Re: D correctly simulated by H proved for THREE YEARS --- rewritten

Am Thu, 13 Jun 2024 10:08:20 -0500 schrieb olcott:
> On 6/13/2024 9:35 AM, Alan Mackenzie wrote:
>> In comp.theory olcott <[email protected]> wrote:
>>> On 6/13/2024 3:49 AM, joes wrote:
>>>> Am Wed, 12 Jun 2024 18:25:14 -0500 schrieb olcott:
>>>>> On 6/12/2024 6:03 PM, Richard Damon wrote:
>>>>>> On 6/12/24 12:57 PM, olcott wrote:
>>>>>>> On 6/12/2024 6:33 AM, Richard Damon wrote:

>>>>> *H DOES NOT simply guess what you think it should do and do that*
Indeed, it must compute and it must be right.
>>>> It has a *specification* it must fulfill.
>> The specification is "calculate whether a particular program with a
>> particular input halts".
> Yes that <is> what the textbooks say, none-the-less halt deciders really
> cannot read textbooks. Instead they go by the behavior that their input
> specifies.
And their programming. And we can change which machine/program we are
looking at, if we've got one that isn't up to spec.

> All halt deciders compute the mapping from their inputs...
All Turing machines do.

> When we actually compute the mapping from the x86 machine language
> finite string input to H(D,D) using the finite string transformation
> rules specified by the semantics of the x86 programming language this
> DOES NOT MAP TO THE BEHAVIOR OF D(D).
Yes, which is why H is wrong.

> When I ask anyone to show the detailed steps of the mapping from the
> machine language finite string of D to the behavior of D(D) *THEY CHANGE
> THE SUBJECT BECAUSE THEY ALREADY KNOW THAT I AM CORRECT*
That mapping is provided by a simulator/UTM. You even use one yourself!
No need to guess the wrong answer.

-- 
joes