Re: The philosophy of computation reformulates existing ideas on a new basis --- getting somewhere

On 11/3/2024 12:20 PM, Richard Damon wrote:

On 11/3/24 9:39 AM, olcott wrote:

>
That is why I used to fully defined semantics of the x86
language to make this 100% perfectly unequivocal.
>
A few lines of x86 code express complex algorithms
succinctly enough that human minds are not totally
overwhelmed by far too much tedious detail.
>

It is not pspecified
in the usual formulation of the problem. Also note that
the behaviour exists before those strings so "describe"
should be and usually is used instead of "specify". The
use of latter may give the false impression that the behaviour
is determined by those strings.
>

>
In order for any machine to compute the mapping from
a finite string it must to so entirely on the basis
of the actual finite string and its specified semantics.

 You have that somewhat backwards. It *CAN* only do what it can compute.
 The mapping is not required to *BE* computable.
 

>
The finite string input to HHH specifies that HHH
MUST EMULATE ITSELF emulating DDD.

 Right, and it must CORRECTLY determine what an unbounded emulation of that input would do, even if its own programming only lets it emulate a part of that.
 

Yes this is exactly correct. I don't understand
why you keep disagreeing with your own self this.

>
The finite string input to HHH1 specifies that HHH1
MUST NOT EMULATE ITSELF emulating DDD.

 But the semantics of the string haven't changed, as the string needs to contain all the details of how the machine it is looking at will work.
 

DDD emulated by HHH specifies that HHH will
emulate itself emulating DDD.
DDD emulated by HHH1 specifies that HHH1 will
NOT emulate itself emulating DDD.

>
Unless HHH rejects its input DDD as non halting the
executed DDD never stops running. This itself proves
that HHH is correct and that DDD is not the same
instance as the one that HHH rejected.

 You have cause and effect backwards.
 

<MIT Professor Sipser agreed to ONLY these verbatim words 10/13/2022>
     If simulating halt decider H correctly simulates its input D
     until H correctly determines that its simulated D would never
     stop running unless aborted then...
The conditional branch instruction criteria has been met.
--
Copyright 2024 Olcott "Talent hits a target no one else can hit; Genius
hits a target no one else can see." Arthur Schopenhauer