Re: Question for Olcott [ Richard continues to be a liar ]

0 views
Skip to first unread message

olcott

unread,
May 23, 2022, 7:50:36 PMMay 23
to
On 5/23/2022 6:33 PM, Richard Damon wrote:
>
> On 5/23/22 3:05 PM, olcott wrote:
>> On 5/23/2022 1:52 PM, Mr Flibble wrote:
>>> A simple multiple choice question for Olcott:
>>>
>>> All things being equal which is more likely:
>>>
>>> (a) Olcott is correct and everybody else is incorrect
>>> (b) Olcott is incorrect and everybody else is correct
>>>
>>> ?
>>>
>>> /Flibble
>>>
>>
>> Believability has the word [lie] embedded directly within itself.
>> Instead of the fake measure of credibility one must employ actual
>> validation.
>>
>> Actual validation conclusively proves that H(P,P)==0
>> is correct. This means that everyone that disagrees is either
>> insufficiently technically competent or a liar.
>
> You consider that H(P,P) == 0 is correct when P(P) halts, when that is
> the DEFINITION of what H(P,P) is supposed to be answering?
The C function H correctly determines that there are no number of steps
(0 to infinity) of its correct simulation of its input: a pair of
pointers to finite strings of x86 machine language that would ever reach
the last instruction of this input.



--
Copyright 2022 Pete Olcott

"Talent hits a target no one else can hit;
Genius hits a target no one else can see."
Arthur Schopenhauer
Reply all
Reply to author
Forward
0 new messages