Thread 16820618

why do we care whether a program will terminate or run forever?

>>16820618 (OP)
Nothing will run forever

>>16820648
Usain Bolt will

745-state busy beaver machine halts iff ZFC is inconsistent.
744-state busy beaver machine halts iff the Riemann hypothesis is false.
Another example is knowing BB(27) allows one to resolve Goldbach s conjecture.

No fucking clue how it’s all related but the mathematicians say so. For the Turing machines related to the consistency of ZFC, I do know that number hasn’t been static; researching where that BB limit specifically is apparently interesting. Originally it was in the thousands, but the latest revision moved it only slightly from BB(748) to BB(745). Important context.. BB(6) hasn’t even been calculated, only estimated.

>>16820618 (OP)
Because if some innocuous-looking problem reduces to the Halting Problem you know it's not generally solvable with computers. People like to think computers can solve all their problems. They are wrong.

>>16820682
Exactly, the retard assumption is that if it reduces to the Halting problem and figuring out the run of BB(whateverthefuck) then it’s trivial for mathematicians to resolve the original problem, but it’s not. If BB(6) is only estimated, BB(27) alone would have an inconceivable number of digits, and don’t even fucking think about BB(745) and BB(744).

>>16820618 (OP)
it means we dont have to waste our time waiting for it to finish

>>16820618 (OP)
Because program run on computer, and every 7/8 people own some kind of computer. Du yu understend??? Du yu get it now, dum dum?:???

>>16820961
I can unplug a computer and the program stops

>>16820682
>People like to think computers can solve all their problems. They are wrong.
Assuming it's not a time/space problem, this has really nasty implications for the correspondence theory of truth

>>16821012
>this has really nasty implications for the correspondence theory of truth
I don't think so, but you're invited to make this case.

>>16821013
Turing-completeness is tightly wound with first and higher-order logic. If there are real, non-trivial facts that are not expressible in a turing-complete grammar, there is disparity between reality and truth. This is already seemingly the implication given the stochastic nature of certain quantum phenomena, but the more indescribable things we can observe, the greater the empirical certainty. Quite frankly, there was never a compelling case made for it in the first place.

>>16821037
>Turing-completeness is tightly wound with first and higher-order logic
Proof?

>>16820618 (OP)
To avoid wasting time and resources trying to do the impossible, retard.

>>16821085
Whats impossible, turning off a computer?

>>16820682
>People like to think computers can solve all their problems. They are wrong.
Name one problem that computers can’t solve

>>16821128
Which part of the Turing machine computational model states that you can just turn off the Turing machine?

>>16821134
>Name one problem that computers can’t solve
The Halting Problem. I just wanna see what kind of schizo cope you had in mind here.

>>16821160
The halting problem is a problem only insofar as it is not an actual problem. Can you describe even a single real world scenario where the crux of the problem was that one wasnt able to detemine if a program would terminate using another program because it is in by conjecture impossible to do so?

>>16821153
Eh, because you just can?

>>16821134
nta, but according to Rice's theorem, all non-trivial semantic properties of programs are undecidable. This means that, in general, most programs cannot be proven to be correct, not malware, or run without error. Obviously, not being able to prove whether or not a program is or has malware is important to those interested in cyber security.
>https://en.wikipedia.org/wiki/Rice's_theorem

Ken Thompson, in his Turing Award lecture "Reflections on Trusting Trust," famously demonstrated how a malicious but trusted developer of a program, such as a C compiler, can introduce trojans into the computers of unwitting users who run the program.
>https://www.cs.cmu.edu/~rdriley/487/papers/Thompson_1984_ReflectionsonTrustingTrust.pdf

>>16821222
>Can you describe even a single real world scenario where the crux of the problem was that one wasnt able to detemine if a program would terminate using another program
Yeah, for example if I want to write a program that tests whether or not another program terminates.

>>16821134
Losing your virginity

Has nobody ever heard of an infinite loop?

>>16822700
a what?

>>16821250
Would solving the halving problem solve such security issues? I think not, prove me wrong

>>16821267
A hypothetical is not a real world case.
Now accept your retardedness and fuck off to reddit

>>16823148
>A hypothetical is not a real world case.
It's only a "hypothetical" because it's impossible. If it was possible, it would be implemented and put to use. In other words, you're only so confused because you did eat breakfast.

>>16821134
loneliness

>>16821134
alopecia

>>16823194
Shieeeet

>>16821134
>Name one problem that computers can’t solve
finding a proof for a theorem.

>>16824814
... or deciding whether two formulae built with the four basic operations and sin() are equivalent.

>>16822700
Its a problem in general, not about specific cases. Can you prove if any program will terminate, not just some of them?
Still no answer why this is an important thing. BTW the answer is you can turn the computer off and on

>>16820618 (OP)
The same reason you enjoy saving photos of questionably young looking anime girls in bikinis.
It gives our ongoing existence some sort of worth.

>>16820674
He retired in 2017

>>16820618 (OP)
because running programs use resources (memory, cpu cycles etc) dingus. I mean if you had it running on a computer that you dont use for anything else, you wouldnnt care. If you use the computer to do other things than just run that program you want to reclaim those resources to do other things

>>16823148
yes it seems like pure autism -
not just having a look to see if you have unwanted infinite loops, but "is there a general algorithm to autistically solve this problem?"

>>16825314
What stupidity, programs freeze all the time, you can force shut them off or just restart the computer.
I'm convinced these midwit challenges like the Riemann hypothesis or this shit are just set as traps to waste your life on nothing

>>16821037
>Turing-completeness is tightly wound with first and higher-order logic
that's not true at all, the set of first-order validities isn't computable and the second-order validities aren't even computably enumerable

>>16820618 (OP)
finite computation power and entropy.

>>16820618 (OP)
You don't want to stop your program only to find out ot was one step away from solving pi or multiplying the next prime number.
Think of all the time you would have wasted.
We've trained beavers to stay busy so this doesn't happen.
I love science.