Path: news.eternal-september.org!eternal-september.org!.POSTED!not-for-mail
From: "Fred. Zwarts" <F.Zwarts@HetNet.nl>
Newsgroups: comp.theory
Subject: Re: ChatGPT agrees that HHH refutes the standard halting problem
 proof method
Date: Wed, 2 Jul 2025 10:37:38 +0200
Organization: A noiseless patient Spider
Lines: 91
Message-ID: <1042r4j$a69o$1@dont-email.me>
References: <103jmr5$3h0jc$1@dont-email.me> <103k0sc$2q38$1@news.muc.de>
 <103k1mc$3j4ha$1@dont-email.me> <103lfn1$ml0$1@dont-email.me>
 <103m813$6dce$1@dont-email.me> <103ol2u$raq9$1@dont-email.me>
 <103onmp$rq7e$1@dont-email.me> <103r0ce$1esb9$1@dont-email.me>
 <103rhf6$1hc53$8@dont-email.me>
 <0c50a8ee4efb36cef4271674792a090125187f9d@i2pn2.org>
 <gPg8Q.1988877$4AM6.189428@fx17.ams4>
 <a60543ff9feb748df80b32970c67bb8c7ab13d89@i2pn2.org>
 <tJA8Q.6$r61e.2@fx11.ams4>
 <5e7f84c84b4ed51e195dd33afd9ed7eca89be454@i2pn2.org>
 <103vduo$2flaf$2@dont-email.me> <104067d$7a12$1@dont-email.me>
 <1040i6c$2ql69$4@dont-email.me>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Injection-Date: Wed, 02 Jul 2025 08:37:40 +0000 (UTC)
Injection-Info: dont-email.me; posting-host="94c83e0600917e76556ff17dba76f9bb";
	logging-data="334136"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1+LiuV/lKfYQdj+qULQceOw"
User-Agent: Mozilla Thunderbird
Cancel-Lock: sha1:kU00pguIPVhjPBLH940NQ7ekfhI=
Content-Language: nl, en-GB
In-Reply-To: <1040i6c$2ql69$4@dont-email.me>

Op 01.jul.2025 om 13:52 schreef olcott:
> On 7/1/2025 3:28 AM, Fred. Zwarts wrote:
>> Op 01.jul.2025 om 03:34 schreef olcott:
>>> On 6/30/2025 8:12 PM, Richard Damon wrote:
>>>> On 6/30/25 2:30 PM, Mr Flibble wrote:
>>>>> On Sun, 29 Jun 2025 22:39:10 -0400, Richard Damon wrote:
>>>>>
>>>>>> On 6/29/25 3:51 PM, Mr Flibble wrote:
>>>>>>> On Sun, 29 Jun 2025 15:00:35 -0400, Richard Damon wrote:
>>>>>>>
>>>>>>>> Remember, the simulator must be simulating the INPUT, and thus 
>>>>>>>> to go
>>>>>>>> past the call HHH instruction, the code must be part of the 
>>>>>>>> input, and
>>>>>>>> the input needs to be a constant.
>>>>>>> No. If HHH is simulating DDD then HHH can detect a call to itself 
>>>>>>> being
>>>>>>> passed DDD within DDD and can assert at that point that the input is
>>>>>>> non-
>>>>>>> halting.
>>>>>>>
>>>>>>> /Flibble
>>>>>>
>>>>>> And thus isn't simu;ating THE INPUT, and that the input isn't a 
>>>>>> PROGRAM.
>>>>>>
>>>>>> Also, what if DDD is using a copy of HHH, as per the proof program,
>>>>>> which might have variations in the code.
>>>>>>
>>>>>> Sorry, just shows you don't understand the problem.
>>>>>
>>>>> No. A simulator does not have to run a simulation to completion if 
>>>>> it can
>>>>> determine that the input, A PROGRAM, never halts.
>>>>>
>>>>> /Flibble
>>>>
>>>> Right, but the program of the input DOES halt.
>>>>
>>>
>>> The directly executed DDD() *IS NOT AN INPUT*
>>> Directly executed Turing machines have always been
>>> outside of the domain of any function computed by
>>> a Turing machine therefore directly executed Turing
>>> machines have never contradicted the decision of
>>> any halt decider.
>>>
>>> Halt deciders compute the mapping from the behavior
>>> that their finite string inputs actually specifies.
>>>
>>>
>>
>> The input is a pointer to a 'finite string' that includes the code of 
>> DDD and all functions called by it, in particular including the code 
>> to abort and halt.
> 
> *I keep correcting you on this and you keep ignoring my correction*

Trying to correct a correct statement is not very professional.

> The measure is DDD simulated by HHH reaching its simulated "return"
> statement final halt state. HHH continues to simulated DDD as a
> pure simulator (that also simulates itself simulating DDD) until
> HHH sees the non-terminating pattern.

As usual claims without evidence.
No, the measure for halting behaviour is reaching the end when executed 
without disturbance. Any disturbance, such as switching off the 
computer, or aborting the simulation (, or changing the initialisation 
such as when cheating with the Root variable) does not count.
Exactly the same input has been proven to specify a halting program by 
world-class simulators and direct execution.
Therefore, HHH is wrong when it thinks there is a non-terminating pattern.
There is no such pattern, there is only a finite recursion.
The reason for this failure of HHH is that it does not count its own 
conditional branch instructions.
You keep ignoring these corrections to your statements.

> 
>> Therefore, even a beginner can see that this input specifies a halting 
>> program. That your HHH cannot see that, does not change the 
>> specification. 
> 
> Five chatbots all agree that the input to HHH(DDD) specifies
> non-terminating recursive emulation even though DDD() halts.

Even more professional experts agree that halting behaviour is specified 
by the input and that HHH is wrong.
We know how easy it is to cheat with chatbots, so what counts more?