XPost: comp.theory, sci.logic   
   From: acm@muc.de   
      
   [ Followup-To: set ]   
      
   In comp.theory olcott  wrote:   
   > On 7/6/2025 5:16 AM, Alan Mackenzie wrote:   
   >> olcott  wrote:   
   >>> On 7/5/2025 2:07 PM, Alan Mackenzie wrote:   
      
   >>>> You lie.  You don't have a proof.  Many people in this group have pointed   
   >>>> out lots of errors in various versions of your purported proof, which you   
   >>>> just ignore.  The section in Professor Linz's book you used to be so fond   
   >>>> of citing will contain plenty of details, if only you would take the   
   >>>> trouble to understand it (assuming you're capable of such understanding).   
      
   >>> I have addressed ....   
      
   >> Meaningless pompous word.   
      
   >>> .... all of those details that you make sure to ignore so that you can   
   >>> baselessly claim that I am wrong.   
      
   >> I vaguely remember rolling my eyes at your hopeless lack of   
   >> understanding.  It was like watching a 7 year old trying to do calculus.   
   >> The basic understanding was simply not there.  Years later, it's still   
   >> not there.   
      
   >> And yes, you are wrong.  The proofs of the halting theorem which involve   
   >> constructing programs which purported halting deciders cannot decide   
   >> correctly are correct.   
      
   > Yet you cannot point to even one mistake because there are none.   
      
   That's what I'm saying.  Those proofs of the halting theorem are free   
   from mistakes.   
      
   More to the point, it is YOU who cannot point to any mistakes in them.   
   They are valid proofs.  Your work, if it contradicts those proofs (which   
   isn't at all clear) can thus be dismissed without further consideration.   
      
   >>> There cannot possibly be *AN ACTUAL INPUT* that does the   
   >>> opposite of whatever its decider decides. All of the examples   
   >>> of this have never been *ACTUAL INPUTS*   
      
   >> That's so sloppily worded, it could mean almost anything.   
      
   > The standard halting problem proof cannot even be constructed.   
      
   It has been constructed, and is valid.  But one would normally talk about   
   formulating a proof, rather than constructing one.   
      
   [ .... ]   
      
   >>> No Turing machine can possibly take another directly executing   
   >>> Turing machine as in input, thus removing these from the   
   >>> domain of every halt decider.   
      
   >> And that, too.   
      
   >>> *Thus the requirement that HHH report on the behavior*   
   >>> *of the directly executed DD has always been bogus*   
      
   >> And that makes your hat trick.   
      
   >>> Turing machine partial halt deciders compute the mapping   
   >>> from their actual inputs to the actual behavior that these   
   >>> inputs specify.   
      
   >> And a fourth.  There's some semblance of truth in there, but it's very   
   >> confused.   
      
      
   > It is not at all confused. I know exactly what it means.   
      
   It's very confused to everybody but you, then.   
      
   >> Sloppy wording is your technique to get people to go down to your level   
   >> of discussion.  That involves many posts trying just to tie you down to   
   >> specific word meanings, and is very tiresome and unrewarding.  I decline   
   >> to get involved any further.   
      
      
   > *Yet as I claimed you found no actual mistake*   
      
   I've found plenty of actual mistakes.  I was a software developer by   
   profession.   
      
   > Let me tell you the punchline so that you can   
   > see why I said those things.   
      
   Despite what I said last post, I will actually go to the trouble of   
   analysing your sloppy expression.   
      
   > Because directly executed Turing machines cannot   
   > possibly be inputs to Turing machine deciders this   
   > makes them outside of the domain of these deciders.   
      
   It's entirely unclear what a "directly executed Turing machine" is.  Most   
   of the time turing machines are theoretical constructs used for proving   
   theorems.  They can be executed, but rarely are.   
      
   It's unclear what you mean by a turing machine being an input to a turing   
   machine.  Read up about universal turing machines to get a bit of   
   background.   
      
   > When a partial halt decider is required to report   
   > on the direct execution of a machine this requirement   
   > is bogus.   
      
   See above.  That paragraph is meaningless.   
      
   > This means that the behavior of DD() is none of the damn   
   > business of HHH, thus does not contradict HHH(DD)==0.   
   > *If you disagree this only proves that you do not understand*   
      
   It's fully obscure what DD() and HHH mean, and thus impossible to   
   affirm or contradict the meaningless "HHH(DD)==0".   
      
   > HHH(DD) does correctly detect that DD simulated by HHH   
   > according to the semantics pf the C programming language   
   > cannot possibly reach its own "return"statement final   
   > halt state.   
      
   See above.  By the way, people concerned with computation theory use   
   turing machines, which are well-defined, simple, and powerful.  They lack   
   the complexity, ambiguity, and unsuitability for theoretical work of real   
   world programming languages like C.   
      
   > *If you disagree this only proves that you do not understand*   
      
   > Any mindless idiot can disagree. Showing an error and proving   
   > that it is an actual mistake requires much more than this.   
      
   Indeed.  All you have done is disagree with one of the proofs of the   
   halting theorem.  You have yet to show an error in it.  That will be   
   difficult, because there aren't any.   
      
   > --   
   > Copyright 2025 Olcott "Talent hits a target no one else can hit; Genius   
   > hits a target no one else can see." Arthur Schopenhauer   
      
   --   
   Alan Mackenzie (Nuremberg, Germany).   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)