Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.theory > #40675 > unrolled thread
| Started by | olcott <NoOne@NoWhere.com> |
|---|---|
| First post | 2021-10-21 10:25 -0500 |
| Last post | 2021-10-21 19:06 -0400 |
| Articles | 20 on this page of 85 — 10 participants |
Back to article view | Back to comp.theory
Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-21 10:25 -0500
Re: Scientific reasoning wij <wyniijj@gmail.com> - 2021-10-21 09:39 -0700
Re: Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-21 12:11 -0500
Re: Scientific reasoning André G. Isaak <agisaak@gm.invalid> - 2021-10-21 11:43 -0600
Re: Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-21 13:03 -0500
Re: Scientific reasoning André G. Isaak <agisaak@gm.invalid> - 2021-10-21 13:58 -0600
Re: Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-21 15:40 -0500
Re: Scientific reasoning André G. Isaak <agisaak@gm.invalid> - 2021-10-21 15:02 -0600
Re: Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-21 16:14 -0500
Re: Scientific reasoning André G. Isaak <agisaak@gm.invalid> - 2021-10-21 15:45 -0600
Re: Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-21 16:59 -0500
Re: Scientific reasoning André G. Isaak <agisaak@gm.invalid> - 2021-10-21 16:12 -0600
Re: Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-21 17:23 -0500
Re: Scientific reasoning Ben Bacarisse <ben.usenet@bsb.me.uk> - 2021-10-21 23:38 +0100
Re: Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-21 18:19 -0500
Re: Scientific reasoning Ben Bacarisse <ben.usenet@bsb.me.uk> - 2021-10-22 03:21 +0100
Re: Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-21 21:34 -0500
Re: Scientific reasoning Richard Damon <Richard@Damon-Family.org> - 2021-10-21 23:08 -0400
Re: Scientific reasoning Ben Bacarisse <ben.usenet@bsb.me.uk> - 2021-10-23 00:41 +0100
Re: Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-22 23:03 -0500
Re: Scientific reasoning Richard Damon <Richard@Damon-Family.org> - 2021-10-23 09:06 -0400
Re: Scientific reasoning Ben Bacarisse <ben.usenet@bsb.me.uk> - 2021-10-24 16:23 +0100
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-24 17:27 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Richard Damon <Richard@Damon-Family.org> - 2021-10-24 19:18 -0400
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Ben Bacarisse <ben.usenet@bsb.me.uk> - 2021-10-26 01:22 +0100
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-25 19:36 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Richard Damon <Richard@Damon-Family.org> - 2021-10-25 20:54 -0400
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Ben Bacarisse <ben.usenet@bsb.me.uk> - 2021-10-26 15:28 +0100
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-26 09:47 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2021-10-26 08:58 -0700
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-26 12:13 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Ben Bacarisse <ben.usenet@bsb.me.uk> - 2021-10-27 00:19 +0100
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-26 19:05 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Ben Bacarisse <ben.usenet@bsb.me.uk> - 2021-10-27 14:15 +0100
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-27 09:21 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Ben Bacarisse <ben.usenet@bsb.me.uk> - 2021-10-27 21:07 +0100
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Richard Damon <Richard@Damon-Family.org> - 2021-10-27 19:46 -0400
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-27 20:39 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Richard Damon <Richard@Damon-Family.org> - 2021-10-27 21:55 -0400
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-27 21:18 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Richard Damon <Richard@Damon-Family.org> - 2021-10-27 22:51 -0400
Re: Scientific reasoning [ if H is correct then it is not incorrect ] André G. Isaak <agisaak@gm.invalid> - 2021-10-27 21:31 -0600
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-27 23:14 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] André G. Isaak <agisaak@gm.invalid> - 2021-10-27 22:32 -0600
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-27 23:55 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Richard Damon <news.x.richarddamon@xoxy.net> - 2021-10-28 06:26 -0400
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Ben Bacarisse <ben.usenet@bsb.me.uk> - 2021-10-26 01:10 +0100
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-25 19:21 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Ben Bacarisse <ben.usenet@bsb.me.uk> - 2021-10-27 00:47 +0100
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-26 19:14 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Richard Damon <Richard@Damon-Family.org> - 2021-10-26 22:27 -0400
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Ben Bacarisse <ben.usenet@bsb.me.uk> - 2021-10-27 14:12 +0100
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-27 08:51 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Ben Bacarisse <ben.usenet@bsb.me.uk> - 2021-10-27 17:38 +0100
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Richard Damon <Richard@Damon-Family.org> - 2021-10-27 19:34 -0400
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Malcolm McLean <malcolm.arthur.mclean@gmail.com> - 2021-10-28 06:52 -0700
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-28 14:13 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-28 10:11 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Daniel Pehoushek <pehoushek1@gmail.com> - 2021-10-28 08:36 -0700
Re: Scientific reasoning [ if H is correct then it is not incorrect ] olcott <NoOne@NoWhere.com> - 2021-10-28 11:26 -0500
Re: Scientific reasoning [ if H is correct then it is not incorrect ] Richard Damon <Richard@Damon-Family.org> - 2021-10-28 20:06 -0400
Re: Scientific reasoning Alan Mackenzie <acm@muc.de> - 2021-10-22 09:55 +0000
Re: Scientific reasoning olcott <polcott2@gmail.com> - 2021-10-22 10:13 -0500
