Re: What is the Result from Invoking this Halt Function?
David C. Ullrich
<olc...@worldnet.att.net> wrote words which said
nothing at all about the following question, which
we repeat for his convenience:
meanwhile, why don't you answer Peter's question.
correcting a typo and clarifying the block structure,
does the following code halt or not?
possible.perfect = 3
loop: sum=0
for all numbers k from 1 .. sqrt(possible.perfect)
if k divides possible.perfect, sum = sum+k
next k
if sum=possible.perfect then print possible.perfect "is an odd perfect
number"; halt. endif
possible.perfect=possible.perfect+2
goto loop
************************
David C. Ullrich
sorry about the inelegant formatting - typing
one-handed for a few weeks...
David C. Ullrich
<olc...@worldnet.att.net> wrote:
>[...]
>The one thing that you (AND EVERYONE ELSE) is doing here
>is completely ignoring the HALTING PROBLEM aspect of this
>question.
one of many things that -you- are ignoring is the
following question:
Peter Olcott
<ste...@nomail.com> wrote in message news:cef9fo$2vlp$1...@msunews.cl.msu.edu...
> In comp.theory Peter Olcott <olc...@worldnet.att.net> wrote:
>
> : <ste...@nomail.com> wrote in message news:cedkdc$1ifo$1...@msunews.cl.msu.edu...
> :> In comp.theory Peter Olcott <olc...@worldnet.att.net> wrote:
> :>
> :> : //
> :> : // All of the functions are assumed to be standard C++
> :> : //
> :> : void LoopIfHalts (string ProgramSourceFile, string InputData)
> :> : {
> :> : if WillHalt01(ProgramSourceFile, InputData)
> :> : while (true)
> :> : ;
> :> : else // to make it easier for people who don't know C++
> :> : return;
> :> : }
> :>
> :> : bool WillHalt02(LoopIfHalts, LoopIfHalts)
> :>
> :> : If we assume that WillHalt01, and WillHalt02 are in separate memory
> :> : space, such that LoopIfHalts only calls WillHalt01, what will be the
> :> : result of invoking WillHalt02 ?
> :>
> :> : ***NOTE*** WillHalt01, and Willhalt02 are identical in every
> :> : way, except that they reside in separate memory space.
> :>
> :>
> :> If you are assuming that WillHalt01 and WillHalt02 are also written
> :> in C++, and if we ignore all machine dependent memory tricks, then
> :> in this case WillHalt02 will return the wrong answer to the halting question.
> :>
> :> If WillHalt02 returns true, then WillHalt01 will also return true because
> :> the code is identical and the input is identical. If WillHalt01 returns true
> :> then LoopIfHalts loops, so WillHalt02 is also wrong.
> :>
> :> If WillHalt02 returns false, then WillHalt01 will also return false because
> :> the code is identical and the input is identical. If WillHalt01 returns false
> :> then LoopIfHalts halts, so WillHalt02 is wrong.
> :>
> :> Note that this argument does not depend on any details about a specific
> :> implementation of WillHalt02, so this argument works for any attempting
> :> solution to the halting problem.
> :>
> :> Stephen
>
> : The one thing that you (AND EVERYONE ELSE) is doing here
> : is completely ignoring the HALTING PROBLEM aspect of this
> : fed back to LoopIfHalts, and LoopIfHalts changes is behavior based
> : on these results. LoopIfHalts does not change its behavior based
> : on the result of WillHalt02. It does not even know about WillHalt02.
> : Willhalt02 is in a separate memory space.
>
> What does that have to do with anything? WillHalt01 and WillHalt02
> are identical code. Given identical input they return identical results.
SORRY BUT THAT IS JUST MINDLESS NONSENSE!!!
That sort of answer could be provided by a mindless automaton,
but not by anyone who read and understood the words, that I
just said immediately above.
I GET IT YOU REFUTE WHAT I AM SAYING WITHOUT
EVER READING ANY OF THE WORDS, AM I RIGHT???
>
> : Thus because LoopIfHalts changes its behavior based on the results
> : provided by WillHalt01, and this changed behavior changes the result
> : which again changes the behavior, this sequence is not at all the same
> : as the sequence where WillHalt02 makes its analysis, and this result
>
> LoopIfHalts does not change its behavior. It is a program, and
> therefore has a behavior determined by the program and the input.
> This behavior never changes. WillHalt01 and WillHalt02 are identical code,
> and given identical input will return identical results.
>
> : In the final analysis WillHalt02 will be able to determine that
> : LoopIfHalts does indeed halt. It will know that its other copy
> : (which is input to the analysis) is smart enough to see that it
> : can not correctly return true, thus must execute the else clause.
>
> But the code is identical. Identical code with identical input
> has to produce identical results. You are claiming that somewhere
> in WillHalt01 and WillHalt02 there is an if statement such as
> if (somecondition)
> and despite the fact all the variables are the same and have the
> same value, the if evalutes to true in one program but false in another.
> How can two identical if statements evaluate differently when all the
> variables have the same value?
>
> Again, if you want to convince me, just write two identical pieces of code
> that produce different results when given the same input.
>
> Stephen
ste...@nomail.com
: <ste...@nomail.com> wrote in message news:cef9fo$2vlp$1...@msunews.cl.msu.edu...
So in your opinion, the idea that two programs with identical code
and identical inputs producing identical results is mindless nonsense?
I guess that sums up your entire position. Programs just magically
do things, independent of the code and independent of the input.
If you allow that, then the Halting problem is solvable.
Stephen
Mitch Harris
>
>I guess that sums up your entire position. Programs just magically
>do things, independent of the code and independent of the input.
>If you allow that, then the Halting problem is solvable.
Really? giving your TMs/programs a "magic" operation would make it
that much -harder- to determine if they halt or not, right?
Or maybe you're saying that the a proposed TM to determine halting
could itself use magic? Then, sure it's be trivially easy (magic can
do anything, right?)
Mitch
PS I'm not talking about oracles, I'm talking about -magic-, the real
stuff.
ste...@nomail.com
: In article <cegkn5$13cd$1...@msunews.cl.msu.edu>, <ste...@nomail.com> wrote:
:>
:>I guess that sums up your entire position. Programs just magically
:>do things, independent of the code and independent of the input.
:>If you allow that, then the Halting problem is solvable.
: Really? giving your TMs/programs a "magic" operation would make it
: that much -harder- to determine if they halt or not, right?
: Or maybe you're saying that the a proposed TM to determine halting
: could itself use magic? Then, sure it's be trivially easy (magic can
: do anything, right?)
: Mitch
It's magic. I do not pretend to understand it. :)
Stephen
>parr(*>
news:AFqOc.353067$Gx4....@bgtnsc04-news.ops.worldnet.att.net...
|
| I thought that you said that you were going to try to help me?
I did say that. I suggested it be done via email ...
I was about to say you didn't contact me. I apologise, the mail
address was faulty, I missed the 'y' out of:
LauryKing at BTInternet dot com
I was hasty in thinking you rejected my offer. Sorry.
I'm still quite happy to try to help on Turing's theorem and the
logic thereof.
--
)>==ss$$%PARR(º> Parr
Peter Olcott
">parr(*>" <gniK...@tenretnitb.moc> wrote in message news:ceh48g$4gb$1...@titan.btinternet.com...
> "Peter Olcott" <olc...@worldnet.att.net> wrote in message
> news:AFqOc.353067$Gx4....@bgtnsc04-news.ops.worldnet.att.net...
> |
> | I thought that you said that you were going to try to help me?
>
> I did say that. I suggested it be done via email ...
I will accept your offer in this public forum only.
Everything must be a matter of permanent public record.
>parr(*>
|
| ">parr(*>" <gniK...@tenretnitb.moc> wrote in message
news:ceh48g$4gb$1...@titan.btinternet.com...
| > "Peter Olcott" <olc...@worldnet.att.net> wrote in message
| > news:AFqOc.353067$Gx4....@bgtnsc04-news.ops.worldnet.att.net...
| > |
| > | I thought that you said that you were going to try to help me?
| >
| > I did say that. I suggested it be done via email ...
|
| I will accept your offer in this public forum only.
| Everything must be a matter of permanent public record.
Don't blame you for not trusting me.
I don't feel capable of reviewing everything that has been said in
sci.comp on this subject and cleaning it up. Can we make a clean
start somewhere?
BTW, I don't know c++, so can't help on this current problem.
--
)>==ss$$%PARR(º> Parr
David C. Ullrich
<olc...@worldnet.att.net> wrote:
>
>">parr(*>" <gniK...@tenretnitb.moc> wrote in message news:ceh48g$4gb$1...@titan.btinternet.com...
>> "Peter Olcott" <olc...@worldnet.att.net> wrote in message
>> news:AFqOc.353067$Gx4....@bgtnsc04-news.ops.worldnet.att.net...
>> |
>> | I thought that you said that you were going to try to help me?
>>
>> I did say that. I suggested it be done via email ...
>
>I will accept your offer in this public forum only.
>Everything must be a matter of permanent public record.
if you want that record to be something other than
an embarassment you should tell us for the record
whether
possible.perfect = 3
loop: sum=0
for all numbers k from 1 .. sqrt(possible.perfect)
if k divides possible.perfect, sum = sum+k
next k
if sum=possible.perfect then print possible.perfect "is an odd perfect
number"; halt. endif
possible.perfect=possible.perfect+2
goto loop
halts. [no, 'this is off-topic' is not an answer to the question -
it's also clearly not so.]
Michael N. Christoff
> "Michael N. Christoff" <mchri...@sympatico.caREMOVETHIS> wrote in
message news:dwxOc.10526$pc.5...@news20.bellglobal.com...> > news:msjOc.153694$OB3.1...@bgtnsc05-news.ops.worldnet.att.net...
> > >
> > > //
> > > // All of the functions are assumed to be standard C++
> > > //
> > > void LoopIfHalts (string ProgramSourceFile, string InputData)
> > > {
> > > if WillHalt01(ProgramSourceFile, InputData)
> > > while (true)
> > > ;
> > > else // to make it easier for people who don't know C++
> > > return;
> > > }
> > >
> > > bool WillHalt02(LoopIfHalts, LoopIfHalts)
> > >
> > > If we assume that WillHalt01, and WillHalt02 are in separate memory
> > > space, such that LoopIfHalts only calls WillHalt01, what will be the
> > > result of invoking WillHalt02 ?
> > >
> > > ***NOTE*** WillHalt01, and Willhalt02 are identical in every
> > > way, except that they reside in separate memory space.
> > >
> >
to
> > LoopIfHalts? But LoopIfHalts takes two parameters so how are you
breaking
> > up LoopIfHalts into ProgramSourceFile and InputData?
>
> WillHalt02() tests to see if LoopIfHalts will halt when testing to see
> if itself will halt. This is only half as convoluted as the original.
>
> WillHalt01 will know that LoopIfHalts will halt, yet not be able
> to report this fact, otherwise it would not halt. In other words
> WillHalt01 will know to refrain from returning true, because it will
> know that if it returns true, that this will be used by LoopIfHalts
> to change its own behavior, thus making it false. In htis case
> the only other option is the else clause, which results in
> LoopIfHalts, halting.
>
This doesn't answer my question.
> > So WillHalt02 checks if LoopIfHalts halts when given as input the source
to
> > LoopIfHalts? But LoopIfHalts takes two parameters so how are you
breaking
> > up LoopIfHalts into ProgramSourceFile and InputData?
l8r, Mike N. Christoff
Peter Olcott
"Michael N. Christoff" <mchri...@sympatico.caREMOVETHIS> wrote in message news:1LwPc.20690$Vm1.2...@news20.bellglobal.com...
Simon G Best
> I don't see any question in here.
If you read Michael N Christoff's posts properly, you would see his
questions. They are:-
> "Michael N. Christoff" <mchri...@sympatico.caREMOVETHIS> wrote in message news:1LwPc.20690$Vm1.2...@news20.bellglobal.com...
>
>>"Peter Olcott" <olc...@worldnet.att.net> wrote in message
>>news:28zOc.355608$Gx4.3...@bgtnsc04-news.ops.worldnet.att.net...
>>
>>>"Michael N. Christoff" <mchri...@sympatico.caREMOVETHIS> wrote in
[...]
>>>>So WillHalt02 checks if LoopIfHalts halts when given as input the source
>>>>to LoopIfHalts? But LoopIfHalts takes two parameters so how are you
>>>>breaking up LoopIfHalts into ProgramSourceFile and InputData?
And then you failed to answer the questions that he asked:-
>>>WillHalt02() tests to see if LoopIfHalts will halt when testing to see
>>>if itself will halt. This is only half as convoluted as the original.
>>>
>>>WillHalt01 will know that LoopIfHalts will halt, yet not be able
>>>to report this fact, otherwise it would not halt. In other words
>>>WillHalt01 will know to refrain from returning true, because it will
>>>know that if it returns true, that this will be used by LoopIfHalts
>>>to change its own behavior, thus making it false. In htis case
>>>the only other option is the else clause, which results in
>>>LoopIfHalts, halting.
To which he correctly responded:-
>>This doesn't answer my question.
You really do need to spend more time reading and less time reacting.
Simon
Peter Olcott
"Simon G Best" <s.g....@btopenworld.com> wrote in message news:410F1487...@btopenworld.com...
> Peter Olcott wrote:
> > I don't see any question in here.
>
> If you read Michael N Christoff's posts properly, you would see his
> questions. They are:-
>
> > "Michael N. Christoff" <mchri...@sympatico.caREMOVETHIS> wrote in message news:1LwPc.20690$Vm1.2...@news20.bellglobal.com...
> >
> >>"Peter Olcott" <olc...@worldnet.att.net> wrote in message
> >>news:28zOc.355608$Gx4.3...@bgtnsc04-news.ops.worldnet.att.net...
