Path: news.eternal-september.org!eternal-september.org!.POSTED!not-for-mail
From: Mikko <mikko.levanto@iki.fi>
Newsgroups: comp.theory
Subject: Re: How do simulating termination analyzers work?
Date: Mon, 23 Jun 2025 10:12:52 +0300
Organization: -
Lines: 107
Message-ID: <103aupj$13t8e$1@dont-email.me>
References: <102sjg5$2k3e9$1@dont-email.me> <1607e7860c899b930b87d371c747708dbeaf1062@i2pn2.org> <102t67r$2o80a$1@dont-email.me> <102u3et$31q0g$4@dont-email.me> <102ufv8$35emj$1@dont-email.me> <1030kqk$3pfor$1@dont-email.me> <10319mv$3u901$7@dont-email.me> <103394q$m26r$1@dont-email.me> <1033pf6$25t1$1@dont-email.me> <1035vdm$10d9c$1@dont-email.me> <1036qg0$16lpk$3@dont-email.me> <1038glb$e9bd$1@dont-email.me> <1039kq9$n1od$1@dont-email.me>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
Injection-Date: Mon, 23 Jun 2025 09:12:52 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="ff9eeb82472defb391a609f5929552bc";
	logging-data="1176846"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1/5UHhBfnn/vJDml4LBIwxr"
User-Agent: Unison/2.2
Cancel-Lock: sha1:9PrXGWUXJpX4P6l507SELCVqt30=

On 2025-06-22 19:16:24 +0000, olcott said:

> On 6/22/2025 3:59 AM, Mikko wrote:
>> On 2025-06-21 17:34:55 +0000, olcott said:
>> 
>>> On 6/21/2025 4:52 AM, Mikko wrote:
>>>> On 2025-06-20 13:59:02 +0000, olcott said:
>>>> 
>>>>> On 6/20/2025 4:20 AM, Fred. Zwarts wrote:
>>>>>> Op 19.jun.2025 om 17:17 schreef olcott:
>>>>>>> On 6/19/2025 4:21 AM, Fred. Zwarts wrote:
>>>>>>>> Op 18.jun.2025 om 15:46 schreef olcott:
>>>>>>>>> On 6/18/2025 5:12 AM, Fred. Zwarts wrote:
>>>>>>>>>> Op 18.jun.2025 om 03:54 schreef olcott:
>>>>>>>>>>> On 6/17/2025 8:19 PM, Richard Damon wrote:
>>>>>>>>>>>> On 6/17/25 4:34 PM, olcott wrote:
>>>>>>>>>>>>> void Infinite_Recursion()
>>>>>>>>>>>>> {
>>>>>>>>>>>>>    Infinite_Recursion();
>>>>>>>>>>>>>    return;
>>>>>>>>>>>>> }
>>>>>>>>>>>>> 
>>>>>>>>>>>>> void Infinite_Loop()
>>>>>>>>>>>>> {
>>>>>>>>>>>>>    HERE: goto HERE;
>>>>>>>>>>>>>    return;
>>>>>>>>>>>>> }
>>>>>>>>>>>>> 
>>>>>>>>>>>>> void DDD()
>>>>>>>>>>>>> {
>>>>>>>>>>>>>    HHH(DDD);
>>>>>>>>>>>>>    return;
>>>>>>>>>>>>> }
>>>>>>>>>>>>> 
>>>>>>>>>>>>> When it is understood that HHH does simulate itself
>>>>>>>>>>>>> simulating DDD then any first year CS student knows
>>>>>>>>>>>>> that when each of the above are correctly simulated
>>>>>>>>>>>>> by HHH that none of them ever stop running unless aborted.
>>>>>>>>>>>> 
>>>>>>>>>>>> WHich means that the code for HHH is part of the input, and thus there 
>>>>>>>>>>>> is just ONE HHH in existance at this time.
>>>>>>>>>>>> 
>>>>>>>>>>>> Since that code aborts its simulation to return the answer that you 
>>>>>>>>>>>> claim, you are just lying that it did a correct simulation (which in 
>>>>>>>>>>>> this context means complete)
>>>>>>>>>>>> 
>>>>>>>>>>> 
>>>>>>>>>>> *none of them ever stop running unless aborted*
>>>>>>>>>> 
>>>>>>>>>> All of them do abort and their simulation does not need an abort.
>>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> *It is not given that any of them abort*
>>>>>>>>> 
>>>>>>>> 
>>>>>>>> At least it is true for all aborting ones, such as the one you 
>>>>>>>> presented in Halt7.c.
>>>>>>> 
>>>>>>> My claim is that each of the above functions correctly
>>>>>>> simulated by any termination analyzer HHH that can possibly
>>>>>>> exist will never stop running unless aborted by HHH.
>>>>>>> Can you affirm or correctly refute this?
>>>>> 
>>>>>> Yes, I confirmed many times that we can confirm this vacuous claim, 
>>>>>> because no such HHH exists. All of them fail to do a correct simulation 
>>>>>> up to the point where they can see whether the input specifies a 
>>>>>> halting program.
>>>>> 
>>>>> if DDD correctly simulated by any simulating termination
>>>>> analyzer HHH never aborts its simulation of DDD then
>>>> 
>>>> that HHH is not interesting.
>>> 
>>> *then the HP proofs are proved to be wrong*
>> 
>> No, they are not. You have not solved the halting problem and that
>> (in addition to all proofs) supports the claim that halting problem
>> is unsolvable.
> 
> ChatGPT corrected my words and agreed that I have
> correctly refuted the generic HP proof technique
> where an input has been defined to only do the
> opposite of whatever value that its decider decides.
> https://chatgpt.com/s/t_6857335b37a08191a077d57039fa4a76

Doesn't matter. Only proofs matter. So far you have not proven anything
and it is unlikely you could prove anything even after asking ChatGPT
for help.

> The ChatGPT that evaluated and affirmed my analysis
> of HHH(DDD) one year ago could only handle 4000 tokens
> thus could not understand HHH(DD).
> 
> ChatGPT with GPT-4-turbo — can handle up to 128,000 tokens
> of context in a single conversation, immediately understood
> HHH(DD) within the context of the conversation of HHH(DDD).

ChatGPT does not understand. Whether you do is still not determined.

Anyway,
>> In order to show that a proof is wrong you need to show an error
>> in the proof. Even then the conclusion is proven unless you can
>> show an error in every proof of that conclusion.

-- 
Mikko