XPost: comp.theory, sci.math   
   From: polcott333@gmail.com   
      
   On 12/18/2025 6:54 PM, dart200 wrote:   
   > On 12/18/25 4:45 PM, polcott wrote:   
   >> On 12/18/2025 6:39 PM, dart200 wrote:   
   >>> On 12/18/25 4:35 PM, Chris M. Thomasson wrote:   
   >>>> On 12/17/2025 11:13 PM, dart200 wrote:   
   >>>>> On 12/17/25 10:17 PM, Chris M. Thomasson wrote:   
   >>>>>> On 12/17/2025 10:14 PM, dart200 wrote:   
   >>>>>>> On 12/17/25 10:04 PM, Chris M. Thomasson wrote:   
   >>>>>>>> On 12/17/2025 9:06 PM, Richard Damon wrote:   
   >>>>>>>> [...]   
   >>>>>>>>> While no halt deciders exist, the "inteface that they define"   
   >>>>>>>>> does.   
   >>>>>>>>   
   >>>>>>>> For a given program, the creator can create a halting decider   
   >>>>>>>> for its logic. Does this path halt or not, well, we know, we   
   >>>>>>>> made the damn thing. But, there is no one universal decider...   
   >>>>>>>>   
   >>>>>>>> [...]   
   >>>>>>>   
   >>>>>>> how do we know that if there is no algo to compute that???   
   >>>>>>>   
   >>>>>>   
   >>>>>> If you create a program you know what paths its going down. If you   
   >>>>>> call into other programs, well, shit happens. But your own logic,   
   >>>>>> you know what you are doing when you made the program.   
   >>>>>   
   >>>>> w/e, i think "your own logic" is following an algo we just haven't   
   >>>>> written down because "muh undecidability"   
   >>>>>   
   >>>>   
   >>>> There can be a specialized halting decider for a program, but not   
   >>>> one decider for any program.   
   >>>   
   >>> thank you for repeating the consensus that has been repeated at me   
   >>> 100s of times now   
   >>>   
   >>   
   >> It is best to think of most of them as not very smart bots.   
   >> On the other hand Claude AI is so fabulous at semantic   
   >> entailment from a correct basis that I got it to override   
   >> its own bias and admit its own mistake this way. It was   
   >> actually capable of introspection.   
   >   
   > bro convincing an AI to say what u want is kinda trivial,   
      
   I got forty pages of pushback before I could rephrase   
   my point from first principles derived from standard   
   definitions.   
      
   > other people   
   > are not, and i've learned a lot more in responding to real people vs AI   
   >   
      
   I have not actually learned anything from anyone   
   here except the most important notion of a computable   
   function.   
      
   Besides that everyone has consistently remained as a   
   broken record of "you are wrong and we can't show why".   
      
   > and i don't feel like having that debate, so we can just agree to   
   > disagree on that point   
   >   
   >>   
   >>> also there are semantic paradoxes that no program could decide upon   
   >>>   
   >>   
   >   
      
      
   --   
   Copyright 2025 Olcott

   
      
   My 28 year goal has been to make 
   
   "true on the basis of meaning expressed in language"
   
   reliably computable.

   
      
   This required establishing a new foundation
   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)