> >>
> >>>"Michael N. Christoff" <mchri...@sympatico.caREMOVETHIS> wrote in
> [...]
> >>>>So WillHalt02 checks if LoopIfHalts halts when given as input the source
> >>>>to LoopIfHalts? But LoopIfHalts takes two parameters so how are you
> >>>>breaking up LoopIfHalts into ProgramSourceFile and InputData?
This question from five posts ago? (when there are that many levels of quote
marks it doesn't look like a current question)
WillHalt(LoopIfHalts, LoopIfHalts, LoopIfHalts)
or LoopIfHalts(DataInput)
Simon G Best
> "Simon G Best" <s.g....@btopenworld.com> wrote in message news:410F1487...@btopenworld.com...
>
>>Peter Olcott wrote:
>>
>>>I don't see any question in here.
>>
>>If you read Michael N Christoff's posts properly, you would see his
>>questions. They are:-
>>
>>>"Michael N. Christoff" <mchri...@sympatico.caREMOVETHIS> wrote in message news:1LwPc.20690$Vm1.2...@news20.bellglobal.com...
>>>
>>>>"Peter Olcott" <olc...@worldnet.att.net> wrote in message
>>>>news:28zOc.355608$Gx4.3...@bgtnsc04-news.ops.worldnet.att.net...
>>>>
>>>>>"Michael N. Christoff" <mchri...@sympatico.caREMOVETHIS> wrote in
>>>>>
>>>>>>to LoopIfHalts? But LoopIfHalts takes two parameters so how are you
>>>>>>breaking up LoopIfHalts into ProgramSourceFile and InputData?
>
> This question from five posts ago? (when there are that many levels of quote
> marks it doesn't look like a current question)
It's only five posts ago because you failed to answer it when it was one
post ago, and failed to even read it when it was only three posts ago.
And you still haven't answered it!
Ignoring what others have bothered to write in their posts doesn't make
you right and Turing wrong. It just means you're being ignorant.
Simon
Robert Low
Simon G Best <s.g....@btopenworld.com> wrote:
>Ignoring what others have bothered to write in their posts doesn't make
>you right and Turing wrong.
But it's his only strategy: averting his gaze from the contradiction
(in the proof that the halting problem is not solvable) is his reason
for claiming that there isn't one.
Daryl McCullough
>>>>>>>So WillHalt02 checks if LoopIfHalts halts when given as input the source
>>>>>>>to LoopIfHalts? But LoopIfHalts takes two parameters so how are you
>>>>>>>breaking up LoopIfHalts into ProgramSourceFile and InputData?
>>
>> This question from five posts ago? (when there are that many levels
>> of quote marks it doesn't look like a current question)
>
>It's only five posts ago because you failed to answer it when it was one
>post ago, and failed to even read it when it was only three posts ago.
Yes, this error, which Peter has made *repeatedly* without bothering
to fix (even though the fix is trivial) is an important clue, I think,
to Peter's problem. He's a programmer, so I'm sure he would never
write
void LoopIfHalts(String M, String x) {
if (H(M,x)) {
while (true) {}
}
}
LoopIfHalts(LoopIfHalts);
He would *know* that makes no sense, because (1) LoopIfHalts
is defined to have two string arguments, and (2) LoopIfHalts
is not a string, it is a program, so it makes no sense to treat
it as if it were a string. (You need a coding function.)
But Peter starts off by assuming the existence of a WillHalt
function, and so he's already in fantasyland, so he thinks
all bets are off. He throws out everything he knows about
programming, including the fact that a function with two
arguments can't be treated like a function with one argument.
He throws out what he knows about programs, that---unless a
program refers to global variables (or "static variables" in
Java), or gets inputs from the user, or reads from some input
stream, or uses a random number generator---if you run the same
program twice on the same arguments, you will typically get the
same result.
Some people have a real problem with conditional reasoning. Once
they start reasoning under a counterfactual assumption, they toss
out everything they know.
--
Daryl McCullough
Ithaca, NY
Peter Olcott
"Daryl McCullough" <da...@atc-nycorp.com> wrote in message news:ceo8q...@drn.newsguy.com...
> Simon G Best says...
>
> >>>>>>>So WillHalt02 checks if LoopIfHalts halts when given as input the source
> >>>>>>>to LoopIfHalts? But LoopIfHalts takes two parameters so how are you
> >>>>>>>breaking up LoopIfHalts into ProgramSourceFile and InputData?
> >>
> >> This question from five posts ago? (when there are that many levels
> >> of quote marks it doesn't look like a current question)
> >
> >It's only five posts ago because you failed to answer it when it was one
> >post ago, and failed to even read it when it was only three posts ago.
>
> Yes, this error, which Peter has made *repeatedly* without bothering
> to fix (even though the fix is trivial) is an important clue, I think,
> to Peter's problem. He's a programmer, so I'm sure he would never
> write
>
> void LoopIfHalts(String M, String x) {
> if (H(M,x)) {
> while (true) {}
> }
> }
>
> LoopIfHalts(LoopIfHalts);
>
> He would *know* that makes no sense, because (1) LoopIfHalts
> is defined to have two string arguments, and (2) LoopIfHalts
> is not a string, it is a program, so it makes no sense to treat
> it as if it were a string. (You need a coding function.)
So I make several trivial errors that have no effect on my main
point, from my single minded focus on this main point. So what?
Peter Olcott
"Robert Low" <mtx...@linux.services.coventry.ac.uk> wrote in message news:ceo5kk$ns9$1...@sunbeam.coventry.ac.uk...
I have now more clearly than ever shown that what was thought to be
a contradiction is not.
Just because I make a lot of mistakes, and am not very clear in what I
say, does not at all logically entail that my primary point is not correct.
To a person that relies upon the third-rate stand-in of credibility rather
a direct measure of validity or soundness it might seem like I must be
wrong. You would think that a guy that could prove Turning wrong would
not make so many mistakes wouldn't you?
Peter Olcott
"Simon G Best" <s.g....@btopenworld.com> wrote in message news:410F9729...@btopenworld.com...
> Peter Olcott wrote:
> > "Simon G Best" <s.g....@btopenworld.com> wrote in message news:410F1487...@btopenworld.com...
> >
> >>Peter Olcott wrote:
> >>
> >>>I don't see any question in here.
> >>
> >>If you read Michael N Christoff's posts properly, you would see his
> >>questions. They are:-
> >>
> >>>"Michael N. Christoff" <mchri...@sympatico.caREMOVETHIS> wrote in message
news:1LwPc.20690$Vm1.2...@news20.bellglobal.com...
> >>>
> >>>>"Peter Olcott" <olc...@worldnet.att.net> wrote in message
> >>>>news:28zOc.355608$Gx4.3...@bgtnsc04-news.ops.worldnet.att.net...
> >>>>
> >>>>>"Michael N. Christoff" <mchri...@sympatico.caREMOVETHIS> wrote in
> >>>>>
> >>>>>>So WillHalt02 checks if LoopIfHalts halts when given as input the source
> >>>>>>to LoopIfHalts? But LoopIfHalts takes two parameters so how are you
> >>>>>>breaking up LoopIfHalts into ProgramSourceFile and InputData?
> >
> > This question from five posts ago? (when there are that many levels of quote
> > marks it doesn't look like a current question)
>
> It's only five posts ago because you failed to answer it when it was one
> post ago, and failed to even read it when it was only three posts ago.
>
> And you still haven't answered it!
I answered it in my last post. It is only a triviality. give up on these
trivialities. I make lots of mistakes, none of these proves me wrong
in my main point.
Kenneth Doyle
> Just because I make a lot of mistakes, and am not very clear in what I
> say, does not at all logically entail that my primary point is not
> correct.
But it does disqualify you from insisting that people aren't getting your
point because they're not reading you carefully.
--
CodeCutter - good, fast and cheap; pick two.
Robert Low
Peter Olcott <olc...@worldnet.att.net> wrote:
>"Robert Low" <mtx...@linux.services.coventry.ac.uk> wrote in message
>> (in the proof that the halting problem is not solvable) is his reason
>> for claiming that there isn't one.
>say, does not at all logically entail that my primary point is not correct.
No. But it does make your argument less convincing.
>wrong. You would think that a guy that could prove Turning wrong would
>not make so many mistakes wouldn't you?
Well, there's something we can all agree on.
Mitch Harris
>
> Some people have a real problem with conditional reasoning. Once
> they start reasoning under a counterfactual assumption, they toss
> out everything they know.
How do you get someone to accept such reasoning? Is it just
practice?
I think parr has the right question with -1*-1 = -1. It's
totally counterintuitive (for some reasonable concept of
intuition). Even knowing and using the rule, an unwanted
conclusion is still easy to deny.
--
Mitch Harris
(remove q to reply)
Simon G Best
> "Simon G Best" <s.g....@btopenworld.com> wrote in message news:410F9729...@btopenworld.com...
[...]
>>And you still haven't answered it!
>
> I answered it in my last post.
Was it where you wrote, "WillHalt(LoopIfHalts, LoopIfHalts, LoopIfHalts)
or LoopIfHalts(DataInput)"? How does that answer the question that he
actually asked?
Nevermind. You're obviously hopelessly incompetent at C++.
> It is only a triviality. give up on these
> trivialities. I make lots of mistakes, none of these proves me wrong
> in my main point.
You make a lot of mistakes, and some of them are fundamental to your
"main point", leaving your "main point" plainly wrong. You keep making
some such mistakes over and over and over again, no matter how many
times they've been pointed out to you and explained to you.
Repeatedly ignoring such corrections does not stop those mistakes from
being mistakes.
Repeatedly ignoring other people's explanations does not stop those
explanations from being correct.
Repeatedly making the same mistakes over and over and over again does
not stop them from being mistakes.
Simon
Kent Paul Dolan
>> You would think that a guy that could prove Turning wrong would
>> not make so many mistakes wouldn't you?
> Well, there's something we can all agree on.
Well, no.
I, for one, would figure that a guy who would make the mistake
of thinking he could prove Turing wrong (and Goedel, and ...)
would be a guy who made lots of other mistakes via careless
thinking and sloppy habits too, and so it has eventuated.
A God who can't write a working C++ program to save his life
isn't a God for programmers.
xanthian.
He also isn't "a programmer", come to that.
--
Posted via Mailgate.ORG Server - http://www.Mailgate.ORG
Robert Low
>"Robert Low" <mtx...@linux.services.coventry.ac.uk> wrote:
>>> You would think that a guy that could prove Turning wrong would
>>> not make so many mistakes wouldn't you?
>> Well, there's something we can all agree on.
>of thinking he could prove Turing wrong (and Goedel, and ...)
>would be a guy who made lots of other mistakes via careless
>thinking and sloppy habits too, and so it has eventuated.
I'm not disagreeing with that, either: that is perfectly
consistent with somebody who actually could prove Turing wrong
not being the sort who'd make lots of sloppy mistakes.
(And now we're back with counterfactuals. Oh dear.)
Michael N. Christoff
>
> "Robert Low" <mtx...@linux.services.coventry.ac.uk> wrote in message
news:ceo5kk$ns9$1...@sunbeam.coventry.ac.uk...
> >
> > Simon G Best <s.g....@btopenworld.com> wrote:
> > >Ignoring what others have bothered to write in their posts doesn't make
> > >you right and Turing wrong.
> >
> > But it's his only strategy: averting his gaze from the contradiction
> > (in the proof that the halting problem is not solvable) is his reason
> > for claiming that there isn't one.
> >
> >
> > --
> > Rob. http://www.mis.coventry.ac.uk/~mtx014/
>
> I have now more clearly than ever shown that what was thought to be
> a contradiction is not.
>
> Just because I make a lot of mistakes, and am not very clear in what I
> say, does not at all logically entail that my primary point is not
correct.
>
True. But to _convince others_ of your point, you must be logically
accurate. This is the stage your 'proof' is at at this point:
http://www.scienceteecher.com/miracle.htm
> To a person that relies upon the third-rate stand-in of credibility rather
> a direct measure of validity or soundness it might seem like I must be
> wrong.
The majority of us have read and understood the proof for ourselves. We are
not 'taking anyone's word' about whether it is true or false or
inconsistent.
> You would think that a guy that could prove Turning wrong would
> not make so many mistakes wouldn't you?
>
And one would hope he could recognize when he was wrong, learn from his
mistakes, and move on to a line of proof that would actually succeed when
the one he was working on unequivocably proved itself to be a dead end.
l8r, Mike N. Christoff
Acme Diagnostics
>
>>> You would think that a guy that could prove Turning wrong would
>>> not make so many mistakes wouldn't you?
>
>> Well, there's something we can all agree on.
>
>Well, no.
>
>I, for one, would figure that a guy who would make the mistake
>of thinking he could prove Turing wrong (and Goedel, and ...)
>would be a guy who made lots of other mistakes via careless
>thinking and sloppy habits too, and so it has eventuated.
>
>A God who can't write a working C++ program to save his life
>isn't a God for programmers.
How can you know he hasn't created those programmers
with his <Twilight Zone theme> M ( I (( N ((( D (((( ?
In the immortal words of "olcott <olc...@uswest.net>":
"It is the mind that creates the physical reality,
not, the other way around... There is a difference
between the projector, and the projected...
Another way of looking at it, is that they are
both the same... Except that in actuality the
mind can influence physical reality to a greater
extent, than the other way around..."
http://www.google.com/groups?selm=39B55772...@uswest.net
So remember: "Another way of looking at it is that they are both the same." -
Olcott, aka "God," author of "Lucid" which removes all ambiguity
from Language...or, as Marx put it philosophically:
Q: What's the difference between a chicken leg?
A: One leg is both the same.
There's more!
