> Except that an execution trace doesn't 'verify' anything. No one
>>>> doubts that your program decides that P(P) == 0. But unless the
>>>> logic underlying your program is actually correct, this result tells
>>>> us nothing. And you haven't provided any sort of proof that the
>>>> logic underlying your program is actually correct.
>>>>
>>>
>>> Halting problem undecidability and infinitely nested simulation
>>>
>>> https://www.researchgate.net/publication/351947980_Halting_problem_undecidability_and_infinitely_nested_simulation
>>>
>>>
>>> Sure I have. When we look as page 3 and 4 of the above paper it is
>>> very easy to see that H does faithfully simulate the execution of P
>>> and that P has no escape from infinite execution in its own code.
>>
>> No. First, it *doesn't* faithfully simulate the code. If H(P, P) is
>> simulating P(P), then P(P) can't call the main H -- it can only invoke
>> a copy of H *within the simulation*.
>>
>
> It is actually and quite factually not a copy.
> I have had to say this hundreds of times now.

It has to be a copy, as H^ isn't the machine H it needs to use a copy.
That is just how Turing Machines work. Turing Machine P has EVERYTHING
it needs as its algorithm in its own code, and EVERYTHING in will use as
its input on its Tape. That code and Tape is distinct from every other
code and tape in the mathematical universe (but their can be copies).
>
> It is the same freaking machine language at the
> same freaking machine language address.

But Linz didn't write his Theorem of x86 assembly language, he wrote it
for Turing Machines.

>
>> Second, even if we ignore the above, the decision criterion you use is
>> flawed because it completely ignores all conditional branches
>> contained within P's copy of H, and it's these conditional branches
>> which prevent P(P) from being an 'infinitely recursive' computation.
>>
>
> Unless we keep the independent variable separate from the dependent
> variable our analysis is incorrect.
>
> The independent variable is the cause. Its value is independent of other
> variables in your study.
>
> The dependent variable is the effect. Its value depends on changes in
> the independent variable.

So the independent variable is H. H is the CAUSE of all the behavior (P
is formed by a fixed algorithm form H). The dependent variable is the
halting behavior of P.

Looking at each possible H, we can see what it does to the P that is
built on it, and what that P actually does, and in EVERY case we find
that H is wrong.

>
> https://www.scribbr.com/methodology/independent-and-dependent-variables/
>
> When we are asking whether or not H must abort its simulation of P we
> examine the simulation of P. We must not examine the behavior of H.

>
>> Once again, traces do not constitute proofs. If you want to be taken
>> seriously, you need to provide some actual *proof* that your algorithm
>> works.
>
> That my proof does not conform to academic conventions does not show
> that it is not a proof. As long as I start with a set of premises that
> can be verified as true and my conclusion logically follows from these
> true premises then I definitely have a proof even if everyone in the
> universe disagrees.
>

It doesn't conform to logical validity.

Yes, start with a set of premisis that are verified to be true, and then
use them within the bounds of their meaning.

Your Premise (1) has validity under one set of definitions of its term,
that is true, but you don't use it with those definitiosn, thus you
don't start with valid premises.

The key fact is that 'its simulation' refers to the specific instance of
the simulation we are looking at, and the need test can be done by
changing JUST THAT INSTANCE, and not other 'copies' of it running in
other instances.

Premise(2) attempts to validate Premise(1) but uses a DIFFERENT meaning
of the words, for which Premise(1) is not true under.

Thus, your logic has failed, and can with the audacity that you do it,
could even be called deciteful.

>>> Premise(1) Every computation that never halts unless its simulation
>>> is aborted is a computation that never halts. This verified as true
>>> on the basis of the meaning of its words.
>>>
>>> Premise(2) The simulation of the input to H(P,P) never halts without
>>> being aborted is a verified fact on the basis of its x86 execution
>>> trace. (shown below).
>>>
>>> Conclusion(3) From the above true premises it necessarily follows
>>> that simulating halt decider H correctly reports that its input:
>>> (P,P) never halts.
>>>
>>> Only by very diligently making sure to not pay attention is it
>>> possible to maintain the ruse of a fake rebuttal.
>>>
>>>> And it isn't correct. Most significantly, your trace skips over
>>>> certain portions of the P being simulated which completely
>>>> invalidates its results.
>>>>
>>>
>>> The question is:
>>>
>>> Does the simulation of P have to be aborted to prevent its infinite
>>> execution?
>>>
>>> As long as the answer to this question is [YES] then Premise(2) is
>>> perfectly satisfied and your objection is utterly pointless.
>>
>> You should cut it out with the asinine repetition.It just makes you
>> look like a three-year old.
>>
>
> If I don't get ridiculous with the repetition in a single reply then I
> need twenty replies to make the same point because this point is
> perpetually ignored. I usually don't repeat the same sentence five times
> until after a key point has been ignored in several replies.
>

Maybe you shouldn't ignore the points that others are making. The fact
is that you keep on insisting on things that just are not true, and for
which you don't even try to present a justification for but just keep
claiming that they are obviously true, when they aren't.

If the actually ARE obviously true, then you should be able to actually
prove this, but you don't even try. Your don't even look at the
rebuttals and even try to point out logica errors in them.

>> And no, the objection is *not* pointless. You can't claim that there
>> is "no escape from infinite execution" when you ignore the conditional
>> statements which would allow P(P) to abort the simulation it is
>> conducting. Those are the conditional branches which provide that escape.
>>
>> André
>>
>
> The question is whether or not H must abort its simulation of P.
> The behavior of H is totally irrelevant to this question.

NO. That is a root of your problem. The Question is does P halt when
given input I. THAT is the question, and the P that we look at is
dependent on your definition of H.

The question of does H need to abort P may be a useful question for
developing the algorithm of H, but it isn't the ultimate question that H
is supposed to be answering. It is just a step in the path to the answer.

>
> Of the two hypothetical possibilities: (independent variable)
> (1) H aborts the simulation of P
> (2) H never aborts the simulation of P
>
> We ask what is the effect on the behavior of P? (dependent variable)
>
> (1) P halts
> (2) P never halts
>
> The above answer applies to every possible encoding of any simulating
> halt decoder H when applied to P or any computational equivalent to P
> such as the Linz ⟨Ĥ⟩.
>

But remember, the behavior of P is the DEPENDENT variable, each answer
corresponds to a given H.

Yes, if H doesn't abort it's P, then that P is non-halting, but that H
doesn't answer either so it fails to be a decider.

Change the independent variable, and you get a DIFFFERNT P, so you can't
just assume the same result.

If H does abort the simulation, and behaves the way you have described
(claiming non-halting) then P Halts. H has NO basis to use the other
case for its decision, as that is a different P, remember, P is a
dependent variable, so needs to be looked at fresh for each different H.