Re: Can returning a value change the value itself (in the Halting Problem)

From: Peter Olcott (olcott_at_worldnet.att.net)
Date: 08/18/04 

Date: Wed, 18 Aug 2004 08:59:09 GMT

"Will Twentyman" <wtwentyman@read.my.sig> wrote in message news:4120eaa5$1_4@newsfeed.slurp.net...
> Peter Olcott wrote:
>
> > Many of the attempts at refuting my refutation of the Halting Problem
> > implicitly assumed that the answer to this question is no. They try to
> > prove that my refutation is wrong by making this false assumption.
> > Instead of keeping this false assumption hidden under the cover of
> > rhetoric, let's analyze it explicitly.
> >
> > The goal of this thread is to analyze each and every step exactly
> > one step at a time until this question has final resolution that most
> > everyone can agree to.
>
> So what's the first step? Note: the subject of the thread does not make
> sense as stated.
>
> --
> Will Twentyman
> email: wtwentyman at copper dot net
>

Is there any possible scenario whereby changing the return protocol
and nothing else can actually change the value of the computation itself?
You have the same program (or TM) and the same data input. Is it
possible that refraining from returning a value can derive a different value
than returning this value to all callers?

I will cut directly to the chase here because this has already taken longer
than I wanted. I posted this on Sunday, and we are just now getting around
to beginning to talk about it. Since my burden of proof is simpler
(simply provide a single example of such a case), it is much easier
to evaluate this single example to see if it is valid, than attempt to examine
to categorically exhaustive set to show that each and every case is invalid.
If my case is valid then providing that it is impossible immediately
becomes impossible.

All Caps Are For Emphasis, (in this case) not Yelling.
In the case of LoopIfHalts, when you return a value to it changes
its behavior. If you tell it that it halts, it goes into an infinite loop.
If you tell it that it does not halt, then it halts. Since the result
that you are determining is whether or not it halts, and providing
this result to LoopIfHalts causes it to do the opposite of what
you tell it. RETURNING THE RESULT PRODUCES AN
ENTIRELY DIFFERENT RESULT THAN REFRAINING
FROM RETURNING THE RESULT IN THIS CASE!

Relevant Pages

  • Re: Can returning a value change the value itself (in the Halting Problem)
    ... >> Many of the attempts at refuting my refutation of the Halting Problem ... In the case of LoopIfHalts, when you return a value to it changes ... If you tell it that it halts, it goes into an infinite loop. ... ENTIRELY DIFFERENT RESULT THAN REFRAINING ...
    (sci.logic)
  • Re: Can returning a value change the value itself (in the Halting Problem)
    ... > Is there any possible scenario whereby changing the return protocol ... > possible that refraining from returning a value can derive a different value ... > In the case of LoopIfHalts, when you return a value to it changes ... If you tell it that it halts, it goes into an infinite loop. ...
    (comp.theory)
  • Re: Can returning a value change the value itself (in the Halting Problem)
    ... > Is there any possible scenario whereby changing the return protocol ... > possible that refraining from returning a value can derive a different value ... > In the case of LoopIfHalts, when you return a value to it changes ... If you tell it that it halts, it goes into an infinite loop. ...
    (sci.logic)
  • Re: What is the Result from Invoking this Halt Function?
    ... > In the case of LoopIfHalts, when you return a value to it changes ... If you tell it that it halts, it goes into an infinite loop. ... > ENTIRELY DIFFERENT RESULT THAN REFRAINING ... a halt analyzer *cannot* refrain from returning a result. ...
    (comp.theory)
  • Re: What is the Result from Invoking this Halt Function?
    ... > In the case of LoopIfHalts, when you return a value to it changes ... If you tell it that it halts, it goes into an infinite loop. ... > ENTIRELY DIFFERENT RESULT THAN REFRAINING ... a halt analyzer *cannot* refrain from returning a result. ...
    (sci.logic)