- - - - -
A poster with the nic 4petesake (4pet...@home.com) identifies
himself as "Peter L. Olcott" in this post (and others):
http://www.google.com/groups?selm=CFVd4.1005$9m1....@news1.rdc1.ne.home.com
google count for "4petesake" = 956.
Spot check of 4petesake writings follows:
"You take the position that God don't exist. I KNOW otherwise." <snip>
"...Shortest path is to study Hinduism, and Buddhism (in all their forms)
along with every form of Mysticism... (Christian, Jewish, and Pagan).
and then test each and every one of these against specific objective
measures of their truth..." <snip> "If you begin to pay attention to nearly
infinite detail of every case of cause-and-effect, you will eventually
KNOW that things are nothing at all like what you supposed that they
were." <snip> "Actually I am a genius..."
http://www.google.com/groups?selm=wbGT3.724$AM2....@news.rdc1.ne.home.com
"I KNOW the mathematics of language from a Graduate course
in Computer Science"
http://www.google.com/groups?selm=CEFW3.1298$AM2....@news.rdc1.ne.home.com
"Absolute truth can be verified analytically..."
http://www.google.com/groups?selm=t0Ix3.458$mH1...@news.rdc1.ne.home.com
"The premise of biblical inerrancy logically entails treating the bible
as a mathematical formalism. Deriving semantics in this way results in
a message quite different from what anyone has yet disseminated."
http://www.google.com/groups?selm=qXAx3.441$mH1...@news.rdc1.ne.home.com
In 1999 in a religous thread, another poster with much engagement
with 4petesake writes this about him:
"here we see that your proof of God
fails by your very own criterion.
I have a sneaking suspicion that your criterion
for knowledge applies to every but yourself.
when it suits you, empirical knowledge is your basis for truth.
when it doesn't suit you, all empirical knowledge is flawed.
I don't even think *you* know what you think.
you contradict yourself so often it's scary.
good luck in sorting out your confused state."
http://www.google.com/groups?selm=385F07BF...@yahoo.com
1999! And did I not identify this unfalsifiability device ("reliability of language"
kook dogma) a while ago here (scroll towards bottom)...
http://www.google.com/groups?selm=40f061a2$0$25801$45be...@newscene.com
Continuing 4petesake spot check...
"...and yet my message is crucial to ALL Christians."
http://www.google.com/groups?selm=EHEx3.449$mH1...@news.rdc1.ne.home.com
"The Holy Spirit though me is the arbiter...
Some of what I say is directly from him, and some is not...
Only the Holy Spirit in you knows for sure which is which..."
http://www.google.com/groups?selm=Lp7Z3.144$HF2....@news.rdc1.ne.home.com
(Interesting exchanges in this message)
"Except the word for the word-of-God is the same root word
as the word LOGIC, itself... LOGOS..."
http://www.google.com/groups?selm=yqBZ3.233$HF2....@news.rdc1.ne.home.com
Wasn't it me who posted something about seeing the secret
truth in an ancient metaphor? (Standard kook fare.)
http://www.google.com/groups?selm=410494ad$0$41741$45be...@newscene.com
Larry
``
Marc Goodman
I volunteer you for the FAQ page ;)
Just as an aside, /of course/ he's a kook.
It wouldn't be any fun poking him if he wasn't,
he'd just say, "oh, I didn't see that, sorry I
was wrong" and move on. Where's the FUN in THAT?
Peter Olcott
"Michael N. Christoff" <mchri...@sympatico.caREMOVETHIS> wrote in message news:ET6Qc.33276$Vm1.6...@news20.bellglobal.com...
>
> "Peter Olcott" <olc...@worldnet.att.net> wrote in message
> news:auXPc.381601$Gx4.2...@bgtnsc04-news.ops.worldnet.att.net...
> The majority of us have read and understood the proof for ourselves. We are
> not 'taking anyone's word' about whether it is true or false or
> inconsistent.
No one (not even one person) has gone through what I have said
point by point, and found any holes in my conclusion.
All that Turing proved was that it is impossible to construct a Halt
Analyzer that always returns the correct result of the Halt Analysis
to the program (or TM) being analyzed.
Dozens of people will claim that this statement is incorrect. Not one
person will ever be able to show exactly why and how this statement
is incorrect.
About all that I have gotten in the way of refutation of this statement is
"you are wrong because you don't know logic". No one ever will
point out any specific error in the above statement specifically because
there is no error to be found. The statement is correct.
Peter Olcott
"Robert Low" <mtx...@linux.services.coventry.ac.uk> wrote in message news:ceqjg2$t2q$1...@sunbeam.coventry.ac.uk...
> Kent Paul Dolan <xant...@well.com> wrote:
> >"Robert Low" <mtx...@linux.services.coventry.ac.uk> wrote:
> >>> You would think that a guy that could prove Turning wrong would
> >>> not make so many mistakes wouldn't you?
> >> Well, there's something we can all agree on.
> >I, for one, would figure that a guy who would make the mistake
> >of thinking he could prove Turing wrong (and Goedel, and ...)
> >would be a guy who made lots of other mistakes via careless
> >thinking and sloppy habits too, and so it has eventuated.
>
>
> I'm not disagreeing with that, either: that is perfectly
> consistent with somebody who actually could prove Turing wrong
> not being the sort who'd make lots of sloppy mistakes.
Although I have heard that Einstein couldn't find his way home,
and would go into stores wandering around aimlessly, only to
later leave without what he came for.
Peter Olcott
"Robert Low" <mtx...@linux.services.coventry.ac.uk> wrote in message news:ceq108$br3$2...@sunbeam.coventry.ac.uk...
>
> Peter Olcott <olc...@worldnet.att.net> wrote:
> >"Robert Low" <mtx...@linux.services.coventry.ac.uk> wrote in message
> >> But it's his only strategy: averting his gaze from the contradiction
> >> (in the proof that the halting problem is not solvable) is his reason
> >> for claiming that there isn't one.
> >Just because I make a lot of mistakes, and am not very clear in what I
> >say, does not at all logically entail that my primary point is not correct.
>
> No. But it does make your argument less convincing.
That is a reasonable statement. If I was careful enough to not
ever make these sorts of errors, then it would take more time
than I can spare.
Simon G Best
>
> No one (not even one person) has gone through what I have said
> point by point, and found any holes in my conclusion.
That's plainly untrue. See the newsgroup comp.theory for plenty of
examples.
> All that Turing proved was that it is impossible to construct a Halt
> Analyzer that always returns the correct result of the Halt Analysis
> to the program (or TM) being analyzed.
Which means that it's impossible for a halt analyzer to be a Turing Machine.
> Dozens of people will claim that this statement is incorrect. Not one
> person will ever be able to show exactly why and how this statement
> is incorrect.
http://www.turingarchive.org/browse.php/B/12
> About all that I have gotten in the way of refutation of this statement is
> "you are wrong because you don't know logic".
Plainly untrue. See the newsgroup sci.logic for plenty of examples.
> No one ever will
> point out any specific error in the above statement specifically because
> there is no error to be found. The statement is correct.
Are you blue in the face, yet?
Quite why you think repeated assertions - especially ones which are now
plainly false - are going to convince anyone is quite something of a
mystery.
Simon
Peter Olcott
"Simon G Best" <s.g....@btopenworld.com> wrote in message news:411185F4...@btopenworld.com...
> Peter Olcott wrote:
> >
> > No one (not even one person) has gone through what I have said
> > point by point, and found any holes in my conclusion.
>
> That's plainly untrue. See the newsgroup comp.theory for plenty of
> examples.
>
> > All that Turing proved was that it is impossible to construct a Halt
> > Analyzer that always returns the correct result of the Halt Analysis
> > to the program (or TM) being analyzed.
>
> Which means that it's impossible for a halt analyzer to be a Turing Machine.
Any computation that can be carried out by mechanical means can be
performed by some Turing Machine.
You must have a very limited understanding of this.
> > Dozens of people will claim that this statement is incorrect. Not one
> > person will ever be able to show exactly why and how this statement
> > is incorrect.
>
> http://www.turingarchive.org/browse.php/B/12
>
> > About all that I have gotten in the way of refutation of this statement is
> > "you are wrong because you don't know logic".
>
> Plainly untrue. See the newsgroup sci.logic for plenty of examples.
Try and quote any. I have not found one single example that was
not incorrect for one reason or another, or only pointed out an
error of no consequence to my primary point. Most were merely
vague generalizations such as "you are wrong because you don't know logic"
>
> > No one ever will
> > point out any specific error in the above statement specifically because
> > there is no error to be found. The statement is correct.
>
> Are you blue in the face, yet?
I was surprised that you did not disagree with it. It seems that most people
disagree with most everything that I say. One guy clearly does disagree
with everything, I guess that's his thing.
Marc Goodman
> All that Turing proved was that it is impossible to construct a Halt
> Analyzer that always returns the correct result of the Halt Analysis
> to the program (or TM) being analyzed.
No. What he showed is that if a Halt determiner existed,
it could be used to construct a paradox. Since paradoxes
are not allowed, and since the construction is trivial, the
only possible conclusion is that the Halt determiner itself
leads directly to the paradox, so the Halt determiner cannot
exist. You've been told this before by many, but you don't
get it because you think that something is being broken,
and you don't realize that it never worked in the first place
so it couldn't be broken. It has nothing to do with
returning the results or not, it still has to work on _other_
programs that return _their_ results, and those other programs
can be paradoxes if the Halt determiner exists.
Halt determiner leads irrevocably to the construction of a paradox,
so Halt determiner cannot exist. Is that plain enough for you?
>
> Dozens of people will claim that this statement is incorrect. Not one
> person will ever be able to show exactly why and how this statement
> is incorrect.
Just did.
>
> About all that I have gotten in the way of refutation of this statement is
> "you are wrong because you don't know logic". No one ever will
> point out any specific error in the above statement specifically because
> there is no error to be found. The statement is correct.
You sir, are a doofus.
Marc Goodman
> That is a reasonable statement. If I was careful enough to not
> ever make these sorts of errors, then it would take more time
> than I can spare.
LOL, you're trying to overturn one of the central theorems in
computer science, but you can't be bothered to spend the time
to correct your errors. What do you do when you're not working
on this that's so much more important?
Robert Low
Peter Olcott <olc...@worldnet.att.net> wrote:
>"Robert Low" <mtx...@linux.services.coventry.ac.uk> wrote in message
>That is a reasonable statement. If I was careful enough to not
>ever make these sorts of errors, then it would take more time
>than I can spare.
If you were careful enough to construct a refutation of
a correct proof it would take quite a while, that's true.
--
Rob. http://www.mis.coventry.ac.uk/~mtx014/
David C. Ullrich
<olc...@worldnet.att.net> wrote:
>
>"Simon G Best" <s.g....@btopenworld.com> wrote in message news:411185F4...@btopenworld.com...
>> Peter Olcott wrote:
>> >
>> > No one (not even one person) has gone through what I have said
>> > point by point, and found any holes in my conclusion.
>>
>> That's plainly untrue. See the newsgroup comp.theory for plenty of
>> examples.
>>
>> > All that Turing proved was that it is impossible to construct a Halt
>> > Analyzer that always returns the correct result of the Halt Analysis
>> > to the program (or TM) being analyzed.
>>
>> Which means that it's impossible for a halt analyzer to be a Turing Machine.
>
>Any computation that can be carried out by mechanical means can be
>performed by some Turing Machine.
or so everyone tends to believe - this is of course not the sort of
thing that one can prove mathematically. but i'm certain simon's
opinion is that it's so.
>You must have a very limited understanding of this.
how in the world do you deduce -that-? did he ever say that he
thinks that a non-tm mechanical halt analyzer -is- possible?
it's you who has a limited understanding of this [no surprise,
given your limited understanding of so many other things.]
here's an explanation. this one is simple enough that you
may actually undestand it:
when a non-kook says
[i] task T is impossible for a tm
he does not mean to allow the possibility that T might
be possible for some non-tm mechanical calcultion.
quite the opposite - what he actually means is
[ii] task T is impossible by any mechanical calculation.
then why doesn't he say [ii], if that's what he means?
because [ii] is not the sort of thing that can be
proved mathematically, because of the missing definition
of 'mechanical calculation'. the point to talking about
tm's is that they give a way to give a precise statement
of things like [ii].
i mean duh, when you start telling people what they
don't understand you should really be a little more
careful.
>> > Dozens of people will claim that this statement is incorrect. Not one
>> > person will ever be able to show exactly why and how this statement
>> > is incorrect.
>>
>> http://www.turingarchive.org/browse.php/B/12
>>
>> > About all that I have gotten in the way of refutation of this statement is
>> > "you are wrong because you don't know logic".
>>
>> Plainly untrue. See the newsgroup sci.logic for plenty of examples.
>
>Try and quote any. I have not found one single example that was
>not incorrect for one reason or another, or only pointed out an
>error of no consequence to my primary point. Most were merely
>vague generalizations such as "you are wrong because you don't know logic"
the fact that you don't believe the refutations are correct doesn't
make them wrong.
>> > No one ever will
>> > point out any specific error in the above statement specifically because
>> > there is no error to be found. The statement is correct.
>>
>> Are you blue in the face, yet?
>
>I was surprised that you did not disagree with it. It seems that most people
>disagree with most everything that I say. One guy clearly does disagree
>with everything, I guess that's his thing.
>
>> Quite why you think repeated assertions - especially ones which are now
>> plainly false - are going to convince anyone is quite something of a
>> mystery.
>>
>> Simon
>>
>
************************
David C. Ullrich
sorry about the inelegant formatting - typing
one-handed for a few weeks...
tom_usenet
His logic was sound; special relativity "trivially" follows from the
assumption that the speed of light is the same for all observers. By
contrast, I suspect that you can find your way home, but don't have
sound maths and logic.
