Path: news.eternal-september.org!eternal-september.org!feeder3.eternal-september.org!news.quux.org!news.nk.ca!rocksolid2!i2pn2.org!.POSTED!not-for-mail
From: Richard Damon <richard@damon-family.org>
Newsgroups: comp.theory
Subject: Re: How could HHH report on the behavior of its caller?
Date: Wed, 14 May 2025 23:14:10 -0400
Organization: i2pn2 (i2pn.org)
Message-ID: <3ef1340d9dc3af09f2aa683fd86689a54ea889a1@i2pn2.org>
References: <1003g30$2pfv4$1@dont-email.me>
 <fcabeaf49bf29636824f5e150e62094c446fc123@i2pn2.org>
 <1003lfm$2u41h$1@dont-email.me>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Injection-Date: Thu, 15 May 2025 03:14:36 -0000 (UTC)
Injection-Info: i2pn2.org;
	logging-data="402983"; mail-complaints-to="usenet@i2pn2.org";
	posting-account="diqKR1lalukngNWEqoq9/uFtbkm5U+w3w6FQ0yesrXg";
User-Agent: Mozilla Thunderbird
X-Spam-Checker-Version: SpamAssassin 4.0.0
Content-Language: en-US
In-Reply-To: <1003lfm$2u41h$1@dont-email.me>

On 5/14/25 11:02 PM, olcott wrote:
> On 5/14/2025 9:54 PM, Richard Damon wrote:
>> On 5/14/25 9:30 PM, olcott wrote:
>>> void DDD()
>>> {
>>>    HHH(DDD);
>>>    return;
>>> }
>>>
>>> int main()
>>> {
>>>    DDD();
>>> }
>>>
>>> If HHH cannot report on the behavior of its caller
>>> because this is a ridiculous requirement then how
>>> can HHH report on the direct execution of DDD()
>>> (AKA its caller).
>>>
>>
>> Because it is given the code of DDD, and thus doesn't need to know 
>> about "It caller"
>>
> 
> Unless it does know about its caller
> The only directly executed DDD() that actually exists
> no HHH can possibly know about any directly executed DDD().
> 
> 

So, how does it know that?

How does that knowledge affect the answer?

And no, it isn't the only directly executed DDD that exists, as programs 
are not "use once".

So, if your claim is that a Halt Decider just can't know the behavior of 
the direct execution of its input for any input, then you are just 
saying that the problme is logically impossible, and thus saying it 
can't be done is just a truism.

If you say it just can't handle this particular input, that is why a 
correct Halt Decider can't exist, we can make an input it can't determine.

Now, of course we can determine the behavior of any input, as that is 
what a UTM does, it recreates the full behavior from a description. The 
only proglem is a Halt Decider isn't supposed to reproduce the behavior, 
just decide on it, and thus can't just use simulation, which actually 
makes your concept broken.

THere is no such machine as a machine that both correct Halt Decides on 
its input and also correctly simulates that input, as the requirements 
are actually contradictory, so a "Simulating Halt Decider" based on 
being the correct simulator itself is just a category error.