XPost: comp.theory, comp.ai.philosophy   
   From: NoOne@nospicedham.NoWhere.com   
      
   On 12/16/2020 10:50 PM, Kaz Kylheku wrote:   
   > On 2020-12-17, olcott  wrote:   
   >> On 12/16/2020 10:24 PM, Kaz Kylheku wrote:   
   >>> On 2020-12-17, olcott  wrote:   
   >>>> On 12/16/2020 7:51 PM, Kaz Kylheku wrote:   
   >>>>> If the progression is caused by an infinite recursion of a function   
   >>>>> with the same paraemters, then all of the simulation levels are   
   >>>>> indistinguishable. No UTM knows how many levels are above,   
   >>>>> and each one has the same number of identical levels below.   
   >>>>>   
   >>>>> That's it.   
   >>>>>   
   >>>>   
   >>>> If we allowed the infinite recursion to infinitely recur this would be   
   >>>> true. We cut off the otherwise infinite recursion at three.   
   >>>>   
   >>>>> While it is possible to break that, by doing so you break the   
   >>>>> premise that we have a pure function.   
   >>>>>   
   >>>>   
   >>>> Halts() is a function of its inputs.   
   >>>   
   >>> Unfortunately "inputs" no longer refers to "those argument things   
   >>> between the parentheses that we are painstakingly keeping the same".   
   >>>   
   >>   
   >> I think that may be simply because you have a narrow minded view of   
   >> inputs as some fixed sized set of objects.   
   >   
   > Oh, I specifically do not. Above, I'm demonstrating flexibility and   
   > insight; I'm recognizing the new situation that there are more inputs   
   > than just the arguments to those function parameters, and I'm   
   > acknowleding that those are bona fide inputs.   
   >   
      
   Great !   
      
   > However, the proofs that are used to demonstrate the nonexistence of a   
   > universal halting aglorithms specify certain functions which have   
   > certain inputs, and only those inputs, and are generally pure.   
      
   The dynamically generated execution trace of the simulation of H_Hat and   
   its input are what Halts bases its halting decision on. That the   
   conventiopal proof may have incorectly assumed static analysis is merely   
   a false assumption on its part.   
      
   > In your coding simulation of those functions, you cannot be adding   
   > additional hidden inputs to the programming language functions   
   > which are supposed to correspond to the proof's functions.   
   >   
      
   The only thing that I actually need to make a correct halting decision   
   is the sequence of user specified x86 instructions that are simulated by   
   the halt decider. A simulator would have access to this sequence.   
      
   >> You are ruling out all dynamically generated inputs.   
   >   
   > Well, not me, personally, but the proofs.   
   >   
   > The halting proofs use certain functions and they specify what exactly   
   > those functions mean. You're changing what the functions mean, so your   
   > code execution produces situations that don't pertain to the original   
   > functions in the proof.   
   >   
      
   The halting proofs merely falsely assume static rather than dynamic   
   analysis. That is their mistake not mine.   
      
      
   --   
   Copyright 2020 Pete Olcott   
      
   "Great spirits have always encountered violent opposition from mediocre   
   minds." Einstein   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)