Tom
Peter Olcott
"David C. Ullrich" <ull...@math.okstate.edu> wrote in message news:kl24h0ps1b3aabdtl...@4ax.com...
> >> > All that Turing proved was that it is impossible to construct a Halt
> >> > Analyzer that always returns the correct result of the Halt Analysis
> >> > to the program (or TM) being analyzed.
> >>
> >> Which means that it's impossible for a halt analyzer to be a Turing Machine.
> >
> >Any computation that can be carried out by mechanical means can be
> >performed by some Turing Machine.
>
> or so everyone tends to believe - this is of course not the sort of
> thing that one can prove mathematically. but i'm certain simon's
> opinion is that it's so.
>
> >You must have a very limited understanding of this.
>
> how in the world do you deduce -that-? did he ever say that he
> thinks that a non-tm mechanical halt analyzer -is- possible?
If a task can be accomplished in C++, (and I have already shown that)
therefore a task can be accomplished by a TM. To state that a task
that can be accomplished in C++ can not be accomplished by a TM,
is to show a fundamental lack of understanding of TM's.
Peter Olcott
"tom_usenet" <tom_u...@hotmail.com> wrote in message news:7q64h0ho0jb01nh87...@4ax.com...
It is sound enough that no one has been able to show any error
in any of my basic reasoning. Several people from the sci.logic
group thought that sound reasoning was crazy idea. Who would
ever think that one should have true premises? (they thought).
Robert Low
Peter Olcott <olc...@worldnet.att.net> wrote:
>It is sound enough that no one has been able to show any error
>in any of my basic reasoning. Several people from the sci.logic
>group thought that sound reasoning was crazy idea. Who would
>ever think that one should have true premises? (they thought).
Who would ever think that you would think a proof
by contradiction should have true premises?
tom_usenet
<olc...@worldnet.att.net> wrote:
>
>"tom_usenet" <tom_u...@hotmail.com> wrote in message news:7q64h0ho0jb01nh87...@4ax.com...
>> On Thu, 05 Aug 2004 00:15:25 GMT, "Peter Olcott"
>> <olc...@worldnet.att.net> wrote:
>
>> His logic was sound; special relativity "trivially" follows from the
>> assumption that the speed of light is the same for all observers. By
>> contrast, I suspect that you can find your way home, but don't have
>> sound maths and logic.
>>
>> Tom
>
>It is sound enough that no one has been able to show any error
>in any of my basic reasoning.
We've been privately e-mailing the real flaw in your reasoning amongst
ourselves; it's a conspiracy.
Tom
David C. Ullrich
<olc...@worldnet.att.net> wrote:
>
>"David C. Ullrich" <ull...@math.okstate.edu> wrote in message news:kl24h0ps1b3aabdtl...@4ax.com...
>> On Thu, 05 Aug 2004 01:32:11 GMT, "Peter Olcott"
>> <olc...@worldnet.att.net> wrote:
>
>> >> > All that Turing proved was that it is impossible to construct a Halt
>> >> > Analyzer that always returns the correct result of the Halt Analysis
>> >> > to the program (or TM) being analyzed.
>> >>
>> >> Which means that it's impossible for a halt analyzer to be a Turing Machine.
>> >
>> >Any computation that can be carried out by mechanical means can be
>> >performed by some Turing Machine.
>>
>> or so everyone tends to believe - this is of course not the sort of
>> thing that one can prove mathematically. but i'm certain simon's
>> opinion is that it's so.
>>
>> >You must have a very limited understanding of this.
>>
>> how in the world do you deduce -that-? did he ever say that he
>> thinks that a non-tm mechanical halt analyzer -is- possible?
>
>If a task can be accomplished in C++, (and I have already shown that)
>therefore a task can be accomplished by a TM. To state that a task
>that can be accomplished in C++ can not be accomplished by a TM,
>is to show a fundamental lack of understanding of TM's.
who has said anything to the contrary?
Simon G Best
>
> If a task can be accomplished in C++, (and I have already shown that)
> therefore a task can be accomplished by a TM.
And any proof about TMs is also a proof about 'C++' programs (assuming
that the Church-Turing Thesis is correct). Therefore, /assuming that
the Church-Turing Thesis is correct/, your claim of what you've
"accomplished in C++" was disproven in 1936.
> To state that a task
> that can be accomplished in C++ can not be accomplished by a TM,
> is to show a fundamental lack of understanding of TM's.
No. What it /would/ mean is that whoever makes such a statement is
disagreeing with the Church-Turing Thesis.
If you /had/ proved that there is a general solution to the Halting
Problem (HP) for Turing Machines (TMs), then what you'd have proved is
that the Church-Turing Thesis is incorrect, *not* that a TM can be a
halt analyzer.
(As far as I know,) The Church-Turing Thesis is unproven. The fact that
no TM could ever be a halt analyzer for TMs is most certainly proven.
So, if your 'C++' 'solution' to the HP for TMs really worked, then what
you'd've done is to /disprove/ the Church-Turing Thesis.
I should emphasise, however, that I'm not in any way conceding that you
have proved that a general solution to the HP for TMs could exist. I'm
just pointing out that it really makes no sense to refer to the
(unproven) Church-Turing Thesis to try to defend your claimed refutation.
In other words: I'm trying to nudge you in the direction of trying to
disprove the Church-Turing Thesis, rather than continue to waste even
more time trying to refute an irrefutable proof. The /reason/ I'm doing
this is because I'm getting bored of the same stuff being repeated over
and over and over again, and I'd like to move on to the Church-Turing
Thesis. I'm confident that you'll make a complete hash of it as well,
but I'd like to learn more about the Church-Turing Thesis from your many
respondents. So, yes, my motivation is quite selfish :-)
Simon
Robert Low
>Peter Olcott wrote:
>> If a task can be accomplished in C++, (and I have already shown that)
>> therefore a task can be accomplished by a TM.
>And any proof about TMs is also a proof about 'C++' programs (assuming
>that the Church-Turing Thesis is correct). Therefore, /assuming that
I don't think that you have to assume to the Church-Turing hypothesis
to say that. If C++ were more powerful than a TM, it would be
rather tricky to write compilers for it with the available
technology.
Peter Olcott
"Robert Low" <mtx...@linux.services.coventry.ac.uk> wrote in message news:cesml8$p58$1...@sunbeam.coventry.ac.uk...
A correct refutation of a correct proof can not exist.
Peter Olcott
"Marc Goodman" <marc.g...@comcast.net> wrote in message news:XfiQc.246667$Oq2.162585@attbi_s52...
Peter Olcott
"Marc Goodman" <marc.g...@comcast.net> wrote in message news:ibgQc.246159$Oq2.130858@attbi_s52...
> Peter Olcott wrote:
> > All that Turing proved was that it is impossible to construct a Halt
> > Analyzer that always returns the correct result of the Halt Analysis
> > to the program (or TM) being analyzed.
>
> No. What he showed is that if a Halt determiner existed,
> it could be used to construct a paradox.
Yet this paradox that he constructed can only possibly exist under
one specific circumstance and no others. If the halt analyzer simply
refrains from ever providing its result to any program that is being
analyzed, then this paradox becomes completely impossible to construct.
That's exactly all that it takes, it takes nothing at all more than this.
It really is just as simple as that.
Before leaping into yet another round of incorrect refutation, try to first
derive a single counter-example that correctly refutes the statement that
I just made immediately above. Don't try to refute me without this
counter-example, it would just waste more time, both mine and yours.
Marc Goodman
> "Marc Goodman" <marc.g...@comcast.net> wrote in message news:ibgQc.246159$Oq2.130858@attbi_s52...
>
>>Peter Olcott wrote:
>>
>>>All that Turing proved was that it is impossible to construct a Halt
>>>Analyzer that always returns the correct result of the Halt Analysis
>>>to the program (or TM) being analyzed.
>>
>>No. What he showed is that if a Halt determiner existed,
>>it could be used to construct a paradox.
>
>
> Yet this paradox that he constructed can only possibly exist under
> one specific circumstance and no others.
It only HAS to exist under this one specific circumstance
to prove the halt analyzer can never exist. 1+2 only has
to be equal to 3 once for 1 to be equal to 1. It doesn't matter
what 1+3, 1+4, 1+5, etc. are equal to.
If the halt analyzer simply
> refrains from ever providing its result to any program that is being
> analyzed, then this paradox becomes completely impossible to construct.
> That's exactly all that it takes, it takes nothing at all more than this.
> It really is just as simple as that.
OK, you can construct a circumstance where a paradox doesn't
occur. Here's another one: use a fish instead of a halt
analyzer. THis doesn't mean that you can't construct the
paradox given a halt analyzer. Any one paradox proves the
halt analyzer doesn't exist, whether your particular construction
is a paradox or not.
>
> Before leaping into yet another round of incorrect refutation, try to first
> derive a single counter-example that correctly refutes the statement that
> I just made immediately above. Don't try to refute me without this
> counter-example, it would just waste more time, both mine and yours.
I just did.
Bill Taylor
> I think parr has the right question with -1*-1 = -1.
It's a lovely question. It's just that most of the answers are stupid.
"Minus times minus results in a plus,
The reason for this though we need not discuss!"
Oggie
Karl Heinz Buchegger
>
> "Marc Goodman" <marc.g...@comcast.net> wrote in message news:ibgQc.246159$Oq2.130858@attbi_s52...
> > Peter Olcott wrote:
> > > All that Turing proved was that it is impossible to construct a Halt
> > > Analyzer that always returns the correct result of the Halt Analysis
> > > to the program (or TM) being analyzed.
> >
> > No. What he showed is that if a Halt determiner existed,
> > it could be used to construct a paradox.
>
> Yet this paradox that he constructed can only possibly exist under
> one specific circumstance and no others. If the halt analyzer simply
> refrains from ever providing its result to any program that is being
> analyzed,
... then it is no longer a halt analyzer
Thank you. Case closed
--
Karl Heinz Buchegger
kbuc...@gascad.at
raydpratt
corporation, and I scored above 99% of the national average for
salespeople in similar positions. I got scores of perfect fives in
all categories except one, where I got a four, and the category was
where idiotic, fascist, Rhenquistian civil-rights-think has to be
accepted as true premises.
I want that intellectual nigger shot, not mimicked in thought, and I
failed the test.
Very Respectfully,
Ray Donald Pratt
David C. Ullrich
<olc...@worldnet.att.net> wrote:
>
>"Marc Goodman" <marc.g...@comcast.net> wrote in message news:ibgQc.246159$Oq2.130858@attbi_s52...
>> Peter Olcott wrote:
>> > All that Turing proved was that it is impossible to construct a Halt
>> > Analyzer that always returns the correct result of the Halt Analysis
>> > to the program (or TM) being analyzed.
>>
>> No. What he showed is that if a Halt determiner existed,
>> it could be used to construct a paradox.
>
>Yet this paradox that he constructed can only possibly exist under
>one specific circumstance and no others. If the halt analyzer simply
>refrains from ever providing its result to any program that is being
>analyzed, then this paradox becomes completely impossible to construct.
>That's exactly all that it takes, it takes nothing at all more than this.
>It really is just as simple as that.
>
>Before leaping into yet another round of incorrect refutation, try to first
>derive a single counter-example that correctly refutes the statement that
>I just made immediately above.
it's amazing how you seem unable to follow the simple proof that this
is not possible, regardless of how many times it's explained. one
more time:
suppose that P is a program that does what you say - a correct halt
analyzer that refrains from returning the result under some
circumstances. now let Q be a program that does exactly what P
does, except that Q always does return 1 or 0. the proof shows
that Q cannot exist. hence P cannot exist. qed.
Simon G Best
> "Marc Goodman" <marc.g...@comcast.net> wrote in message news:ibgQc.246159$Oq2.130858@attbi_s52...
>
>>Peter Olcott wrote:
>>
>>>All that Turing proved was that it is impossible to construct a Halt
>>>Analyzer that always returns the correct result of the Halt Analysis
>>>to the program (or TM) being analyzed.
>>
>>No. What he showed is that if a Halt determiner existed,
>>it could be used to construct a paradox.
>
>
> Yet this paradox that he constructed can only possibly exist under
> one specific circumstance and no others. If the halt analyzer simply
> refrains from ever providing its result to any program that is being
> analyzed, then this paradox becomes completely impossible to construct.
> That's exactly all that it takes, it takes nothing at all more than this.
> It really is just as simple as that.
If the halt analyzer is a Turing Machine, then it can't know who or what
invoked it. Therefore, it does not know whether or not to refrain from
returning its results. As a Turing Machine, it will *always* behave
*exactly* the same when given the same inputs. Therefore, if it fails
to return a result to a program which it's analyzing, it'll also fail to
return a result (for the same inputs) in all other circumstances. That
means that a Turing Machine can't be a halt analyzer.
> Before leaping into yet another round of incorrect refutation, try to first
> derive a single counter-example that correctly refutes the statement that
> I just made immediately above. Don't try to refute me without this
> counter-example, it would just waste more time, both mine and yours.
The error you keep making is that you rely on a Turing Machine being
able to tell whether or not it was invoked by the program being
analyzed. It can't. Therefore, we do not need to come up with *any*
counter-examples (though we have repeatedly pointed out the
counter-example provided by Alan Turing in 1936). It's already plain
and obvious from the definition of a Turing Machine that it cannot know
who or what invoked it, and that your reliance on it being able to tell
who or what invoked it means that you haven't refuted Turing's proof.
But, to spell it all out again, here's my thought experiment and
accompanying proof:-
Imagine that you are a halt analyzer, WillHalt. You are given a tape
with the following source code and input on it:-
Source code:
bool LoopIfHalts(SourceCode s) {
if (WillHalt(s, s)) {
while(true); // Will loop forever.
return true;
else
return false;
}
Input:
bool LoopIfHalts(SourceCode s) {
if (WillHalt(s, s)) {
while(true); // Will loop forever.
return true;
else
return false;
}
***IMPORTANT:***
However, as WillHalt, you do not know who or what gave you this tape. It
might have been LoopIfHalts; or it might have been someone or something
else.