Re: Scientific reasoning Alan Mackenzie <acm@muc.de> - 2021-10-22 17:06 +0000
Re: Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-22 12:20 -0500
Re: Scientific reasoning Richard Damon <Richard@Damon-Family.org> - 2021-10-22 19:15 -0400
Re: Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-22 23:00 -0500
Re: Scientific reasoning Richard Damon <Richard@Damon-Family.org> - 2021-10-23 05:53 -0400
Re: Scientific reasoning André G. Isaak <agisaak@gm.invalid> - 2021-10-21 16:55 -0600
Re: Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-21 18:22 -0500
Re: Scientific reasoning André G. Isaak <agisaak@gm.invalid> - 2021-10-21 18:48 -0600
Re: Scientific reasoning olcott <NoOne@NoWhere.com> - 2021-10-21 20:02 -0500
Re: Scientific reasoning Richard Damon <Richard@Damon-Family.org> - 2021-10-22 08:04 -0400
Re: Scientific reasoning H(P,P) and Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ olcott <NoOne@NoWhere.com> - 2021-10-22 08:33 -0500
Re: Scientific reasoning H(P,P) and Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ Richard Damon <Richard@Damon-Family.org> - 2021-10-22 19:12 -0400
Re: Scientific reasoning H(P,P) and Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ olcott <NoOne@NoWhere.com> - 2021-10-22 22:58 -0500
Re: Scientific reasoning H(P,P) and Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ Richard Damon <Richard@Damon-Family.org> - 2021-10-23 05:59 -0400
Re: Scientific reasoning H(P,P) and Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ olcott <NoOne@NoWhere.com> - 2021-10-23 08:44 -0500
Re: Scientific reasoning H(P,P) and Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ Richard Damon <Richard@Damon-Family.org> - 2021-10-23 11:46 -0400
Re: Scientific reasoning H(P,P) and Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ [ infinite loops ] olcott <NoOne@NoWhere.com> - 2021-10-23 11:59 -0500
Re: Scientific reasoning H(P,P) and Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ [ infinite loops ] Richard Damon <Richard@Damon-Family.org> - 2021-10-23 13:13 -0400
Re: Scientific reasoning H(P,P) and Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ [ infinite loops ] olcott <NoOne@NoWhere.com> - 2021-10-23 12:53 -0500
Re: Scientific reasoning H(P,P) and Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ [ infinite loops ] Richard Damon <Richard@Damon-Family.org> - 2021-10-23 14:16 -0400
Re: Scientific reasoning Richard Damon <Richard@Damon-Family.org> - 2021-10-21 23:13 -0400
Re: Scientific reasoning Richard Damon <Richard@Damon-Family.org> - 2021-10-21 19:06 -0400
Page 2 of 5 — ← Prev page 1 [2] 3 4 5 Next page →
| From | Richard Damon <Richard@Damon-Family.org> |
|---|---|
| Date | 2021-10-23 09:06 -0400 |
| Message-ID | <_MTcJ.2929$k35.881@fx33.iad> |
| In reply to | #40736 |
On 10/23/21 12:03 AM, olcott wrote: > On 10/22/2021 6:41 PM, Ben Bacarisse wrote: >> olcott <NoOne@NoWhere.com> writes: >> >>> On 10/21/2021 9:21 PM, Ben Bacarisse wrote: >>>> olcott <NoOne@NoWhere.com> writes: >>>> >>>>> On 10/21/2021 5:38 PM, Ben Bacarisse wrote: >>>>>> olcott <NoOne@NoWhere.com> writes: >>>>>> >>>>>>> q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn >>>>>>> We can tell that Ĥq0 correctly determines the halt status of ⟨Ĥ⟩ ⟨Ĥ⟩ >>>>>>> entirely on the basis of the meaning of the words. >>>>>> We can tell, based entirely on how Ĥ is defined, that no such Ĥ can >>>>>> exit. >>>>> >>>>> You can only "tell" this by making sure to dismiss what I say >>>>> out-of-hand without carefully evaluating it point-by-point. >>>> >>>> Everyone (except you) can tell it by reading the proof given in Linz. >>>> Everything Linz says applies to what you consider your very special >>>> "simulating halt deciders". >>>> >>> >>> The halt decider must only correctly decide whether or not its input >>> halts on its input. As long as this decision is correct then it is >>> impossible for anything else to show that the halt decider is incorrect. >>> >>> The halt decider must only correctly decide whether or not its input >>> halts on its input. As long as this decision is correct then it is >>> impossible for anything else to show that the halt decider is incorrect. >>> >>> The halt decider must only correctly decide whether or not its input >>> halts on its input. As long as this decision is correct then it is >>> impossible for anything else to show that the halt decider is incorrect. >>> >>> The halt decider must only correctly decide whether or not its input >>> halts on its input. As long as this decision is correct then it is >>> impossible for anything else to show that the halt decider is incorrect. >>> >>> The halt decider must only correctly decide whether or not its input >>> halts on its input. As long as this decision is correct then it is >>> impossible for anything else to show that the halt decider is incorrect. >> >> You are being childish again. It's not a good look. You should try >> engaging with the point I made. >> > > The words above are like say that when we know that we have a black cat > then we know that we have a cat, and yet you still disagree. > > You can some how show that a black cat is not a cat? But it isn't the case you claim. Your statement is gibberish. The phrase "whether or not its input halts on its input" is an improper sentence. Inputs don't halt or not. The ACTUAL condition the decider is supposed to be using is something like: "The Computation REPRESENTED by its input" or "The Machine REPRESENTED by the first part of its input applied to the rest of its input" Note the key word added, REPRESENTED. The input is the representation of H^ and a second copy of that. THus both of these phrases refer to the actual operation of H^ applied to a representation of H^, which you have previously admitted is Halting. You repeatedly MISINTERPRET these conditions to be the (partial) simulation by H of te input, which is incorrect. That is why you keep on trying to enter your Dog in the Cat show, because you think it is close enough.
