XPost: comp.theory, sci.math, comp.ai.philosophy
From: polcott333@gmail.com
On 12/26/2025 7:41 PM, Richard Damon wrote:
> On 12/26/25 8:17 PM, olcott wrote:
>> On 12/26/2025 4:29 PM, Richard Damon wrote:
>>> On 12/26/25 1:07 PM, olcott wrote:
>>>> On 12/26/2025 11:26 AM, Richard Damon wrote:
>>>>> On 12/26/25 12:18 PM, olcott wrote:
>>>>>> On 12/26/2025 11:07 AM, Richard Damon wrote:
>>>>>>> On 12/26/25 11:56 AM, olcott wrote:
>>>>>>>> On 12/26/2025 10:24 AM, Richard Damon wrote:
>>>>>>>>> On 12/26/25 10:20 AM, olcott wrote:
>>>>>>>>>> On 12/26/2025 9:05 AM, Richard Damon wrote:
>>>>>>>>>>> On 12/26/25 8:54 AM, olcott wrote:
>>>>>>>>>>>> On 12/26/2025 6:59 AM, Richard Damon wrote:
>>>>>>>>>>>>> On 12/25/25 11:51 PM, olcott wrote:
>>>>>>>>>>>>>> On 12/25/2025 10:32 PM, Richard Damon wrote:
>>>>>>>>>>>>>>> On 12/25/25 10:37 PM, olcott wrote:
>>>>>>>>>>>>>>>> On 12/25/2025 9:17 PM, Richard Damon wrote:
>>>>>>>>>>>>>>>>> On 12/25/25 10:12 PM, olcott wrote:
>>>>>>>>>>>>>>>>>> Three different LLMs have been totally convinced
>>>>>>>>>>>>>>>>>> a total of 50 times, you just don't understand.
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>> LLM LIE, so are not reliable sources.
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> *Anyone that disagrees with this is not telling the truth*
>>>>>>>>>>>>>>>> "Any result that cannot be derived as a pure function
>>>>>>>>>>>>>>>> of finite strings is uncomputable."
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>> But Halting *IS* a "pure function of finite strings"
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>> And it is uncomputable
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> Not exactly. Usually ⟨M⟩ simulated by H == UTM(⟨M⟩)
>>>>>>>>>>>>>> Sometimes ⟨M⟩ simulated by H != UTM(⟨M⟩)
>>>>>>>>>>>>>
>>>>>>>>>>>>> Only if H doesn't CORRECTLY simulate (M).
>>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> Correctly simulated is defined by the semantics
>>>>>>>>>>>> of C applied to the finite string input for
>>>>>>>>>>>> the N steps until H sees the repeating pattern.
>>>>>>>>>>>
>>>>>>>>>>> So, how does that differ from what the program actually does?
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Ah great this is the first time that you didn't
>>>>>>>>>> just dodge that out of hundreds of times.
>>>>>>>>>>
>>>>>>>>>> When-so-ever an input finite string ⟨M⟩ does not
>>>>>>>>>> cheat and call its own decider the input finite
>>>>>>>>>> string to H(⟨M⟩) is a valid proxy for UTM(⟨M⟩).
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>> So, you didn't answer the question.
>>>>>>>>>
>>>>>>>>> How does H CORRECTLY simulate the input and get a different
>>>>>>>>> result from what the program does?
>>>>>>>>>
>>>>>>>>
>>>>>>>> The finite string P H
>>>>>>>> is not a valid proxy to UTM(P).
>>>>>>>
>>>>>>> So, you don't understand that a string is a string and you can
>>>>>>> copy it elsewhere?
>>>>>>>
>>>>>>
>>>>>> There is a key semantic difference between a finite
>>>>>> string that describes behavior and the exact sequence
>>>>>> of steps that a finite string input specifies to a
>>>>>> specific instance of a decider.
>>>>>
>>>>> Really?
>>>>>
>>>>> And why is that?
>>>>>
>>>>> Since the DEFINITION of semantics for strings representing programs
>>>>> is the operation of that program.
>>>>>
>>>>> Note, the string represents what it represents to EVERYTHING.
>>>>>
>>>>
>>>> That definition has always been less than 100%
>>>> precisely accurate even when one takes the vague
>>>> term: "represents" with a more precise term of
>>>> the art-meaning.
>>>
>>> Nope, nothing can be more accurate than the actual definition,
>>>
>>> You got a source for your claim, or is this just another lie out of
>>> your insanity.
>>>
>>>>
>>>> I simply bypass all of that by defining the new
>>>> idea of the sequence of steps that a finite string
>>>> input instance specifies to its decider instance.
>>>
>>> But you don't GET to define the new idea, not without admitting you
>>> are leaving Computation Theory.
>>>
>>> All you are doing is admitting that you logic is built on lying,
>>>
>>>>
>>>> That is a level of precision that no one bothered
>>>> to think about for 90 years. That this level of
>>>> detail is empirically proven to make an actual
>>>> difference conclusively validates it.
>>>
>>> No, your level of stupidity, thinking you get to redefine thing, is
>>> show to be great.
>>>
>>> As I said, All you have done is proved that you don't know what you
>>> are talking about, but are just making up lies.
>>>
>>> If you can't prove your claim in the system, from the defined
>>> definition, your claims are just admitted lies.
>>>
>>
>> Turing machine deciders: Transform finite string
>> inputs by finite string transformation rules into
>> {Accept, Reject} values.
>>
>> All that I am doing is exploring the exact details
>> of that. That no one else bothered to explore these
>> exact details is no mistake of mine.
>>
> So, your H / HH / HHH can be halt decider, at least if you define them
> in a way that meets the requirements, which your code doesn't, since
> their transform depends on hidden state.
>
Implementation details are irrelevant to theoretical limits.
H does apply finite string transformation rules to its
input P deriving {Reject}.
Four LLM systems have now fully agreed with all of
my reasoning about the general subject of undecidability.
ChatGPT and Claude AI have agreed in fresh brand new
conversations a dozen times each.
It initially took them fifty pages of dialogue to get it.
I am now down to 15 pages on each system.
It is not that these LLM systems are terribly faulty.
It is that conventional wisdom about undecidability
across computer science , math and logic is a foundational
error.
> The problem is you forget to define what it means to be a Halt Decider,
> or any form of XXXX Decider. Your problem is "Halting" is defined as a
> property of the actual machine being talked about, which can be
> expressed in terms of a UTM processing the string representation of it.
>
> You then get this crasy idea (which is just a lie) that you can just
> ignore the behavior of the CORRECT simulation of that input, as shown by
> what the UTM does, and try to define it's incorrect simulation (since it
> just stops short based on its own error) as being correct.
>
> And then, you show your problem by just refusing to even try to answer
> with a justification on why your idea is correct.
>
> How can your H have "Correctly Simuated" and input that "Correctly
[continued in next message]
--- SoupGate-DOS v1.05
* Origin: you cannot sedate... all the things you hate (1:229/2)
|