As a halt analyzer, you have to determine whether or not
LoopIfHalts(LoopIfHalts) would halt, and you have to be able to return
your result. If LoopIfHalts(LoopIfHalts) would halt, return 'true';
otherwise, if it would not halt, return 'false'.
What result, if any, do you return?
A: 'true'.
B: 'false'.
C: No result is returned.
D: A result is returned, but it is neither 'true' nor 'false'.
Bear in mind that, as WillHalt, you do not know whether or not you were
invoked by LoopIfHalts. As a Turing Machine, you simply do not know who
or what gave you that tape.
And now for the proof of my refutation of your claims :-)
(1) WillHalt as a Turing Machine
No matter how many times you would be invoked, you would *never* know
who or what is invoking you. Therefore, whatever your answer is, it
would have to be the same for *all* invocations in which the tapes
you're given are identical to the one above. Otherwise, you would not
be behaving as a Turing Machine, as Turing Machines are incapable of
giving different results for the same inputs (as is obvious from the
definition of Turing Machines).
(2) WillHalt Can't Return 'true'
If your answer is A, 'true', then that would mean what you, as WillHalt,
would return 'true' *regardless* of who or what invoked you (because you
have no way of knowing who or what gave you the tape). That means you
would return 'true' when invoked by LoopIfHalts, and 'true' in all other
invocations. So, when LoopIfHalts(LoopIfHalts) invokes you, you would
return 'true', and LoopIfHalts would then never halt. When invoked as
just WillHalt(LoopIfHalts, LoopIfHalts), you would again return 'true',
but this would now be plainly incorrect. Therefore, if your answer is
A, you would be failing to match the definition of a halt analyzer.
(3) WillHalt Can't Return 'false'
If your answer is B, 'false', then that would mean what you, as
WillHalt, would return 'false' *regardless* of who or what invoked you
(because you have no way of knowing who or what gave you the tape). That
means you would return 'false' when invoked by LoopIfHalts, and 'false'
in all other invocations. So, when LoopIfHalts(LoopIfHalts) invokes
you, you would return 'false', and LoopIfHalts would then halt. When
invoked as just WillHalt(LoopIfHalts, LoopIfHalts), you would again
return 'false', but this would now be plainly incorrect. Therefore, if
your answer is B, you would be failing to match the definition of a halt
analyzer.
(4) WillHalt Must Return Something.
If your answer is C, then you would never return anything for that tape
*regardless* of who or what invoked you (because you have no way of
knowing who or what gave you the tape). Therefore, as you would never
return a result for tapes identical to the one above, you would be
failing to match the definition of a halt analyzer.
(5) WillHalt Must Return either 'true' or 'false'.
If your answer is D, then, whatever your result would be, you would
*always* return that *same* result for that tape *regardless* of who or
what invoked you (because you have no way of knowing who or what gave
you the tape). Therefore, as you would always return that result for
tapes identical to the one above, and as that result would be neither
'true' nor 'false', you would be failing to match the definition of a
halt analyzer.
(6) WillHalt Can't Be a Turing Machine and a Halt Analyzer
As now clearly proven, if WillHalt is a halt analyzer, and if WillHalt
is a Turing Machine (1), it cannot return either 'true' or 'false'
(2,3), but must return either 'true' or 'false' (4,5). This would
clearly be impossible, a plain contradiction. Therefore, it has been
proven that it is impossible for WillHalt to be both a halt analyzer and
a Turing Machine. This proof clearly applies to any replacement for
WillHalt. Therefore, your claim that it is possible for a Turing
Machine to be a halt analyzer is refuted.
(7) Turing Refuted Peter Olcott's Claim in 1936
As this proof here is the same as Turing's proof, it is also now proven
that Turing refuted your claim in 1936 :-)
Simon
David Bandel
i thought you got above 99%? that's hardly failing.. or do you mean
that part was the only part of the test that mattered to them? if so
then it's your own fault. logical implications are not hard to
understand.
Acme Diagnostics
Marc Goodman <marc.g...@comcast.net> wrote:
Hi Marc. Thanks for the kind words you posted a few days ago:
>I volunteer you for the FAQ page ;)
I have a question for you. First, here are two short excerpts from
what you wrote to Peter:
8/5/04
>No. What he showed is that if a Halt determiner existed,
>it could be used to construct a paradox. Since paradoxes
>are not allowed, and since the construction is trivial, the
>only possible conclusion is that the Halt determiner itself
>leads directly to the paradox, so the Halt determiner cannot
>exist. <snip>
>
>Halt determiner leads irrevocably to the construction of a paradox,
>so Halt determiner cannot exist. Is that plain enough for you?
>
>8/6/04:
>
>Exactly. He constructed a paradox using only a Halt
>analyzer, an if statement and goto.
Here is my question:
Is there any *linkable* proof of the Halting Problem for a computer
language that does not contrive a logic paradox and require
self-reference to make the proof?
I have cleaned up the question since posting it to Chris Menzel.
I am comfortable with the word "paradox" here. It seems you are too,
so even though there is probably no definitive answer to the question,
I would like to hear your opinion, or guess if so.
I'm not interested in Turing Machines at all, but only the proof "for
a computer language." That distinction is made explicit on this page:
http://www.csee.umbc.edu/~squire/cs451_l26.html
(I am asking this same question of Owen in a companion post.)
Thanks,
Larry
Kent Paul Dolan
> Is there any *linkable* proof of the Halting
> Problem for a computer language that does not
> contrive a logic paradox and require
> self-reference to make the proof?
> I'm not interested in Turing Machines at all, but
> only the proof "for a computer language." That
> distinction is made explicit on this page:
Since any interesting computer programming language
reduces to a Turing machine (granted at usually huge
losses in performance), then if you are interested
in "computer languages", you are by implication also
interested in Turing machines. The two are joined at
the hip, you can't get one and exclude the other.
As to the proof, what other kind would you expect?
Remember, the claim for a halting problem solver is
that it can accept any program and any input for
that program, and determine whether the former, run
on the latter, halts, and make that determination in
finite time whether the answer is yes or no.
The alternative to doing a proof by contradiction
once and for all that such a beast cannot exist
would seem to be producing a counter example, one by
one, for each of a countable infinity of potential
halting problem solvers.
That is further muddled by the datum that the way
some of them fail would be to produce an incorrect
answer, but the way others of them fail would be to
run without halting.
How would you differentiate this latter group from a
set that just ran longer than the lifetime of the
universe?
You'd need a halting problem solver, and that makes
a loop in logic.
This doesn't seem like an especially fruitful
approach, to put things mildly.
"Proof by contradiction" is one of the strongest
tools of mathematics/logic/computer theory, in
particular because it is one of the few to handle an
infinite number of cases with convenience. Not
having such a tool is crippling. Don't reject it
without an overwhelmingly strong reason to do so.
Mere "dislike" isn't nearly a strong enough reason.
xanthian.
--
Posted via Mailgate.ORG Server - http://www.Mailgate.ORG
Eray Ozkural exa
> Here is my question:
>
> Is there any *linkable* proof of the Halting Problem for a computer
> language that does not contrive a logic paradox and require
> self-reference to make the proof?
You mean proof of the undecidability theorem?
Maybe you were confused by the proof by contradiction as in the
following sketch?
http://en.wikipedia.org/wiki/Halting_problem
I bet Peter would have a hard time understanding any of this post, so
please don't bother explaining anything to him....
I suppose the one you are looking for is Turing's *original* proof.
What self-reference? What logic paradox? I don't think those are
necessary to arrive at the concept of undecidability. Maybe you are
referring to some conceptual tools which we sometimes use to make the
proofs sound familiar or more comprehensible to some people (as it is
sometimes used in an intuitive explanation of the proof of Godel's
first incompleteness theorem?)
This seems to be good enough:
http://www.cse.ohio-state.edu/~gurari/theory-bk/theory-bk-fourse5.html
And please somebody, make the wikipedia entry more informative. It
would be so much easier after telling it to net.kooks like Peter for
several years!
Regards,
--
Eray
Peter Olcott
"David C. Ullrich" <ull...@math.okstate.edu> wrote in message news:f7v6h0dmtd1strl0f...@4ax.com...
Slight little error here that makes all the difference.
You say that the conclusion that Q can't exist entails that P can't exist.
Yet it is not that Q can't exist. Merely that Q sometimes reports results
that are not correct.
This is a perfect example of a (otherwise) very well formed refutation,
that just happens to be incorrect. I think that this is the basic form of
all of the official and published refutations.
This is either the key most crucial point that I am missing, or the key
crucial point that everyone else (except me) has missed for sixty eight
years.
Let's take this ALL THE WAY. Please explain to me exactly and precisely
how is it that Q can not exist? The way that I see it can perfectly well
exist, yet sometimes provide incorrect results. I see nothing at all about
Q that would in any possible way, by any stretch of the imagination
prevent it from existing.
Peter Olcott
> <olc...@worldnet.att.net> wrote:
> it's amazing how you seem unable to follow the simple proof that this
> is not possible, regardless of how many times it's explained. one
> more time:
>
> suppose that P is a program that does what you say - a correct halt
> analyzer that refrains from returning the result under some
> circumstances. now let Q be a program that does exactly what P
> does, except that Q always does return 1 or 0. the proof shows
> that Q cannot exist. hence P cannot exist. qed.
If you mean does exactly what P does AND still provide correct
results everytime?
It is an error to allow what is being studied to in any way effect the
outcome of the study. The only sure way to correctly segregate the
independent and dependent variables is to refrain from providing the
results of the analysis to the test subject. It is a specific error to
provides the results of the study to the test subject in any possible
case (such as the Halting Problem Proof) where the test subject can
in any possible way effect the outcome of the study.
This explains why the different outcomes are possible in the two
different cases, and that the latter case is specifically an error.
It should have been completely obvious long ago that the latter
program is erroneously formed, whereas the former program
is correctly formed, and an incorrect program would not ever
be expected to produce the same results as the correct program.
Marc Goodman
> Here is my question:
>
> Is there any *linkable* proof of the Halting Problem for a computer
> language that does not contrive a logic paradox and require
> self-reference to make the proof?
Beats me :)
I studied this stuff over 20 years ago, before the web even
existed. And, given the nature of the proof, it's not like
I felt the need to check back every few weeks to find out
if there had been any new developments :) Once a decade,
"Oh, the halting problem is still undecidable, what a
surprise" is more than enough.
I hope some younger person who is more familiar with what
current references are available can step in and help you.
> I have cleaned up the question since posting it to Chris Menzel.
> I am comfortable with the word "paradox" here. It seems you are too,
> so even though there is probably no definitive answer to the question,
> I would like to hear your opinion, or guess if so.
Well, I just checked the dictionary after reading what another
poster (Chris Menzel?) had to say about prefering
"contradictory" to "paradoxical." And, at least some of
the meanings for paradox basically say, "reasoning to
contradictory conclusions from acceptable premises based
on valid deduction" and "a seemingly contradictory statement
that may nonetheless be true." I really don't like either
of these meanings when applied to the current discussion,
so I recant my use of the term "paradox" and will make
an effort to use "contradictory" instead.
I can still learn, it just takes me longer and sometimes
my brane starts to smoke and catch fire.
I hope Owen can help more!
-Marc
Marc Goodman
> Let's take this ALL THE WAY. Please explain to me exactly and precisely
> how is it that Q can not exist? The way that I see it can perfectly well
> exist, yet sometimes provide incorrect results. I see nothing at all about
> Q that would in any possible way, by any stretch of the imagination
> prevent it from existing.
The key is the statement, "sometimes provide incorrect results."
Turing was _only_ talking about the case where Q returns correct
results for every single possible input. If you say, "well,
sure, but we can construct Q' that works virtually all the time"
Then, you know what? I agree (at least for certain values
of "virtually all the time.").
Yes, I agree, it is possible to construct a program Q' that
can decide whether some/many/most/virtually all programs
halt or not (at least in principle). Here's the news flash:
Turing would agree too. But, that doesn't disprove Turing's
proof, because he was only talking about a Halt Analyzer that
worked in _all_ cases, and returned its results in _all_ cases.
Such a Halt Analyzer, Q, cannot exist as Turing proved.
So, what you've "proved" is something everyone would agree
with (and say "so what") if you didn't pretend it was a
disproof or refutation of Turing's proof.
This CORRECT REFUTATION has been provided to you MANY times
over the last month. If you get it this time, you can stop
with the bullshit about "no one has successfully refuted my
argument."
Bwahahahaha, like that will ever happen.
Marc Goodman
> If you mean does exactly what P does AND still provide correct
> results everytime?
>
> It is an error to allow what is being studied to in any way effect the
> outcome of the study.
int Factorial(int n)
{
return(n*Factorial(n-1));
}
The function Factorial is being "studied" exactly the same
way LoopIfHalts is being "studied" (it's return value affects
the behavior and return value of the calling function, where
the calling function and the called function are the same.
How is this an error? Or, would you like to clarify exactly
when code is "allowed" to recurse and when it isn't?
The only sure way to correctly segregate the
> independent and dependent variables is to refrain from providing the
> results of the analysis to the test subject. It is a specific error to
> provides the results of the study to the test subject in any possible
> case (such as the Halting Problem Proof) where the test subject can
> in any possible way effect the outcome of the study.
I see your argument has mutated again. Would you please
at least acknowledge that your program DOES NOT return the
correct value for while(Halt(LIH)) before moving on to
the next half-baked scheme to disprove Turing's proof?
Thanks in advance.
