home bbs files messages ]

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

   comp.ai.philosophy      Perhaps we should ask SkyNet about this      59,235 messages   

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

   Message 58,080 of 59,235   
   dbush to olcott   
   Re: Never any actual rebuttal to HHH(DD)   
   20 Oct 25 23:05:09   
   
   [continued from previous message]   
      
    > "minor remark" becomes true by being a hypothetical: if the moon is made   
    > of cheese, the Martians can look forward to a fine fondue.  But,   
    > personally, I think the professor is more straight talking than that,   
    > and he simply took as a method that can work for some inputs.  That's   
    > the only way is could be seen as a "minor remark" with being accused of   
    > being disingenuous.   
      
   On 8/23/2024 9:10 PM, Mike Terry wrote:   
    > So that PO will have no cause to quote me as supporting his case:  what   
    > Sipser understood he was agreeing to was NOT what PO interprets it as   
    > meaning.  Sipser would not agree that the conclusion applies in PO's   
    > HHH(DDD) scenario, where DDD halts.   
      
   On 5/2/2025 9:16 PM, Mike Terry wrote:   
    > PO is trying to interpret Sipser's quote:   
    >   
    > --- Start Sipser quote   
    >       If simulating halt decider H correctly simulates its input D   
    >       until H correctly determines that its simulated D would never   
    >       stop running unless aborted then   
    >   
    >       H can abort its simulation of D and correctly report that D   
    >       specifies a non-halting sequence of configurations.   
    > --- End Sipser quote   
    >   
    > The following interpretation is ok:   
    >   
    >      If H is given input D, and while simulating D gathers enough   
    >      information to deduce that UTM(D) would never halt, then   
    >      H can abort its simulation and decide D never halts.   
    >   
    > I'd say it's obvious that this is what Sipser is saying, because it's   
    > natural, correct, and relevant to what was being discussed (valid   
    > strategy for a simulating halt decider).  It is trivial to check that   
    > what my interpretation says is valid:   
    >   
    >     if UTM(D) would never halt, then D never halts, so if H(D) returns   
    >     never_halts then that is the correct answer for the input.  QED  :)   
      
   --- 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