[toc] | [prev] | [next] | [standalone]
| From | Ben Bacarisse <ben.usenet@bsb.me.uk> |
|---|---|
| Date | 2021-10-24 16:23 +0100 |
| Message-ID | <87pmru30r1.fsf@bsb.me.uk> |
| In reply to | #40736 |
olcott <NoOne@NoWhere.com> writes: > On 10/22/2021 6:41 PM, Ben Bacarisse wrote: >> olcott <NoOne@NoWhere.com> writes: >> >>> On 10/21/2021 9:21 PM, Ben Bacarisse wrote: >>>> olcott <NoOne@NoWhere.com> writes: >>>> >>>>> On 10/21/2021 5:38 PM, Ben Bacarisse wrote: >>>>>> olcott <NoOne@NoWhere.com> writes: >>>>>> >>>>>>> q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn >>>>>>> We can tell that Ĥq0 correctly determines the halt status of ⟨Ĥ⟩ ⟨Ĥ⟩ >>>>>>> entirely on the basis of the meaning of the words. >>>>>> We can tell, based entirely on how Ĥ is defined, that no such Ĥ can >>>>>> exit. >>>>> >>>>> You can only "tell" this by making sure to dismiss what I say >>>>> out-of-hand without carefully evaluating it point-by-point. >>>> >>>> Everyone (except you) can tell it by reading the proof given in Linz. >>>> Everything Linz says applies to what you consider your very special >>>> "simulating halt deciders". >>>> >>> >>> The halt decider must only correctly decide whether or not its input halts on its input. As long as this decision is correct then it is >>> impossible for anything else to show that the halt decider is incorrect. >>> >>> The halt decider must only correctly decide whether or not its input halts on its input. As long as this decision is correct then it is >>> impossible for anything else to show that the halt decider is incorrect. >>> >>> The halt decider must only correctly decide whether or not its input halts on its input. As long as this decision is correct then it is >>> impossible for anything else to show that the halt decider is incorrect. >>> >>> The halt decider must only correctly decide whether or not its input halts on its input. As long as this decision is correct then it is >>> impossible for anything else to show that the halt decider is incorrect. >>> >>> The halt decider must only correctly decide whether or not its input halts on its input. As long as this decision is correct then it is >>> impossible for anything else to show that the halt decider is incorrect. >> >> You are being childish again. It's not a good look. You should try >> engaging with the point I made. > > The words above are like say that when we know that we have a black > cat then we know that we have a cat, and yet you still disagree. We know there are black cats. We know they are all cats. We know there are no halt deciders using a few trivial steps that you pretend to not (or perhaps really don't) understand. It's rather obvious that can't fault the logic in the proof, because you don't even try. At the key step you simply remove the condition that applies and you replace it with your own words. That's not how mathematics works! If you write the line q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn and you follow that with any condition other than "if Ĥ applied ⟨Ĥ⟩ does not halt" then you are not talking about Linz's Ĥ. You can't specify what it /is/ you are talking about (14 years trying attest to that) but it's not Linz's Ĥ with any other condition than the one he writes. > You can some how show that a black cat is not a cat? A simulating halt decider is a member of the class of halt deciders. That's an empty class. >> You've blown the excuse that you only care about this one case, because >> you are referring to Linz's Ĥ. That is an empty class of TMs because of >> how it is specified. If you want to cling to the "only this one case" >> argument you must stop talking about Linz's nonexistent TMs and define >> your own. Unless you admit you are not talking about Linz's Ĥ I will >> keep pointing out why you are wrong about it. -- Ben.
[toc] | [prev] | [next] | [standalone]
| From | olcott <NoOne@NoWhere.com> |
|---|---|
| Date | 2021-10-24 17:27 -0500 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <cOCdnTGmbLdHQOj8nZ2dnUU7-fvNnZ2d@giganews.com> |
| In reply to | #40772 |
On 10/24/2021 10:23 AM, Ben Bacarisse wrote:
> olcott <NoOne@NoWhere.com> writes:
>
>> On 10/22/2021 6:41 PM, Ben Bacarisse wrote:
>>> olcott <NoOne@NoWhere.com> writes:
>>>
>>>> On 10/21/2021 9:21 PM, Ben Bacarisse wrote:
>>>>> olcott <NoOne@NoWhere.com> writes:
>>>>>
>>>>>> On 10/21/2021 5:38 PM, Ben Bacarisse wrote:
>>>>>>> olcott <NoOne@NoWhere.com> writes:
>>>>>>>
>>>>>>>> q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn
>>>>>>>> We can tell that Ĥq0 correctly determines the halt status of ⟨Ĥ⟩ ⟨Ĥ⟩
>>>>>>>> entirely on the basis of the meaning of the words.
>>>>>>> We can tell, based entirely on how Ĥ is defined, that no such Ĥ can
>>>>>>> exit.
>>>>>>
>>>>>> You can only "tell" this by making sure to dismiss what I say
>>>>>> out-of-hand without carefully evaluating it point-by-point.
>>>>>
>>>>> Everyone (except you) can tell it by reading the proof given in Linz.
>>>>> Everything Linz says applies to what you consider your very special
>>>>> "simulating halt deciders".
>>>>>
>>>>
>>>> The halt decider must only correctly decide whether or not its input halts on its input. As long as this decision is correct then it is
>>>> impossible for anything else to show that the halt decider is incorrect.
>>>>
>>>> The halt decider must only correctly decide whether or not its input halts on its input. As long as this decision is correct then it is
>>>> impossible for anything else to show that the halt decider is incorrect.
>>>>
>>>> The halt decider must only correctly decide whether or not its input halts on its input. As long as this decision is correct then it is
>>>> impossible for anything else to show that the halt decider is incorrect.
>>>>
>>>> The halt decider must only correctly decide whether or not its input halts on its input. As long as this decision is correct then it is
>>>> impossible for anything else to show that the halt decider is incorrect.
>>>>
>>>> The halt decider must only correctly decide whether or not its input halts on its input. As long as this decision is correct then it is
>>>> impossible for anything else to show that the halt decider is incorrect.
>>>
>>> You are being childish again. It's not a good look. You should try
>>> engaging with the point I made.
>>
>> The words above are like say that when we know that we have a black
>> cat then we know that we have a cat, and yet you still disagree.
>
> We know there are black cats. We know they are all cats. We know there
> are no halt deciders using a few trivial steps that you pretend to not
> (or perhaps really don't) understand.
>
Of course when you say there are no halt decider you do not mean that
they are no halt deciders you mean that there are no universal halt
deciders. It would be better if you say exactly what you mean and mean
exactly what you say.
void Infinite_Loop()
{
HERE: goto HERE;
}
_Infinite_Loop()
[00000ab0](01) 55 push ebp
[00000ab1](02) 8bec mov ebp,esp
[00000ab3](02) ebfe jmp 00000ab3
[00000ab5](01) 5d pop ebp
[00000ab6](01) c3 ret
Size in bytes:(0007) [00000ab6]
Surely you would agree that the above infinite loop can be correctly
decided as not halting.
