Re: What is the Result from Invoking this Halt Function?
From: Karl Heinz Buchegger (kbuchegg_at_gascad.at)
Date: 08/13/04
- Next message: r.e.s.: "Re: Yet another Attempt at Disproving the Halting Problem"
- Previous message: Karl Heinz Buchegger: "Re: What is the Result from Invoking this Halt Function?"
- In reply to: Peter Olcott: "Re: What is the Result from Invoking this Halt Function?"
- Next in thread: Peter Olcott: "Re: What is the Result from Invoking this Halt Function?"
- Reply: Peter Olcott: "Re: What is the Result from Invoking this Halt Function?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Fri, 13 Aug 2004 11:02:34 +0200
Peter Olcott wrote:
>
> "Karl Heinz Buchegger" <kbuchegg@gascad.at> wrote in message news:411B5DCC.65841593@gascad.at...
> > Peter Olcott wrote:
> > >
> > > "Karl Heinz Buchegger" <kbuchegg@gascad.at> wrote in message news:411B2795.C89F597C@gascad.at...
>
> > And your method is not a correct refutation because the orignal
> > proof still works if one returns a result. And since it has been shown
> > that 'not returning a result' is simply impossible iff a result
> > is generated at all(and it doesn't matter how it is generated), your method
> > vanishes into thin air.
> > It's as simple as that.
>
> In order to show that my method does not work it must be shown
> that for each and every possible circumstance there does not exist
> even one single circumstance where my method would work.
No, we don't have to do that.
In order to show that your method is flawed we just need to find a way
around it. Then we are back at the 'original' proof and all things
remain the same.
>
> Since selectively refraining from returning a result is one single circumstance
> whereby my method works, the only possible refutation would entail
> finding a way such that there can not possibly exist any possible means
> for my program to every selectively refrain from returing a result.
Nonsense. Your method still produces a result. And we can find ways
(always) to use that result to construct the paradox. To use the
return value is just one way to do it, the simplest in the case
of C++. But it need not be so.
Just by refraining to do something, paradoxes don't vanish into thin air.
>
> Not Returning A result Is Not Impossible
It is. It may not look like a 'return' statement (and you know this
otherwise you would not have included the reference to 'read only
screen memory and such things), but the end effect is the same.
There are always ways to feed back the result to LoopIfHalts. And
that is all that we need to show that your 'proof' is based on
something that dosn't exist in reality.
Note: We are in a better situation then you. Whatever you come
up with to make this 'return' impossible, all we need is just
one way that circumvents this. You on the other hand have to
show that there is no way to construct that loop. No way - as in
never, under no circumstances. Just claiming: 'And what if there
is no web cam around' is not enough. You have to show that we
cannot use the output of your machine (program) to construct
a paradox, under no circumstances, no matter what we do.
> The Halt Analyzer could ask the UTM to determine its invocation
> context before it even begins processing.
And it has already been shown that his is beyond the cpabilities
of a TM. So this is not an option. I already have invited you
to include this code in a C++ WillHalt function in a framework
I have provided (since I am not that fluent with TM, but I am
fluent in C++), but you refrained from doing so (pun intended).
-- Karl Heinz Buchegger kbuchegg@gascad.at
- Next message: r.e.s.: "Re: Yet another Attempt at Disproving the Halting Problem"
- Previous message: Karl Heinz Buchegger: "Re: What is the Result from Invoking this Halt Function?"
- In reply to: Peter Olcott: "Re: What is the Result from Invoking this Halt Function?"
- Next in thread: Peter Olcott: "Re: What is the Result from Invoking this Halt Function?"
- Reply: Peter Olcott: "Re: What is the Result from Invoking this Halt Function?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|