XPost: comp.theory, comp.software-eng   
   From: user7160@newsgrouper.org.invalid   
      
   On 1/15/26 7:28 PM, Richard Damon wrote:   
   > On 1/15/26 7:23 AM, dart200 wrote:   
   >   
   >> bro stick a giant dildo up ur asshole u hypocritical fuckface...   
   >>   
   >> when i tried to suggest improvements to the computational model, like   
   >> RTMs, u then told me i *can't* do that because muh ct-thesis, and here   
   >> u are crying about how no superior method has been found as if u'd   
   >> ever even tried to look past the ct-thesis...   
   >   
   > No, you didn't suggest improvements to the model, you just showed you   
   > don't knoww what that means.   
   >   
   > You don't get to change what a "computation" is, that isn't part of the   
   > "model".   
      
   you honestly could have just said that cause the rest of this is just u   
   repeating urself as if that makes it more correct   
      
   >   
   > The model would be the format of the machine, and while your RTM might   
   > be a type of machine that could be thought of, they don't do   
   > COMPUTATIONS, as it violates the basic rules of what a compuation IS.   
   >   
   > Computations are specific algorithms acting on just the input data.   
   >   
   > A fundamental property needed to reach at least Turing Complete ability,   
   > is the ability to cascade algorithms.   
   >   
   > Your RTM break that capability, and thus become less than Turing Complete.   
      
   i'm sorry, RTMs are literally just TMs with one added instruction that   
   dumps static meta-data + copies tape ... how have they *lost* power with   
   that??? clearly they can express anything that TMs can ...   
      
   >   
   > And, any algorithm that actually USES their capability to detect if they   
   > have been nested will become incorrect as a decider, as a decider is a   
   > machine that computes a specific mapping of its input to its output, and   
   > if that result changes in the submachine, only one of the answers it   
   > gives (as a stand-alone, or as the sub-machine) can be right, so you   
   > just show that it gave a wrong answer.   
      
   u have proof that doesn't work yet you keep asserting this is the "one   
   true way". seems like u just enjoy shooting urself in the foot, with the   
   only actual rational way being it's just the "one true way"   
      
   >   
   > This is sort of like the problem with a RASP machine architecture, sub-   
   > machines on such a platform are not necessarily computations, if they   
   > use the machines capability to pass information not allowed by the rules   
   > of a computation. Your RTM similarly break that property.   
   >   
   > Remember, Computations are NOT just what some model of processing   
   > produce, but specifically is defined based on producing a specific   
   > mapping of input to output, so if (even as a sub-machine) a specific   
   > input might produce different output, your architecture is NOT doing a   
   > computation.   
   >   
   > And without that property, using what the machine could do, becomes a   
   > pretty worthless criteria, as you can't actually talk much about it.   
      
   the output is still well-defined and deterministic at runtime,   
      
   context-dependent computations are still computations. the fact TMs   
   don't capture them is an indication that the ct-thesis may be false   
      
   --   
   arising us out of the computing dark ages,   
   please excuse my pseudo-pyscript,   
   ~ nick   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)