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 58,669 of 59,235    |
|    olcott to Mikko    |
|    Re: on mathematical ghosts (2/3)    |
|    15 Dec 25 09:42:32    |
      [continued from previous message]              >>>>>>>>>> BB(n) is, by definitiion a "finite" number. Talking about the       >>>>>>>>>> "limit" of a finite number is a misuse of the term.       >>>>>>>>>       >>>>>>>>> i mean the natural number limit L >5 at which point BB(L)       >>>>>>>>> becomes fundamentally *unknowable* due to some L-state machine       >>>>>>>>> being fundamentally undecidable.       >>>>>>>>>       >>>>>>>>> if L doesn't exist, that would make halting generally       >>>>>>>>> decidable, so therefore L must exist       >>>>>>>>>       >>>>>>>>> if L does exist, then there must be some L-state machine U       >>>>>>>>> which cannot be decided on *by any partial* decider, because       >>>>>>>>> the BB computation would find it and use it       >>>>>>>>>       >>>>>>>>>>       >>>>>>>>>> We can sometimes establish upper and lower bounds on the value       >>>>>>>>>> of BB(n), is that what you mean by "a limit L"?       >>>>>>>>>>       >>>>>>>>>>>       >>>>>>>>>>> if you believe the halting problem, then BB must have a limit       >>>>>>>>>>> L, or else halting becomes generally solvable using the BB       >>>>>>>>>>> function. see, if you can compute the BB number for any N-       >>>>>>>>>>> state machines, then for any N-state machine u can just run       >>>>>>>>>>> the N- state machine until BB number of steps. any machine       >>>>>>>>>>> that halts on or before BB(N) steps halts, any that run past       >>>>>>>>>>> must be nonhalting       >>>>>>>>>>       >>>>>>>>>> No, if we could establish an upper limit for BB(n) for all n,       >>>>>>>>>> then we could solve the hatling problem, as we have an upper       >>>>>>>>>> limit for the number of steps we need to simulate the machine.       >>>>>>>>>>       >>>>>>>>>> BB(n) has a value, but for sufficiently large values of n, we       >>>>>>>>>> don't have an upper bound for BB(n).       >>>>>>>>>>       >>>>>>>>>>>       >>>>>>>>>>> and the problem with allowing for partial decidability is       >>>>>>>>>>> that BB can run continually run more and more deciders in       >>>>>>>>>>> parallel, on every N- state machine, until one comes back       >>>>>>>>>>> with an halting answer, for every N-state machine, which then       >>>>>>>>>>> it can the use to decide what the BB number is for any N ...       >>>>>>>>>>       >>>>>>>>>> So, what BB are you running? Or are you misusing "running" to       >>>>>>>>>> try to mean somehow trying to calculate?       >>>>>>>>>>>       >>>>>>>>>>> contradicting the concept it must have a limit L, where some       >>>>>>>>>>> L- state machine cannot be decidable by *any* partial decider       >>>>>>>>>>> on the matter,       >>>>>>>>>>       >>>>>>>>>> No, it can have a limit, just not a KNOWN limit.       >>>>>>>>>       >>>>>>>>> consensus is there can a known limit L to the BB function, and       >>>>>>>>> proofs have been put out in regards to this       >>>>>>>>>       >>>>>>>>>>       >>>>>>>>>>>       >>>>>>>>>>> so no richard, partial decidability does not work if BB is to       >>>>>>>>>>> have a limit       >>>>>>>>>>>       >>>>>>>>>>       >>>>>>>>>> You only have the problem is BB has a KNOWN limit. Again, you       >>>>>>>>>> trip up on assuming you can know any answer you want.       >>>>>>>>>>       >>>>>>>>>> That some things are not knowable breaks your logic.       >>>>>>>>>       >>>>>>>>       >>>>>>>> I just glanced at your paper and skipped to the conclusion.       >>>>>>>> Why do we care about the undecidability of the halting problem?       >>>>>>>> Because undecidability in general (if it is correct) shows       >>>>>>>> that truth itself is broken. Truth itself cannot be broken.       >>>>>>>> This is the only reason why I have worked on these things       >>>>>>>> for 28 years.       >>>>>>>       >>>>>>> because it makes us suck as developing and maintaining software,       >>>>>>> and as a 35 year old burnt out SWE, i'm tired of living in a       >>>>>>> world running off sucky software. it really is limiting our       >>>>>>> potential, and i want my soon to be born son to have a far better       >>>>>>> experience with this shit than i did.       >>>>>>>       >>>>>>> a consequence of accepting the halting problem is then       >>>>>>> necessarily accepting proof against *all* semantic deciders,       >>>>>>> barring us from agreeing on what such general deciders might be       >>>>>>>       >>>>>>       >>>>>> Exactly: Tarski even "proved" that we can't even directly       >>>>>> compute what is true. This lets dangerous liars get away       >>>>>> with their dangerous lies.       >>>>>>       >>>>>>> this has lead to not only an unnecessary explosion in complexity       >>>>>>> of software engineering, because we can't generally compute       >>>>>>> semantic (turing) equivalence,       >>>>>>>       >>>>>>> but the general trend in deploying software that doesn't have       >>>>>>> computed semantic proofs guaranteeing they actually do what we       >>>>>>> want them to do.       >>>>>>       >>>>>> Yes without computing halting total proof of       >>>>>> correctness is impossible.       >>>>>>       >>>>>>> "testing" is poor substitute for doing so, but that's the most we       >>>>>>> can agree upon due to the current theory of computing.       >>>>>>>       >>>>>>> i think my ideas might contribute to dealing with incompleteness       >>>>>>> in fundamental math more generally ... like producing more       >>>>>>> refined limits to it's philosophical impact. tho idk if it can be       >>>>>>> gotten rid of completely, anymore than we can get rid of the       >>>>>>> words "this statement is false"       >>>>>>>       >>>>>>       >>>>>> I don't think that there actually are any limits       >>>>>> except for expressions requiring infinite proofs.       >>>>>>       >>>>>>> but i am currently focused on the theory of computing and not       >>>>>>> anything more generally. the fundamental objects comprising the       >>>>>>> theory of computing (machines) are far more constrained in their       >>>>>>> definitions than what set theory needs to encompass, and within       >>>>>>> those constraints i think i can twist the consensus into some       >>>>>>> contradiction that are just entirely ignorant of atm       >>>>>>>       >>>>>>       >>>>>> I have explored all of the key areas. None of them       >>>>>> can be made as 100% perfectly concrete and unequivocal       >>>>>> as computing.       >>>>>>       >>>>>>> that's the slam dunk left that i need. i have a means to rectify       >>>>>>> whatever contradiction we find thru the use of RTMs, but i'm       >>>>>>> still teasing out the contradiction that will *force* others to       >>>>>>> notice       >>>>>>>       >>>>>>       >>>>>> I do have my refutation of the halting problem itself       >>>>>> boiled down to a rough draft of two first principles.       >>>>>>       >>>>>> When the halting problem requires a halt decider       >>>>>> to report on the behavior of a Turing machine this       >>>>>> is always a category error because Turing machines       >>>>>> only take finite string inputs.       >>>>>>       >>>>>> The corrected halting problem requires a Turing       >>>>>> machine decider to report in the behavior that its       >>>>>> actual finite string input actually specifies.       >>>>>>       >>>>>>       >>>>>       >>>>> polcott, i'm working on making the halting problem complete and       >>>>> consistent in regards to a subset of the improved "reflective       >>>>> turing machines" that encompasses all useful computations       >>>>>       >>>>> i'm sorry, but not about trying to reaffirm the halting function as       >>>>> still uncomputable by calling it a category error       >>>>>       >>>>       >>>> I do compute the halting function correctly.       >>>       >>> the halting *function* is an abstract mathematical object that maps a       >>> machine description to whether the machine described halts or not,       >>> not the associated machine description that attempts to compute this       >>>       >>       >> All Turing machines only compute the mapping       >> from input finite strings to some value.       >> On this basis I do compute halting correctly.       >>       >>>> I have been doing this for more than three years.       >>>> We probably should not be spamming alt.buddha.short.fat.guy       >>>              [continued in next message]              --- 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