> It's rather obvious that can't fault the logic in the proof, because you
> don't even try.
As long as I can show how the impossible input is correctly decided as
not halting I have totally refuted the proof. No more details than this
are required.
If someone claims X cannot possibly be done and I show X being done,
then I have totally proved them wrong. That you disagree with this is
either ridiculous or dishonest.
> At the key step you simply remove the condition that
> applies and you replace it with your own words. That's not how
> mathematics works!
> If you write the line
>
> q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn
>
> and you follow that with any condition other than "if Ĥ applied ⟨Ĥ⟩ does
> not halt" then you are not talking about Linz's Ĥ.
If the halt decider at Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ correctly determines that its input
never halts then even the fact that Ĥ applied to ⟨Ĥ⟩ halts cannot
possibly contradict this.
YOU DISAGREE WITH THIS:
If X is correct about Y then X is not incorrect about Y.
If the halt decider at Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ correctly determines that its input
never halts then even the fact that Ĥ applied to ⟨Ĥ⟩ halts cannot
possibly contradict this.
YOU DISAGREE WITH THIS:
If X is correct about Y then X is not incorrect about Y.
If the halt decider at Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ correctly determines that its input
never halts then even the fact that Ĥ applied to ⟨Ĥ⟩ halts cannot
possibly contradict this.
YOU DISAGREE WITH THIS:
If X is correct about Y then X is not incorrect about Y.
--
Copyright 2021 Pete Olcott
"Great spirits have always encountered violent opposition from mediocre
minds." Einstein
[toc] | [prev] | [next] | [standalone]
| From | Richard Damon <Richard@Damon-Family.org> |
|---|---|
| Date | 2021-10-24 19:18 -0400 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <_QldJ.33878$1n1.18946@fx48.iad> |
| In reply to | #40782 |
On 10/24/21 6:27 PM, olcott wrote:
> On 10/24/2021 10:23 AM, Ben Bacarisse wrote:
>> olcott <NoOne@NoWhere.com> writes:
>>
>>> On 10/22/2021 6:41 PM, Ben Bacarisse wrote:
>>>> olcott <NoOne@NoWhere.com> writes:
>>>>
>>>>> On 10/21/2021 9:21 PM, Ben Bacarisse wrote:
>>>>>> olcott <NoOne@NoWhere.com> writes:
>>>>>>
>>>>>>> On 10/21/2021 5:38 PM, Ben Bacarisse wrote:
>>>>>>>> olcott <NoOne@NoWhere.com> writes:
>>>>>>>>
>>>>>>>>> q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn
>>>>>>>>> We can tell that Ĥq0 correctly determines the halt status of
>>>>>>>>> ⟨Ĥ⟩ ⟨Ĥ⟩
>>>>>>>>> entirely on the basis of the meaning of the words.
>>>>>>>> We can tell, based entirely on how Ĥ is defined, that no such Ĥ can
>>>>>>>> exit.
>>>>>>>
>>>>>>> You can only "tell" this by making sure to dismiss what I say
>>>>>>> out-of-hand without carefully evaluating it point-by-point.
>>>>>>
>>>>>> Everyone (except you) can tell it by reading the proof given in Linz.
>>>>>> Everything Linz says applies to what you consider your very special
>>>>>> "simulating halt deciders".
>>>>>>
>>>>>
>>>>> The halt decider must only correctly decide whether or not its
>>>>> input halts on its input. As long as this decision is correct then
>>>>> it is
>>>>> impossible for anything else to show that the halt decider is
>>>>> incorrect.
>>>>>
>>>>> The halt decider must only correctly decide whether or not its
>>>>> input halts on its input. As long as this decision is correct then
>>>>> it is
>>>>> impossible for anything else to show that the halt decider is
>>>>> incorrect.
>>>>>
>>>>> The halt decider must only correctly decide whether or not its
>>>>> input halts on its input. As long as this decision is correct then
>>>>> it is
>>>>> impossible for anything else to show that the halt decider is
>>>>> incorrect.
>>>>>
>>>>> The halt decider must only correctly decide whether or not its
>>>>> input halts on its input. As long as this decision is correct then
>>>>> it is
>>>>> impossible for anything else to show that the halt decider is
>>>>> incorrect.
>>>>>
>>>>> The halt decider must only correctly decide whether or not its
>>>>> input halts on its input. As long as this decision is correct then
>>>>> it is
>>>>> impossible for anything else to show that the halt decider is
>>>>> incorrect.
>>>>
>>>> You are being childish again. It's not a good look. You should try
>>>> engaging with the point I made.
>>>
>>> The words above are like say that when we know that we have a black
>>> cat then we know that we have a cat, and yet you still disagree.
>>
>> We know there are black cats. We know they are all cats. We know there
>> are no halt deciders using a few trivial steps that you pretend to not
>> (or perhaps really don't) understand.
>>
>
> Of course when you say there are no halt decider you do not mean that
> they are no halt deciders you mean that there are no universal halt
> deciders. It would be better if you say exactly what you mean and mean
> exactly what you say.
He means there are no even partial halt deciders that get the H^
computation correctly.
Yours doesn't (it says non-halting when the actual computation is
Halting), so you still haven't shown a member of that even partial set.
>
> void Infinite_Loop()
> {
> HERE: goto HERE;
> }
>
> _Infinite_Loop()
> [00000ab0](01) 55 push ebp
> [00000ab1](02) 8bec mov ebp,esp
> [00000ab3](02) ebfe jmp 00000ab3
> [00000ab5](01) 5d pop ebp
> [00000ab6](01) c3 ret
> Size in bytes:(0007) [00000ab6]
>
> Surely you would agree that the above infinite loop can be correctly
> decided as not halting.
But that isn't the computation that you claim to be able to disprove
Linz with.
>
>> It's rather obvious that can't fault the logic in the proof, because you
>> don't even try.
>
> As long as I can show how the impossible input is correctly decided as
> not halting I have totally refuted the proof. No more details than this
> are required.