> This explains why the different outcomes are possible in the two
> different cases, and that the latter case is specifically an error.
Where's the error exactly? One invocation halts if the other
invocation loops, and vice versa. The code compiles. The
syntax is correct. Where's only a while loop and a function
call. Where is this supposed error?
> It should have been completely obvious long ago that the latter
> program is erroneously formed, whereas the former program
> is correctly formed, and an incorrect program would not ever
> be expected to produce the same results as the correct program.
I still don't see the error. It must be in the function
"Halt," eh?
Marc Goodman
Here is the correct definition for the phrase "Q can't exist":
A halt function Q that correctly returns either 1 or 0 for all
inputs cannot exist. Saying "Q sometimes reports results that
are not correct" EXACTLY matches that definition. If Q sometimes
returns incorrect results, thet Q is NOT a program that correctly
returns either 1 or 0 for all inputs. I know you will find some
loophole where my English can be twisted around into a new
straw-man argument, but I hope I've made it sufficiently difficult
that it takes you some effort this time.
Marc Goodman
> int Factorial(int n)
> {
> return(n*Factorial(n-1));
> }
Oopsie,
int Factorial(int n)
{
if(n==0)
return(1);
return(n*Factorial(n-1));
}
Duh.
Acid Pooh
The fact that he wants "that intellectual nigger shot" means that the
company is better off without him. Let's try to keep the racial slurs
outside of both the workplace _and_ sci.logic, k?
'cid 'ooh
Acme Diagnostics
"Kent Paul Dolan" <xant...@well.com> wrote:
>"Acme Diagnostics" <LFinez...@partpostmark.net> wrote:
>
>> Is there any *linkable* proof of the Halting
>> Problem for a computer language that does not
>> contrive a logic paradox and require
>> self-reference to make the proof?
>
For there not to be one. <g>
Thanks for inferencing my meaning of the word "paradox."
>"Proof by contradiction" is one of the strongest
>tools of mathematics/logic/computer theory, in
>particular because it is one of the few to handle an
>infinite number of cases with convenience. Not
>having such a tool is crippling. Don't reject it
>without an overwhelmingly strong reason to do so.
>
>Mere "dislike" isn't nearly a strong enough reason.
I suspect you are including that for the gribble who may be
overhearing, or perhaps I was ambiguous. I've known about
albino crows since about age 9.
Thanks a lot for answering. I infer there is no link, as you are
the best link finder I know; or you knew there would be no
link. Probably the latter. Either way works for me.
Larry
Acme Diagnostics
Marc Goodman <marc.g...@comcast.net> wrote:
>Acme Diagnostics wrote:
>> Here is my question:
>>
>> Is there any *linkable* proof of the Halting Problem for a computer
>> language that does not contrive a logic paradox and require
>> self-reference to make the proof?
>
>Beats me :)
Good enough that you know of no such link. Thanks.
Larry
David C. Ullrich
<olc...@worldnet.att.net> wrote:
so we're back to the stupidest of your arguments. nobody disputes that
it's possible to make a halt analyzer that doesn't always work. [note
the initial hypothesis 'suppose that P is a program that does what
you say - a correct halt analyzer that refrains from returning the
result under some circumstances.' note the word 'correct']
it's really hard to understand how you can fail to see how stupid
you're being here. it's like i claim that the theorem that there
are no -non-trivial integer solutions to x^3 + y^3 = z^3 is wrong.
i give x = 2, y = 2, z = 2 as a counterexample - somehow i think
this counts as a counterexample even though those values don't
actually satisfy the equation...
>This is a perfect example of a (otherwise) very well formed refutation,
>that just happens to be incorrect. I think that this is the basic form of
>all of the official and published refutations.
>
>This is either the key most crucial point that I am missing, or the key
>crucial point that everyone else (except me) has missed for sixty eight
>years.
and you really think that the second alternative is correct.
totally wacky.
>Let's take this ALL THE WAY. Please explain to me exactly and precisely
>how is it that Q can not exist?
turing proved that a Q -with- the properties i specified cannot
exist. the proof has been presented many times already.
>The way that I see it can perfectly well
>exist, yet sometimes provide incorrect results. I see nothing at all about
>Q that would in any possible way, by any stretch of the imagination
>prevent it from existing.
answer a question for me. with a yes or no: does the python program
def halts(tm, input):
return true
count as a counterexample to the theorem?
Simon G Best
> "David C. Ullrich" <ull...@math.okstate.edu> wrote in message news:f7v6h0dmtd1strl0f...@4ax.com...
>
>>is not possible, regardless of how many times it's explained. one
>>more time:
>>
>>suppose that P is a program that does what you say - a correct halt
>>analyzer that refrains from returning the result under some
>>circumstances. now let Q be a program that does exactly what P
>>does, except that Q always does return 1 or 0. the proof shows
>>that Q cannot exist. hence P cannot exist. qed.
>
> Slight little error here that makes all the difference.
> You say that the conclusion that Q can't exist entails that P can't exist.
> Yet it is not that Q can't exist. Merely that Q sometimes reports results
> that are not correct.
Q always returns what P would return when P is invoked under those
circumstances in which P would return a result. Therefore, if Q
"sometimes reports results that are not correct", then it must be that P
"sometimes reports results that are not correct", and therefore P is not
a halt analyzer.
> This is a perfect example of a (otherwise) very well formed refutation,
> that just happens to be incorrect. I think that this is the basic form of
> all of the official and published refutations.
>
> This is either the key most crucial point that I am missing, or the key
> crucial point that everyone else (except me) has missed for sixty eight
> years.
The most crucial point that you are deliberately missing is that a
Turing Machine doesn't know who or what gave it its input tape. What
David Ullrich is doing is pointing out that even if we overlook this
fact, we can still prove that P cannot exist.
If it comes down to it, we can have Q invoke P, and lie to P by
pretending that it's something else that's invoking P. Turing Machines
only 'know' what we tell them, and we can lie to them (very easily,
because they only have themselves and their tapes to go on, and we have
complete control over what's on the tapes).
> Let's take this ALL THE WAY. Please explain to me exactly and precisely
> how is it that Q can not exist?
That's already been done. See my thought experiment and accompanying
proof. As Q, unlike P, /always/ returns its result /under all
circumstances/, my thought experiment and proof apply to Q.
> The way that I see it can perfectly well
> exist, yet sometimes provide incorrect results. I see nothing at all about
> Q that would in any possible way, by any stretch of the imagination
> prevent it from existing.
Then you are conceding that Turing's proof is correct.
Simon
Simon G Best
> "David C. Ullrich" <ull...@math.okstate.edu> wrote in message news:f7v6h0dmtd1strl0f...@4ax.com...
>
>>On Fri, 06 Aug 2004 01:47:34 GMT, "Peter Olcott"
>><olc...@worldnet.att.net> wrote:
>
>>it's amazing how you seem unable to follow the simple proof that this
>>is not possible, regardless of how many times it's explained. one
>>more time:
>>
>>suppose that P is a program that does what you say - a correct halt
>>analyzer that refrains from returning the result under some
>>circumstances. now let Q be a program that does exactly what P
>>does, except that Q always does return 1 or 0. the proof shows
>>that Q cannot exist. hence P cannot exist. qed.
>
> If you mean does exactly what P does AND still provide correct
> results everytime?
Yes, that's exactly what he means.
> It is an error to allow what is being studied to in any way effect the
> outcome of the study.
It's an "error" that we can 'commit'. The fact that we can 'commit'
such an "error" is what proves that no Turing Machine could ever be a
halt analyzer.
> The only sure way to correctly segregate the
> independent and dependent variables is to refrain from providing the
> results of the analysis to the test subject.
A Turing Machine can only achieve that if it refrains from providing the
results of that analysis in *all* circumstances. After all, its input
tape is the same in *all* such circumstances.
> It is a specific error to
> provides the results of the study to the test subject in any possible
> case (such as the Halting Problem Proof) where the test subject can
> in any possible way effect the outcome of the study.
And we can 'commit' that "error", which is what proves that no Turing
Machine could ever be a halt analyzer. Just asserting that it's Not
Allowed simply isn't a refutation of Turing's proof.
> This explains why the different outcomes are possible in the two
> different cases,
No it doesn't. You're relying on the assertion that it's possible to
try to prove that it's possible. That's not an explanation.
To explain how it's possible, you have to explain how it's possible for
a Turing Machine to behave differently for the same inputs under
different circumstances. It plainly isn't possible, and you know this,
which is obviously why you're ignoring my thought experiment.
> and that the latter case is specifically an error.
An "error" we can commit, just as Turing proved.
> It should have been completely obvious long ago that the latter
> program is erroneously formed, whereas the former program
> is correctly formed, and an incorrect program would not ever
> be expected to produce the same results as the correct program.
So, having failed to refute Turing's proof, you're now resorting to
insisting that the situation Turing used is Not Allowed, even though you
know that such an insistance is a /very/ basic error.
Simon
Peter Olcott
"Simon G Best" <s.g....@btopenworld.com> wrote in message news:4114CC26...@btopenworld.com...
> Peter Olcott wrote:
> > "David C. Ullrich" <ull...@math.okstate.edu> wrote in message news:f7v6h0dmtd1strl0f...@4ax.com...
> >
> >>On Fri, 06 Aug 2004 01:47:34 GMT, "Peter Olcott"
> >><olc...@worldnet.att.net> wrote:
> >
> >>it's amazing how you seem unable to follow the simple proof that this
> >>is not possible, regardless of how many times it's explained. one
> >>more time:
> >>
> >>suppose that P is a program that does what you say - a correct halt
> >>analyzer that refrains from returning the result under some
> >>circumstances. now let Q be a program that does exactly what P
> >>does, except that Q always does return 1 or 0. the proof shows
> >>that Q cannot exist. hence P cannot exist. qed.
> >
> > If you mean does exactly what P does AND still provide correct
> > results everytime?
>
> Yes, that's exactly what he means.
>
> > It is an error to allow what is being studied to in any way effect the
> > outcome of the study.
>
> It's an "error" that we can 'commit'.
So you commit an error in one case and refrain from committing an error
in the second case. Why is it not absurd to expect different results? Why
is it that the erroneous program's behavior can be used to say anything at
all about the correct program's behavior? The correct program can even
correctly process the erroneous program, as input. Because it could not
do this in the past, the prior examples of the erroneous program did
indicate that the correct program could not exist. This is no longer the case.
> The fact that we can 'commit'
> such an "error" is what proves that no Turing Machine could ever be a
> halt analyzer.
>
> > The only sure way to correctly segregate the
> > independent and dependent variables is to refrain from providing the
> > results of the analysis to the test subject.
>
> A Turing Machine can only achieve that if it refrains from providing the
> results of that analysis in *all* circumstances. After all, its input
> tape is the same in *all* such circumstances.
That simply does not follow. RAM is always the same, one cell has the
ability to hold any byte sized piece of data just like the rest. However
there is a clear difference between a simply function call, and a function
call to a function call. In C++ this difference is represented in the program
stack. If we created a C++ to UTM compiler, then this Universal Turing
Machine would have the same stack that C++ has, specifically because
it would implement the C++ stack. The UTM could be defined such that
it's initial position on the tape is zero (the tape having its cells numbered
from zero to infinity, and from zero to negative infinity). In this case Halt
merely looks at the program stack, and if the return address is not zero
it refrains from returning a result. This way it always returns a correct value
when called directly, and, always refrains from returning a value when
called by any other function.
> > It is a specific error to
> > provides the results of the study to the test subject in any possible
> > case (such as the Halting Problem Proof) where the test subject can
> > in any possible way effect the outcome of the study.
>
> And we can 'commit' that "error", which is what proves that no Turing
> Machine could ever be a halt analyzer. Just asserting that it's Not
> Allowed simply isn't a refutation of Turing's proof.
>
> > This explains why the different outcomes are possible in the two
> > different cases,
>
> No it doesn't. You're relying on the assertion that it's possible to
> try to prove that it's possible. That's not an explanation.
>
> To explain how it's possible, you have to explain how it's possible for
> a Turing Machine to behave differently for the same inputs under
> different circumstances. It plainly isn't possible, and you know this,
> which is obviously why you're ignoring my thought experiment.
I just pointed out the step-by-step item-by-item detailed case of
exactly how, where, and why the inputs are not the same. I just
now thought of this case in answer to your specific objection. It
does appear that these dialogues are having a significant beneficial
impact on the quality of my presentation. Thanks for your help.
Peter Olcott
"Simon G Best" <s.g....@btopenworld.com> wrote in message news:4114C80...@btopenworld.com...
> Peter Olcott wrote:
> > "David C. Ullrich" <ull...@math.okstate.edu> wrote in message news:f7v6h0dmtd1strl0f...@4ax.com...
> >
> >>it's amazing how you seem unable to follow the simple proof that this
> >>is not possible, regardless of how many times it's explained. one
> >>more time:
> >>
> >>suppose that P is a program that does what you say - a correct halt
> >>analyzer that refrains from returning the result under some
> >>circumstances. now let Q be a program that does exactly what P
> >>does, except that Q always does return 1 or 0. the proof shows
> >>that Q cannot exist. hence P cannot exist. qed.
> >
> > Slight little error here that makes all the difference.
> > You say that the conclusion that Q can't exist entails that P can't exist.
> > Yet it is not that Q can't exist. Merely that Q sometimes reports results
> > that are not correct.
>
> Q always returns what P would return when P is invoked under those
> circumstances in which P would return a result.
I have already explained exactly why this is not true several hundred times.
Returning the result produces an entirely different execution trace than refraining
from returning a result for the specific program being analyzed.
> Therefore, if Q
> "sometimes reports results that are not correct", then it must be that P
> "sometimes reports results that are not correct", and therefore P is not
> a halt analyzer.
