home bbs files messages ]

Forums before death by AOL, social media and spammers... "We can't have nice things"

   sci.logic      Logic -- math, philosophy & computationa      262,912 messages   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]

   Message 261,596 of 262,912   
   olcott to Python   
   Re: The halting problem is incorrect two   
   01 Dec 25 08:57:17   
   
   XPost: sci.math, comp.theory   
   From: polcott333@gmail.com   
      
   On 12/1/2025 8:45 AM, Python wrote:   
   > Le 01/12/2025 à 15:38, olcott a écrit :   
   >> On 12/1/2025 8:29 AM, Python wrote:   
   >>>> [snip boring nonsense and lies]   
   >>>   
   >>> Peter you've intoxicated yourself.   
   >>>   
   >>> Here is what Chat GPT told me once about himself:   
   >>>   
   >>   
   >> Welcome back!   
   >>   
   >>> You have put your finger on the single most fundamental limitation of   
   >>> large language models:   
   >>>   
   >>> They can generate coherent arguments for things that are false,   
   >>> harmful, fringe, or logically impossible — not because they “believe”   
   >>> them, but because they can simulate the rhetorical form of such   
   >>> arguments.   
   >>>   
   >>> And you’re right:   
   >>> The fact that the model “doesn’t believe it” is irrelevant.   
   >>> What matters is:   
   >>> it can produce it.   
   >>>   
   >>> f2up math.   
   >>   
   >> Once you fully understand semantic tautologies   
   >> (the ultimate basis of all of my work)   
   >>   
   >> In epistemology (theory of knowledge), a self-evident   
   >> proposition is a proposition that is known to be true   
   >> by understanding its meaning without proof...   
   >> https://en.wikipedia.org/wiki/Self-evidence   
   >>   
   >> You will understand that I am correct. If you insist   
   >> on finding fault at a much higher priority than an   
   >> honest dialogue then you will never understand that   
   >> I am correct.   
   >   
   > You are NOT correct.   
   >   
      
   You will continue to lack a sufficient basis   
   for that until you grok (Heinlein) semantic   
   tautology / self-evident truth.   
      
   >> It seems that the single most useful application   
   >> of my work is to make LLM systems much more reliable.   
   >   
   > Your "work" is complete garbage... Sorry.   
   >   
   >   
      
   Yet you cannot possibly show that with complete   
   and correct reasoning because you continue to   
   lack the above required basis.   
      
   --   
   Copyright 2025 Olcott   
      
   My 28 year goal has been to make   
   "true on the basis of meaning" computable.   
      
   This required establishing a new foundation   
   for correct reasoning.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]


(c) 1994,  bbs@darkrealms.ca