Except that you haven't proved that the computation is non-halting, in
fact you have public agreed that it is halting.
You only make the absurd claim that a wrong answer is somehow because H
hasn't gone far enough to prove it Halting.
>
> If someone claims X cannot possibly be done and I show X being done,
> then I have totally proved them wrong. That you disagree with this is
> either ridiculous or dishonest.
>
Except that you HAVEN'T proved that X can be done.
You have provided a Halt Decider that says that a computaton that
actually halts in short order is non-halting.
All you have proved is that you H can't actually prove that this
computation is Halting, so it erroneously calls it non-halting.
>> At the key step you simply remove the condition that
>> applies and you replace it with your own words. That's not how
>> mathematics works!
>
>> If you write the line
>>
>> q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn
>>
>> and you follow that with any condition other than "if Ĥ applied ⟨Ĥ⟩ does
>> not halt" then you are not talking about Linz's Ĥ.
>
> If the halt decider at Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ correctly determines that its input
> never halts then even the fact that Ĥ applied to ⟨Ĥ⟩ halts cannot
> possibly contradict this.
Except that H^(<H^>) does halt Your line shows it getting to H^qn which
is a halting state, so it IS Halting.
>
> YOU DISAGREE WITH THIS:
> If X is correct about Y then X is not incorrect about Y.
Except that X IS incorrect and you haven't actually proved that it was
correct.
You refuse to answer to comments pointing out the flaws in your
reasoning, so your proof is still discredited.
>
>
>
> If the halt decider at Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ correctly determines that its input
> never halts then even the fact that Ĥ applied to ⟨Ĥ⟩ halts cannot
> possibly contradict this.
?? Facts can't argue with claims?
Isn't that the whole basis of your arguement? You are trying to provide
a Halt decider that actually gets the right answer to disprove the prove
that it is impossible, since you can't find the actual error in the proof.
>
> YOU DISAGREE WITH THIS:
> If X is correct about Y then X is not incorrect about Y.
>
>
But it isn't.
>
> If the halt decider at Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ correctly determines that its input
> never halts then even the fact that Ĥ applied to ⟨Ĥ⟩ halts cannot
> possibly contradict this.
>
But the Computation that the input is a representation of DOES Halt, and
that is what the question is about.
You still don't understand that NO input halts or doesn't halt, only
machines. Data doesn't have a 'Halting Property', only machines applied
to data. You use the wrong machine applied to the data.
> YOU DISAGREE WITH THIS:
> If X is correct about Y then X is not incorrect about Y.
>
No, it isn't. Maybe you have shown that X is correcct about Z, but it
isn;t correct about Y.
[toc] | [prev] | [next] | [standalone]
| From | Ben Bacarisse <ben.usenet@bsb.me.uk> |
|---|---|
| Date | 2021-10-26 01:22 +0100 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <87lf2g1vpb.fsf@bsb.me.uk> |
| In reply to | #40783 |
Richard Damon <Richard@Damon-Family.org> writes: > On 10/24/21 6:27 PM, olcott wrote: >> On 10/24/2021 10:23 AM, Ben Bacarisse wrote: >>> olcott <NoOne@NoWhere.com> writes: >>>> The words above are like say that when we know that we have a black >>>> cat then we know that we have a cat, and yet you still disagree. >>> >>> We know there are black cats. We know they are all cats. We know there >>> are no halt deciders using a few trivial steps that you pretend to not >>> (or perhaps really don't) understand. >>> >> Of course when you say there are no halt decider you do not mean that >> they are no halt deciders you mean that there are no universal halt >> deciders. It would be better if you say exactly what you mean and >> mean exactly what you say. > > He means there are no even partial halt deciders that get the H^ > computation correctly. No, I meant what I said that there are no halt deciders. The term "halt decider" (except, perhaps, in some weird contexts) always means a universal one. PO is ether lying about using Ĥ to mean what Linz does, or lying about how it is specified. Linz's Ĥ is derived from a universal decider H. > Yours doesn't... I would rather not talk about "yours" (i.e. PO's thingy -- whatever that is) until he admits he's not using Ĥ to talk about Linz's class of TMs called Ĥ. But that would mean he has to define what he's talking about (the famous "pathological" inputs) and that's eluded him for more than a decade. I think he gets given far too much licence. If he writes Ĥ and claims that it might exist because it comes from a partial halt decider, he should be called out for that. -- Ben.