>
> > This is a perfect example of a (otherwise) very well formed refutation,
> > that just happens to be incorrect. I think that this is the basic form of
> > all of the official and published refutations.
> >
> > This is either the key most crucial point that I am missing, or the key
> > crucial point that everyone else (except me) has missed for sixty eight
> > years.
>
> The most crucial point that you are deliberately missing is that a
> Turing Machine doesn't know who or what gave it its input tape. What
This is already specifically addressed on the prior response to you.
> David Ullrich is doing is pointing out that even if we overlook this
> fact, we can still prove that P cannot exist.
>
> If it comes down to it, we can have Q invoke P, and lie to P by
> pretending that it's something else that's invoking P. Turing Machines
> only 'know' what we tell them, and we can lie to them (very easily,
> because they only have themselves and their tapes to go on, and we have
> complete control over what's on the tapes).
This is already specifically addressed on the prior response to you.
> > Let's take this ALL THE WAY. Please explain to me exactly and precisely
> > how is it that Q can not exist?
>
> That's already been done. See my thought experiment and accompanying
> proof. As Q, unlike P, /always/ returns its result /under all
> circumstances/, my thought experiment and proof apply to Q.
>
> > The way that I see it can perfectly well
> > exist, yet sometimes provide incorrect results. I see nothing at all about
> > Q that would in any possible way, by any stretch of the imagination
> > prevent it from existing.
>
> Then you are conceding that Turing's proof is correct.
>
> Simon
>
Not at all. Q and P are not linked.
Peter Olcott
to you. I have already said it all there.
"David C. Ullrich" <ull...@math.okstate.edu> wrote in message news:uad9h0tcdoh2gtjll...@4ax.com...
Peter Olcott
"Acid Pooh" <pooh...@yahoo.com> wrote in message news:d7ba1f79.04080...@posting.google.com...
> dwb...@yahoo.com (David Bandel) wrote in message news:<a88af92f.04080...@posting.google.com>...
> > rayd...@hotmail.com (raydpratt) wrote in message news:<17b9a07c.04080...@posting.google.com>...
> > i thought you got above 99%? that's hardly failing.. or do you mean
> > that part was the only part of the test that mattered to them? if so
> > then it's your own fault. logical implications are not hard to
> > understand.
>
> The fact that he wants "that intellectual nigger shot" means that the
> company is better off without him. Let's try to keep the racial slurs
> outside of both the workplace _and_ sci.logic, k?
>
> 'cid 'ooh
I myself consider that a swear word to never be uttered.
Peter Olcott
"Marc Goodman" <marc.g...@comcast.net> wrote in message news:x6_Qc.258872$Oq2.3255@attbi_s52...
> Peter Olcott wrote:
> > "David C. Ullrich" <ull...@math.okstate.edu> wrote in message news:f7v6h0dmtd1strl0f...@4ax.com...
> >>suppose that P is a program that does what you say - a correct halt
> >>analyzer that refrains from returning the result under some
> >>circumstances. now let Q be a program that does exactly what P
> >>does, except that Q always does return 1 or 0. the proof shows
> >>that Q cannot exist. hence P cannot exist. qed.
> >
> >
> > Slight little error here that makes all the difference.
> > You say that the conclusion that Q can't exist entails that P can't exist.
> > Yet it is not that Q can't exist. Merely that Q sometimes reports results
> > that are not correct.
>
> Here is the correct definition for the phrase "Q can't exist":
> A halt function Q that correctly returns either 1 or 0 for all
> inputs cannot exist.
That's the kind of precise and exact statement that I was looking
for. If everyone could strive to be as precise and exact as this
statement, it would take far fewer dialogues to make any point.
(I really meant the above statement when I wrote it, the statement
below has occurred to me when I was forming my elaboration.)
Whoops. I have already defined a process by which this statement
can be directly refuted. Here is a revised version that I can not
currently directly refute.
A halt function Q that correctly returns either 1 or 0 for all nputs
(in every possible invocation) cannot exist.
I also have a really simple way to determine the type of invocation
in one of my replies to Simon Best.
Peter Olcott
"Marc Goodman" <marc.g...@comcast.net> wrote in message news:T1_Qc.258860$Oq2.27375@attbi_s52...
> Peter Olcott wrote:
> > If you mean does exactly what P does AND still provide correct
> > results everytime?
> >
> > It is an error to allow what is being studied to in any way effect the
> > outcome of the study.
>
> int Factorial(int n)
> {
> return(n*Factorial(n-1));
> }
>
> The function Factorial is being "studied" exactly the same
> way LoopIfHalts is being "studied" (it's return value affects
> the behavior and return value of the calling function, where
> the calling function and the called function are the same.
> How is this an error?
It's not really being studied, like LoopIfHalts is being studied?
> Or, would you like to clarify exactly
> when code is "allowed" to recurse and when it isn't?
>
> The only sure way to correctly segregate the
> > independent and dependent variables is to refrain from providing the
> > results of the analysis to the test subject. It is a specific error to
> > provides the results of the study to the test subject in any possible
> > case (such as the Halting Problem Proof) where the test subject can
> > in any possible way effect the outcome of the study.
>
> I see your argument has mutated again. Would you please
> at least acknowledge that your program DOES NOT return the
> correct value for while(Halt(LIH)) before moving on to
> the next half-baked scheme to disprove Turing's proof?
> Thanks in advance.
In one of my replies to Simon Best, I explained exactly how it could
easily distinguish between the two different invocation types, and thus
be able to correctly determine whether or not LoopIfHalts halts.
By the way, LoopIfHalts still always halts.
> > This explains why the different outcomes are possible in the two
> > different cases, and that the latter case is specifically an error.
>
> Where's the error exactly? One invocation halts if the other
The error (as any scientist would know) is in providing tainting
information to the subject of the study, thus altering the study's
outcome.
Peter Olcott
"Marc Goodman" <marc.g...@comcast.net> wrote in message news:rMYQc.259682$XM6.175024@attbi_s53...
How exactly and precisely does the ability to create Q a program
that sometimes returns incorrect results have anything at all to do
with P? P NEVER returns incorrect results. In those cases where
the results would otherwise be incorrect, it refrains from returning
anything.
In one of my replies to Simon Best, I have shown exactly how the
Halt analyzer could determine when to not return results as opposed
to the single case where it can always return the correct result for every
possible input.
David C. Ullrich
<olc...@worldnet.att.net> wrote:
>Go read my responses to Simon Best's responses to me responding
>to you. I have already said it all there.
the answer to the following question is either yes or no.
typing one or the other of those two words would take a
lot less time than what you've said here.
[if you don't simply answer the question then people will
assume you don't want to, for the obvious reason...]
>"David C. Ullrich" <ull...@math.okstate.edu> wrote in message news:uad9h0tcdoh2gtjll...@4ax.com...
>>[...]
Simon G Best
> "Simon G Best" <s.g....@btopenworld.com> wrote in message news:4114CC26...@btopenworld.com...
>
>>Peter Olcott wrote:
>>
>>>It is an error to allow what is being studied to in any way effect the
>>>outcome of the study.
>>
>>It's an "error" that we can 'commit'.
>
> So you commit an error in one case and refrain from committing an error
> in the second case. Why is it not absurd to expect different results? Why
> is it that the erroneous program's behavior can be used to say anything at
> all about the correct program's behavior? The correct program can even
> correctly process the erroneous program, as input. Because it could not
> do this in the past, the prior examples of the erroneous program did
> indicate that the correct program could not exist. This is no longer the case.
You've lost me. Which Turing Machine is the "erroneous" Turing Machine?
Do you mean Q? Or LoopIfHalts?
>>A Turing Machine can only achieve that if it refrains from providing the
>>results of that analysis in *all* circumstances. After all, its input
>>tape is the same in *all* such circumstances.
>
> That simply does not follow.
Yes it does. Turing Machines are idempotent (in the computing sense).
> RAM is always the same, one cell has the
> ability to hold any byte sized piece of data just like the rest. However
> there is a clear difference between a simply function call, and a function
> call to a function call. In C++ this difference is represented in the program
> stack.
If you rely on what's on the stack, then you're failing to model these
Turing Machines correctly. A Turing Machine cannot tell who or what
gave it its input tape. You have to comply with that in your 'C++'
versions, which means you can't make any such use of the call stack, if
you are to model the relevant Turing Machines correctly.
You've had this error explained to you before, but again you've messed
up the modelling.
> If we created a C++ to UTM compiler, then this Universal Turing
> Machine would have the same stack that C++ has, specifically because
> it would implement the C++ stack. The UTM could be defined such that
> it's initial position on the tape is zero (the tape having its cells numbered
> from zero to infinity, and from zero to negative infinity). In this case Halt
> merely looks at the program stack, and if the return address is not zero
> it refrains from returning a result. This way it always returns a correct value
> when called directly, and, always refrains from returning a value when
> called by any other function.
Okay, I'll write a UTM simulator in C++, and run your UTM C++ programs
on the simulated UTM. As the simulated UTM can't know that it's being
simulated (because all it has is itself and its (simulated) input tape),
we can treat it as being an ordinary WillHalt, and Turing's proof still
applies.
But you've messed up the modelling anyway. This is why I suggested that
you should stop trying to do it in C++ (or whatever you think C++ is),
and instead just do it terms of Turing Machines.
[...]
>
> I just pointed out the step-by-step item-by-item detailed case of
> exactly how, where, and why the inputs are not the same. I just
> now thought of this case in answer to your specific objection. It
> does appear that these dialogues are having a significant beneficial
> impact on the quality of my presentation. Thanks for your help.
No.
The /correct/ way to model your stack-analyzing C++ solution is as follows:-
H is a halt-analyzing Turing Machine which takes /three/ arguments, S,
D, and C. The first, S, is a description of a Turing Machine. The
second, D, is a description of the input to be given to that Turing
Machine. The third, C, is all the other information that H needs in
order for it to work. 'C' can stand for 'context', 'conditions',
'circumstances', 'call-stack', 'computer state', or whatever you like
(as long as it can be encoded as a sequence of symbols on H's input
tape, of course).
I'll use C_0 to refer to whatever C would be when H is called as a
program rather than being called as a subroutine or 'function' within
another program. C_0 could be your zero return address on your
call-stack, for example. You do, of course, get to completely specify
and define what C_0 should be, if you want.
So, when using H as a complete program, rather than as part of another
program, in order to find out whether or not Turing Machine T would halt
on input x, the correct invocation would be 'H(T, x, C_0)'. When using
H as part of another program, C_0 would be the /wrong/ C argument.
I can now define the following Turing Machine, D.
D(x) --> if(H(x, x, C_0)) loop forever; else return false.
Yes! I've given H the wrong C argument! This is Not Allowed, but I'm
doing it anyway. If a Turing Machine H can exist, then so can another
Turing Machine which gives H the wrong C argument. Turing's proof again
applies; therefore, no Turing Machine could ever be a halt analyzer.
So, when you mess up the modelling, I can just use a simulated UTM to
recreate the situation Turing used, and Turing's proof applies. When
your strategy is correctly modelled, I can just define a Turing Machine
which gives H the wrong C argument, and Turing's proof applies.
That's two refutations for the price of one :-)
Simon
Simon G Best
> "Simon G Best" <s.g....@btopenworld.com> wrote in message news:4114C80...@btopenworld.com...
>
>>Peter Olcott wrote:
>>
>>>"David C. Ullrich" <ull...@math.okstate.edu> wrote in message news:f7v6h0dmtd1strl0f...@4ax.com...
>>>
>>>>analyzer that refrains from returning the result under some
>>>>circumstances. now let Q be a program that does exactly what P
>>>>does, except that Q always does return 1 or 0. the proof shows
>>>>that Q cannot exist. hence P cannot exist. qed.
>>>
>>>Slight little error here that makes all the difference.
>>>You say that the conclusion that Q can't exist entails that P can't exist.
>>>Yet it is not that Q can't exist. Merely that Q sometimes reports results
>>>that are not correct.
>>
>>Q always returns what P would return when P is invoked under those
>>circumstances in which P would return a result.
>
> I have already explained exactly why this is not true several hundred times.
And you have got it wrong "several hundred times."
Given P, we can easily construct Q so that Q will do exactly what David
Ullrich said it would do.
We can do this by using a UTM to simulate P as a program, and wrapping
that UTM-with-program up in a standard halt analyzer interface. The
wrapped-up UTM-with-program is then Q. The simulated P won't know that
it's being simulated by the UTM, because the UTM is a UTM.
Simon
Marc Goodman
>>So, what you've "proved" is something everyone would agree
>>with (and say "so what") if you didn't pretend it was a
>>disproof or refutation of Turing's proof.
>>
>>This CORRECT REFUTATION has been provided to you MANY times
>>over the last month. If you get it this time, you can stop
>>with the bullshit about "no one has successfully refuted my
>>argument."
>>
>>Bwahahahaha, like that will ever happen.
>>
>
>
> How exactly and precisely does the ability to create Q a program
> that sometimes returns incorrect results have anything at all to do
> with P?
Because Turing wasn't proving ANYTHING about P. He was
ONLY talking about Q. Turing proved Q could not exist,
he didn't prove that P could not exist, and never claimed
to. Showing P can exist does not disprove Turing's proof
in any way.
P NEVER returns incorrect results. In those cases where
> the results would otherwise be incorrect, it refrains from returning
> anything.
That's exactly why it's not the Turing Machine Turing was
talking about.
> In one of my replies to Simon Best, I have shown exactly how the
> Halt analyzer could determine when to not return results as opposed
> to the single case where it can always return the correct result for every
> possible input.
I believe (for the sake of argument) you can construct P.
I accept that as given.
You still haven't shown this has anything to do with Q, so
you still haven't said _anything_ about Turing's proof.