[toc] | [prev] | [next] | [standalone]
| From | olcott <NoOne@NoWhere.com> |
|---|---|
| Date | 2021-10-25 19:36 -0500 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <mqCdnfN7E-Ue0Or8nZ2dnUU7-N3NnZ2d@giganews.com> |
| In reply to | #40819 |
On 10/25/2021 7:22 PM, Ben Bacarisse wrote: > Richard Damon <Richard@Damon-Family.org> writes: > >> On 10/24/21 6:27 PM, olcott wrote: >>> On 10/24/2021 10:23 AM, Ben Bacarisse wrote: >>>> olcott <NoOne@NoWhere.com> writes: > >>>>> The words above are like say that when we know that we have a black >>>>> cat then we know that we have a cat, and yet you still disagree. >>>> >>>> We know there are black cats. We know they are all cats. We know there >>>> are no halt deciders using a few trivial steps that you pretend to not >>>> (or perhaps really don't) understand. >>>> >>> Of course when you say there are no halt decider you do not mean that >>> they are no halt deciders you mean that there are no universal halt >>> deciders. It would be better if you say exactly what you mean and >>> mean exactly what you say. >> >> He means there are no even partial halt deciders that get the H^ >> computation correctly. > > No, I meant what I said that there are no halt deciders. The term "halt > decider" (except, perhaps, in some weird contexts) always means a > universal one. > > PO is ether lying about using Ĥ to mean what Linz does, or lying about > how it is specified. Linz's Ĥ is derived from a universal decider H. > Linz's "proof" that no universal halt decider exists is entirely based on his belief that H cannot decide ⟨Ĥ⟩ ⟨Ĥ⟩ correctly. >> Yours doesn't... > > I would rather not talk about "yours" (i.e. PO's thingy -- whatever that > is) until he admits he's not using Ĥ to talk about Linz's class of TMs > called Ĥ. But that would mean he has to define what he's talking about > (the famous "pathological" inputs) and that's eluded him for more than a > decade. > > I think he gets given far too much licence. If he writes Ĥ and claims > that it might exist because it comes from a partial halt decider, he > should be called out for that. > -- Copyright 2021 Pete Olcott "Great spirits have always encountered violent opposition from mediocre minds." Einstein
[toc] | [prev] | [next] | [standalone]
| From | Richard Damon <Richard@Damon-Family.org> |
|---|---|
| Date | 2021-10-25 20:54 -0400 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <vlIdJ.486$Kw9.162@fx45.iad> |
| In reply to | #40822 |
On 10/25/21 8:36 PM, olcott wrote: > On 10/25/2021 7:22 PM, Ben Bacarisse wrote: >> Richard Damon <Richard@Damon-Family.org> writes: >> >>> On 10/24/21 6:27 PM, olcott wrote: >>>> On 10/24/2021 10:23 AM, Ben Bacarisse wrote: >>>>> olcott <NoOne@NoWhere.com> writes: >> >>>>>> The words above are like say that when we know that we have a black >>>>>> cat then we know that we have a cat, and yet you still disagree. >>>>> >>>>> We know there are black cats. We know they are all cats. We know >>>>> there >>>>> are no halt deciders using a few trivial steps that you pretend to not >>>>> (or perhaps really don't) understand. >>>>> >>>> Of course when you say there are no halt decider you do not mean that >>>> they are no halt deciders you mean that there are no universal halt >>>> deciders. It would be better if you say exactly what you mean and >>>> mean exactly what you say. >>> >>> He means there are no even partial halt deciders that get the H^ >>> computation correctly. >> >> No, I meant what I said that there are no halt deciders. The term "halt >> decider" (except, perhaps, in some weird contexts) always means a >> universal one. >> >> PO is ether lying about using Ĥ to mean what Linz does, or lying about >> how it is specified. Linz's Ĥ is derived from a universal decider H. >> > > Linz's "proof" that no universal halt decider exists is entirely based > on his belief that H cannot decide ⟨Ĥ⟩ ⟨Ĥ⟩ correctly. And your H doesn't decide correctly, so Linz is still right. Since your H says H(<H^>,<H^>) is non-halting, Linz's H^(<H^>) is Halting, so H is wrong. It doesn't matter that H's simulation never got to that point, the actual computation that the input is a representation of does, and THAT is what matters. This is an inevitable result if H is actually the Computation that is required of it. > > >>> Yours doesn't... >> >> I would rather not talk about "yours" (i.e. PO's thingy -- whatever that >> is) until he admits he's not using Ĥ to talk about Linz's class of TMs >> called Ĥ. But that would mean he has to define what he's talking about >> (the famous "pathological" inputs) and that's eluded him for more than a >> decade. >> >> I think he gets given far too much licence. If he writes Ĥ and claims >> that it might exist because it comes from a partial halt decider, he >> should be called out for that. >> > >
[toc] | [prev] | [next] | [standalone]
| From | Ben Bacarisse <ben.usenet@bsb.me.uk> |
|---|---|
| Date | 2021-10-26 15:28 +0100 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <87k0hzzwqw.fsf@bsb.me.uk> |
| In reply to | #40822 |
olcott <NoOne@NoWhere.com> writes: > Linz's "proof" that no universal halt decider exists is entirely based > on his belief that H cannot decide ⟨Ĥ⟩ ⟨Ĥ⟩ correctly. It's a proof, not a "proof". It only becomes a "proof" when someone finds a mistake in it. And it does not rely on his beliefs. That's not the word to use when talking about simple logic like this: If a TM, H, existed such that Hq0 <M> s ⊦* Hqy if M applied to s halts, and Hq0 <M> s ⊦* Hqn if M applied to s does not halt, we could construct from it an H' such that H'q0 <M> s ⊦* oo if M applied to s halts, and H'q0 <M> s ⊦* H'qn if M applied to s does not halt. And from that we could construct an Ĥ such that Ĥq0 <M> ⊦* Ĥqx <M> <M> ⊦* oo if M applied to <M> halts, and Ĥq0 <M> ⊦* Ĥqx <M> <M> ⊦* Ĥqn if M applied to <M> does not halt. Setting M to Ĥ, so <M> becomes <Ĥ>, we see that the existence of H (as Linz defines it) logically entails that a TM Ĥ that does this Ĥq0 <Ĥ> ⊦* Ĥqx <Ĥ> <Ĥ> ⊦* oo if Ĥ applied to <Ĥ> halts, and Ĥq0 <Ĥ> ⊦* Ĥqx <Ĥ> <Ĥ> ⊦* Ĥqn if Ĥ applied to <Ĥ> does not halt. would also have to exist. But, as Linz says, "this is clearly nonsense". To call such trivial sequence of deductions a belief is just rhetorical spin. -- Ben.