Daryl McCullough
>How exactly and precisely does the ability to create Q a program
>that sometimes returns incorrect results have anything at all to do
>with P? P NEVER returns incorrect results. In those cases where
>the results would otherwise be incorrect, it refrains from returning
>anything.
Oh, brother. A solution to the halting problem would be a program
H(x,y) that *always* returns a result, and returns true if x is a
code for a program that halts on input y, and returns false if x
is *not* a code for a program that halts on input y.
Do you agree that there is no such program?
Once again, the criteria are these:
1. H(x,y) always returns either true or false (1 or 0, if you prefer)
2. If x is a code for a program with argument that halts on input y,
then H(x,y) returns true (or 1).
3. If x is *not* a code for a program with argument that halts on input
y, then H(x,y) returns false.
Do you agree that there is no program satisfying these criteria?
--
Daryl McCullough
Ithaca, NY
Daryl McCullough
>A halt function Q that correctly returns either 1 or 0 for all nputs
>(in every possible invocation) cannot exist.
That's what Turing proved. That's what it means for the Halting Problem
to be unsolvable.
Marc Goodman
> Peter Olcott says...
>
>>A halt function Q that correctly returns either 1 or 0 for all nputs
>>(in every possible invocation) cannot exist.
>
> That's what Turing proved. That's what it means for the Halting Problem
> to be unsolvable.
>
I expect it will be two days before Peter comes back with
his next attempt. Meanwhile, he will just add this one to
his previous "successful" attempts that were "never refuted."
The City of Defensive Illusion
> On Mon, 02 Aug 2004 23:55:24 GMT, "Peter Olcott"
> <olc...@worldnet.att.net> wrote:
>
> >
> > I don't see any question in here.
> >
> You are blind, right?
My only request regarding this is that all future posts to this thread
include the newsgroup "talk.bizarre". The reason for this is as
follows. When I get done looking at the thread, through either posting
to it, or sniggling snidely as I look upon your minced words with
disdain, I like to quickly get back to talk.bizarre central, by
clicking on it (I use google to access USENET) in one of the subject
lines, and if it is not in the message, then I have to look through
all the messages in the thread until I find one with talk.bizarre
included. Please, make my life empirically more fun. Thanks. Carry on.
A
Jim Burns
That is, is it possible to construct a halting-analyzer for
standard Turning machines using quantum logic gates?
I've been following the Peter Olcott saga somewhat. I think
someone said (many someones said) that Turing's proof only rules
out the construction of a halting-analyzer for Turing machines (or
equivalent) by using a Turing machine (or equivalent). For example,
it would not logically rule out asking an oracle that just knew
whether TM x would halt on input y.
I suspect that quantum computers may be considered stronger
than TMs, since we know they would be able to perform certain tasks
much faster than traditional computers, factoring very large
numbers, for example. I may be wrong about this, since we can
factor these large numbers (in principle, given enough
universe-sized computers), but that is one point I hope to
have cleared up.
It seems appropriate to ask about quantum logic here
in sci.logic. I've heard just a bit about it, for example,
http://plato.stanford.edu/entries/qt-quantlog/
It's not really clear to me what exactly it is, except that it
apparently is not logic, but orthologic. Thus it seems that
quantum computers may not need to follow all the rules that
classical computers do. Another question I have is whether
there is a nuts-and-bolts introduction to using orthologic,
at the level of what would be truth tables in classical logic.
Proofs of soundness and completeness or proofs that QM implies an
orthologic would be fascinating eventually, but I think I may need
to get a gut-level feel for what orthologic is before I can absorb
these deeper proofs.
Jim Burns
Kenneth Doyle
news:Qa5Rc.183363$OB3.1...@bgtnsc05-news.ops.worldnet.att.net:
> Returning the result produces an entirely different execution trace
> than refraining from returning a result for the specific program being
> analyzed.
>
But if you're printing the result on the screen then you are returning a
result. You're just returning it to a different location; a memory
location instead of the accumulator. In TM terms, all you're doing is
printing the result to a different area on the tape.
--
CodeCutter - good, fast and cheap; pick two.
Kenneth Doyle
> In those cases where
> the results would otherwise be incorrect, it refrains from returning
> anything.
>
Not if it prints the result on the screen. In TM terms, that's equivalent
to returning a result.
Peter Olcott
"Kenneth Doyle" <nob...@notmail.com> wrote in message news:Xns953F6F61D18FE...@61.9.191.5...
> "Peter Olcott" <olc...@worldnet.att.net> wrote in
> news:oT5Rc.183485$OB3....@bgtnsc05-news.ops.worldnet.att.net:
>
> > In those cases where
> > the results would otherwise be incorrect, it refrains from returning
> > anything.
> >
>
> Not if it prints the result on the screen. In TM terms, that's equivalent
> to returning a result.
>
In any case I am not talking about putting it on the screen.
That is one way that works. What I am talking about here
is not ever returning anything at all when halt is called from
any other function, and returning 1 or 0 (or true or false)
definitively determining whether the program (or TM) being
analyzed will halt or not, in the cases where halt is not called
from any other function.
Peter Olcott
"Jim Burns" <burn...@osu.edu> wrote in message news:41156FF3...@osu.edu...
> Does quantum mechanics trump Turing's Halting Problem?
> That is, is it possible to construct a halting-analyzer for
> standard Turning machines using quantum logic gates?
>
> I've been following the Peter Olcott saga somewhat. I think
> someone said (many someones said) that Turing's proof only rules
> out the construction of a halting-analyzer for Turing machines (or
> equivalent) by using a Turing machine (or equivalent). For example,
> it would not logically rule out asking an oracle that just knew
> whether TM x would halt on input y.
We don't need to get into that. If we are going to talk about things
such as oracles we might as well talk about fairy god mothers.
Peter Olcott
"Marc Goodman" <marc.g...@comcast.net> wrote in message news:G%aRc.106430$eM2.8698@attbi_s51...
A halt function P that correctly returns either 1 or 0 for all inputs
can exist using my ideas.
The reason that you guys never get me is you keep missing these
subtle little distinctions. Glance at a few words, then refute, refute, refute.
Peter Olcott
"Daryl McCullough" <da...@atc-nycorp.com> wrote in message news:cf3cg...@drn.newsguy.com...
A halt function P that correctly returns either 1 or 0 for all inputs
Peter Olcott
"Daryl McCullough" <da...@atc-nycorp.com> wrote in message news:cf3cc...@drn.newsguy.com...
> Peter Olcott says...
>
> >How exactly and precisely does the ability to create Q a program
> >that sometimes returns incorrect results have anything at all to do
> >with P? P NEVER returns incorrect results. In those cases where
> >the results would otherwise be incorrect, it refrains from returning
> >anything.
>
> Oh, brother. A solution to the halting problem would be a program
> H(x,y) that *always* returns a result,
FALSE ASSUMPTION !!!
> and returns true if x is a
> code for a program that halts on input y, and returns false if x
> is *not* a code for a program that halts on input y.
>
> Do you agree that there is no such program?
Do I agree that water at room temperature is very dry?
(Water is wet, by the way).
> Once again, the criteria are these:
>
> 1. H(x,y) always returns either true or false (1 or 0, if you prefer)
FALSE ASSUMPTION
> 2. If x is a code for a program with argument that halts on input y,
> then H(x,y) returns true (or 1).
> 3. If x is *not* a code for a program with argument that halts on input
> y, then H(x,y) returns false.
>
> Do you agree that there is no program satisfying these criteria?
If this program always returned this value to write only memory,
then your reasoning fails. A program meeting the above criteria
could exist iff it always returned its value to write only memory.
Write only memory is by no means outside of the scope of a TM.
Peter Olcott
"Marc Goodman" <marc.g...@comcast.net> wrote in message news:8f8Rc.243775$JR4.145340@attbi_s54...
> Peter Olcott wrote:
> > How exactly and precisely does the ability to create Q a program
> > that sometimes returns incorrect results have anything at all to do
> > with P?
>
> Because Turing wasn't proving ANYTHING about P. He was
> ONLY talking about Q. Turing proved Q could not exist,
> he didn't prove that P could not exist, and never claimed
> to. Showing P can exist does not disprove Turing's proof
> in any way.
So then you admit that all this discussion about Q is merely
a red herring, and has nothing to do with the Halting Problem?
> P NEVER returns incorrect results. In those cases where
> > the results would otherwise be incorrect, it refrains from returning
> > anything.
>
> That's exactly why it's not the Turing Machine Turing was
> talking about.
Yup he proved the a solution to the Halting Problem was impossible
specifically because he found one artificially contrived example that
did not work. I know of a car that doesn't run, does this prove that
all cars do not run???
> > In one of my replies to Simon Best, I have shown exactly how the
> > Halt analyzer could determine when to not return results as opposed
> > to the single case where it can always return the correct result for every
> > possible input.
>
> I believe (for the sake of argument) you can construct P.
> I accept that as given.
> You still haven't shown this has anything to do with Q, so
> you still haven't said _anything_ about Turing's proof.
>
Q has nothing to do with showing that solving the Halting Problem
is impossible.
Peter Olcott
"Simon G Best" <s.g....@btopenworld.com> wrote in message news:41150737...@btopenworld.com...
> Peter Olcott wrote:
> > I have already explained exactly why this is not true several hundred times.
> [...]
>
> And you have got it wrong "several hundred times."
>
> Given P, we can easily construct Q so that Q will do exactly what David
> Ullrich said it would do.
>
> We can do this by using a UTM to simulate P as a program, and wrapping
> that UTM-with-program up in a standard halt analyzer interface. The
> wrapped-up UTM-with-program is then Q. The simulated P won't know that
> it's being simulated by the UTM, because the UTM is a UTM.
>
> Simon
>
There are so many details that are missing, but, it looks like all
that you are doing is creating a halt analyzer that will not work.
This in no possible way shows in any possible case why my halt
analyzer will not work.
Peter Olcott
"Simon G Best" <s.g....@btopenworld.com> wrote in message news:4114EF1E...@btopenworld.com...
> Peter Olcott wrote:
> > "Simon G Best" <s.g....@btopenworld.com> wrote in message news:4114CC26...@btopenworld.com...
> >
> >>Peter Olcott wrote:
> >>
> >>>It is an error to allow what is being studied to in any way effect the
> >>>outcome of the study.
> >>
> >>It's an "error" that we can 'commit'.
> >
> > So you commit an error in one case and refrain from committing an error
> > in the second case. Why is it not absurd to expect different results? Why
> > is it that the erroneous program's behavior can be used to say anything at
> > all about the correct program's behavior? The correct program can even
> > correctly process the erroneous program, as input. Because it could not
> > do this in the past, the prior examples of the erroneous program did
> > indicate that the correct program could not exist. This is no longer the case.
>
> You've lost me. Which Turing Machine is the "erroneous" Turing Machine?
> Do you mean Q? Or LoopIfHalts?
Q
> >>A Turing Machine can only achieve that if it refrains from providing the
> >>results of that analysis in *all* circumstances. After all, its input
> >>tape is the same in *all* such circumstances.
> >
> > That simply does not follow.
>
> Yes it does. Turing Machines are idempotent (in the computing sense).
So in other words you are merely assuming that the TM can not tell
the difference between different types of invocations. This assumption
is not correct. I have a simple way to prove that it is not correct, yet
will refrain from providing this way until I get agreement that the idea
would otherwise work.
>
> > RAM is always the same, one cell has the
> > ability to hold any byte sized piece of data just like the rest. However
> > there is a clear difference between a simply function call, and a function
> > call to a function call. In C++ this difference is represented in the program
> > stack.
>
> If you rely on what's on the stack, then you're failing to model these
> Turing Machines correctly.
Yup further investigation showed that there were a lot of holes in this
method. This biggest hole was assuming that LoopIfHalts knew or cared
about stack frames. LoopIfHalts could be written in TM native code.
> you should stop trying to do it in C++ (or whatever you think C++ is),
> and instead just do it terms of Turing Machines.
This might be a good idea.
> > I just pointed out the step-by-step item-by-item detailed case of
> > exactly how, where, and why the inputs are not the same.
> No.
Agreed, not yet. I do have a whole other method that would work.
It doesn't bother with C++. It goes straight to the TM level of specification.
> The /correct/ way to model your stack-analyzing C++ solution is as follows:-
I have totally abandoned the whole stack based approach.
> Simon
>
Peter Olcott
"David C. Ullrich" <ull...@math.okstate.edu> wrote in message news:lsq9h092aj9osv58c...@4ax.com...
> On Sat, 07 Aug 2004 14:11:04 GMT, "Peter Olcott"
> <olc...@worldnet.att.net> wrote:
>
> >Go read my responses to Simon Best's responses to me responding
> >to you. I have already said it all there.
>
> the answer to the following question is either yes or no.
> typing one or the other of those two words would take a
> lot less time than what you've said here.
>
> [if you don't simply answer the question then people will
> assume you don't want to, for the obvious reason...]
>
> >"David C. Ullrich" <ull...@math.okstate.edu> wrote in message news:uad9h0tcdoh2gtjll...@4ax.com...
> >>[...]
> >>
> >> answer a question for me. with a yes or no: does the python program
> >>
> >> def halts(tm, input):
> >> return true
> >>
> >> count as a counterexample to the theorem?
All that this seems to be is a halt function that would wrong quite
often, otherwise I have no idea what you are saying?
The Ghost In The Machine
<citydel...@yahoo.com>
wrote
on 7 Aug 2004 14:21:03 -0700
<9a12b70c.04080...@posting.google.com>:
Somehow, that actually works in both sci.logic and talk.bizarre.
There's a certain type of logic to it, and it's bizarre.
Sadly, I don't see comp.theory, but 2 out of 3 ain't bad. :-)
Followups.
--
#191, ewi...@earthlink.net
It's still legal to go .sigless.