[toc] | [prev] | [next] | [standalone]
| From | olcott <NoOne@NoWhere.com> |
|---|---|
| Date | 2021-10-26 09:47 -0500 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <K_GdnULDGdaciOX8nZ2dnUU7-NnNnZ2d@giganews.com> |
| In reply to | #40853 |
On 10/26/2021 9:28 AM, Ben Bacarisse wrote: > olcott <NoOne@NoWhere.com> writes: > >> Linz's "proof" that no universal halt decider exists is entirely based >> on his belief that H cannot decide ⟨Ĥ⟩ ⟨Ĥ⟩ correctly. > > It's a proof, not a "proof". It only becomes a "proof" when someone > finds a mistake in it. And it does not rely on his beliefs. That's not > the word to use when talking about simple logic like this: > Because this is necessarily true (only a fool would disagree) Simulating Halt Decider Theorem (Olcott 2021): Whenever simulating halt decider H correctly determines that input P never reaches its final state (whether or not its simulation of P is aborted) then H correctly decides that P never halts. As soon as I show that Ĥq0 <Ĥ> <Ĥ> meets the above theorem then Linz has been refuted. Only a fool would disagree. > If a TM, H, existed such that > > Hq0 <M> s ⊦* Hqy if M applied to s halts, and > Hq0 <M> s ⊦* Hqn if M applied to s does not halt, > > we could construct from it an H' such that > > H'q0 <M> s ⊦* oo if M applied to s halts, and > H'q0 <M> s ⊦* H'qn if M applied to s does not halt. > > And from that we could construct an Ĥ such that > > Ĥq0 <M> ⊦* Ĥqx <M> <M> ⊦* oo if M applied to <M> halts, and > Ĥq0 <M> ⊦* Ĥqx <M> <M> ⊦* Ĥqn if M applied to <M> does not halt. > > Setting M to Ĥ, so <M> becomes <Ĥ>, we see that the existence of H (as > Linz defines it) logically entails that a TM Ĥ that does this > > Ĥq0 <Ĥ> ⊦* Ĥqx <Ĥ> <Ĥ> ⊦* oo if Ĥ applied to <Ĥ> halts, and > Ĥq0 <Ĥ> ⊦* Ĥqx <Ĥ> <Ĥ> ⊦* Ĥqn if Ĥ applied to <Ĥ> does not halt. > > would also have to exist. But, as Linz says, "this is clearly > nonsense". > > To call such trivial sequence of deductions a belief is just rhetorical > spin. > -- Copyright 2021 Pete Olcott "Great spirits have always encountered violent opposition from mediocre minds." Einstein
[toc] | [prev] | [next] | [standalone]
| From | Malcolm McLean <malcolm.arthur.mclean@gmail.com> |
|---|---|
| Date | 2021-10-26 08:58 -0700 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <51a1fe9f-3dd5-41cf-95a2-9018c459647fn@googlegroups.com> |
| In reply to | #40855 |
On Tuesday, 26 October 2021 at 15:47:36 UTC+1, olcott wrote: > On 10/26/2021 9:28 AM, Ben Bacarisse wrote: > > olcott <No...@NoWhere.com> writes: > > > >> Linz's "proof" that no universal halt decider exists is entirely based > >> on his belief that H cannot decide ⟨Ĥ⟩ ⟨Ĥ⟩ correctly. > > > > It's a proof, not a "proof". It only becomes a "proof" when someone > > finds a mistake in it. And it does not rely on his beliefs. That's not > > the word to use when talking about simple logic like this: > > > Because this is necessarily true (only a fool would disagree) > > Simulating Halt Decider Theorem (Olcott 2021): > Whenever simulating halt decider H correctly determines that input P > never reaches its final state (whether or not its simulation of P is > aborted) then H correctly decides that P never halts. > Well that hasn't been thought through. Assuming simulating halt decider H to be a Turing machine, if H's simulation of P is not halted, H never terminates, so, using Turing's model, you never see the result of the decision. (In Turing's model, intermediate states of the tape whilst the machine is running can't be treated as results. What looks like a result might be scratch space. You have to wait until the machine has halted to read the result).
[toc] | [prev] | [next] | [standalone]
| From | olcott <NoOne@NoWhere.com> |
|---|---|
| Date | 2021-10-26 12:13 -0500 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <rZ6dnSDSG6ykquX8nZ2dnUU7-fXNnZ2d@giganews.com> |
| In reply to | #40858 |
On 10/26/2021 10:58 AM, Malcolm McLean wrote:
> On Tuesday, 26 October 2021 at 15:47:36 UTC+1, olcott wrote:
>> On 10/26/2021 9:28 AM, Ben Bacarisse wrote:
>>> olcott <No...@NoWhere.com> writes:
>>>
>>>> Linz's "proof" that no universal halt decider exists is entirely based
>>>> on his belief that H cannot decide ⟨Ĥ⟩ ⟨Ĥ⟩ correctly.
>>>
>>> It's a proof, not a "proof". It only becomes a "proof" when someone
>>> finds a mistake in it. And it does not rely on his beliefs. That's not
>>> the word to use when talking about simple logic like this:
>>>
>> Because this is necessarily true (only a fool would disagree)
>>
>> Simulating Halt Decider Theorem (Olcott 2021):
>> Whenever simulating halt decider H correctly determines that input P
>> never reaches its final state (whether or not its simulation of P is
>> aborted) then H correctly decides that P never halts.
>>
> Well that hasn't been thought through. Assuming simulating halt decider H
> to be a Turing machine, if H's simulation of P is not halted, H never terminates,
> so, using Turing's model, you never see the result of the decision.
>> Simulating Halt Decider Theorem (Olcott 2021):
>> Whenever simulating halt decider H correctly determines that input P
WOULD
>> never
REACH
its final state (whether or not its simulation of P is
>> aborted) then H correctly decides that P never halts.
> (In Turing's model, intermediate states of the tape whilst the machine is
> running can't be treated as results. What looks like a result might be scratch
> space. You have to wait until the machine has halted to read the result).
>
void Infinite_Loop()
{
HERE: goto HERE;
}
_Infinite_Loop()
[00000ab0](01) 55 push ebp
[00000ab1](02) 8bec mov ebp,esp
[00000ab3](02) ebfe jmp 00000ab3
[00000ab5](01) 5d pop ebp
[00000ab6](01) c3 ret
Size in bytes:(0007) [00000ab6]
I provide this as a simplest example.
It doesn't take a PhD to understand that the simulation of
_Infinite_Loop() never reaches machine address 0xab6.
// Simplified Linz Ĥ (Linz:1990:319)
// Strachey(1965) CPL translated to C
01 void P(u32 x)
02 {
03 if (H(x, x))
04 HERE: goto HERE;
05 }
The above code can be analyzed using a similar axiomatic basis to
determine that no simulation of the input to H(P,P) ever gets past line 03.
--
Copyright 2021 Pete Olcott
"Great spirits have always encountered violent opposition from mediocre
minds." Einstein
[toc] | [prev] | [next] | [standalone]
| From | Ben Bacarisse <ben.usenet@bsb.me.uk> |
|---|---|
| Date | 2021-10-27 00:19 +0100 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <875ytjz84x.fsf@bsb.me.uk> |
| In reply to | #40855 |
olcott <NoOne@NoWhere.com> writes: > On 10/26/2021 9:28 AM, Ben Bacarisse wrote: >> olcott <NoOne@NoWhere.com> writes: >> >>> Linz's "proof" that no universal halt decider exists is entirely based >>> on his belief that H cannot decide ⟨Ĥ⟩ ⟨Ĥ⟩ correctly. >> It's a proof, not a "proof". It only becomes a "proof" when someone >> finds a mistake in it. And it does not rely on his beliefs. That's not >> the word to use when talking about simple logic like this: > > Because this is necessarily true (only a fool would disagree) > > Simulating Halt Decider Theorem (Olcott 2021): > Whenever simulating halt decider H correctly determines that input P > never reaches its final state (whether or not its simulation of P is > aborted) then H correctly decides that P never halts. What is H? What is P? What is a correct decision? What does it mean to say an input does or does not halt? Linz's proof does not leave any of these things unspecified. He defines what H should do in all cases. He does not introduce new symbols like P without definition. (The last time you used P it as a scrap of C code, not an input string to a Turning machine.) > As soon as I show that Ĥq0 <Ĥ> <Ĥ> meets the above theorem then Linz > has been refuted. Only a fool would disagree. No. You need to point out where Linz goes wrong, not pontificate about an ill-defined "theorem". >> If a TM, H, existed such that >> Hq0 <M> s ⊦* Hqy if M applied to s halts, and >> Hq0 <M> s ⊦* Hqn if M applied to s does not halt, >> we could construct from it an H' such that >> H'q0 <M> s ⊦* oo if M applied to s halts, and >> H'q0 <M> s ⊦* H'qn if M applied to s does not halt. >> And from that we could construct an Ĥ such that >> Ĥq0 <M> ⊦* Ĥqx <M> <M> ⊦* oo if M applied to <M> halts, and >> Ĥq0 <M> ⊦* Ĥqx <M> <M> ⊦* Ĥqn if M applied to <M> does not halt. >> Setting M to Ĥ, so <M> becomes <Ĥ>, we see that the existence of H (as >> Linz defines it) logically entails that a TM Ĥ that does this >> Ĥq0 <Ĥ> ⊦* Ĥqx <Ĥ> <Ĥ> ⊦* oo if Ĥ applied to <Ĥ> halts, and >> Ĥq0 <Ĥ> ⊦* Ĥqx <Ĥ> <Ĥ> ⊦* Ĥqn if Ĥ applied to <Ĥ> does not halt. >> would also have to exist. But, as Linz says, "this is clearly >> nonsense". And there was you chance to point out where Linz goes wrong. Why did you not take it? Because there is no error in the logic and you know it. -- Ben.
[toc] | [prev] | [next] | [standalone]
| From | olcott <NoOne@NoWhere.com> |
|---|---|
| Date | 2021-10-26 19:05 -0500 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <AKCdnfTc5oEsCuX8nZ2dnUU7-RXNnZ2d@giganews.com> |
| In reply to | #40871 |
On 10/26/2021 6:19 PM, Ben Bacarisse wrote: > olcott <NoOne@NoWhere.com> writes: > >> On 10/26/2021 9:28 AM, Ben Bacarisse wrote: >>> olcott <NoOne@NoWhere.com> writes: >>> >>>> Linz's "proof" that no universal halt decider exists is entirely based >>>> on his belief that H cannot decide ⟨Ĥ⟩ ⟨Ĥ⟩ correctly. >>> It's a proof, not a "proof". It only becomes a "proof" when someone >>> finds a mistake in it. And it does not rely on his beliefs. That's not >>> the word to use when talking about simple logic like this: >> >> Because this is necessarily true (only a fool would disagree) >> >> Simulating Halt Decider Theorem (Olcott 2021): >> Whenever simulating halt decider H correctly determines that input P >> never reaches its final state (whether or not its simulation of P is >> aborted) then H correctly decides that P never halts. > > What is H? a simulating halt decider > What is P? input to Simulating Halt Decider What is a correct decision? Reports whether or not simulated input reaches its final state > What does it mean > to say an input does or does not halt? the simulation of the input never reaches its final state (whether or not its simulation of P is aborted) > Linz's proof does not leave any > of these things unspecified. He defines what H should do in all cases. > He does not introduce new symbols like P without definition. (The last > time you used P it as a scrap of C code, not an input string to a > Turning machine.) > >> As soon as I show that Ĥq0 <Ĥ> <Ĥ> meets the above theorem then Linz >> has been refuted. Only a fool would disagree. > > No. You need to point out where Linz goes wrong, not pontificate about > an ill-defined "theorem". Linz is wrong is his final conclusion at the top of page 320. https://www.liarparadox.org/Peter_Linz_HP(Pages_315-320).pdf q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn The input to Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ specifies infinitely nested simulation thus Ĥq0 correctly transitions to Ĥqn. The mistake that Linz makes is that he is wrongly assuming that Ĥq0 is deciding whether or not itself halts. Ĥq0 is NOT deciding whether or not itself halts. Ĥq0 is deciding whether or its input: ⟨Ĥ⟩ ⟨Ĥ⟩ halts. He confused himself by not provided the full specification for Ĥ. He abbreviated it to be concise. -- Copyright 2021 Pete Olcott "Great spirits have always encountered violent opposition from mediocre minds." Einstein
[toc] | [prev] | [next] | [standalone]
| From | Ben Bacarisse <ben.usenet@bsb.me.uk> |
|---|---|
| Date | 2021-10-27 14:15 +0100 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <87cznqy5fe.fsf@bsb.me.uk> |
| In reply to | #40874 |
olcott <NoOne@NoWhere.com> writes:
> On 10/26/2021 6:19 PM, Ben Bacarisse wrote:
>> olcott <NoOne@NoWhere.com> writes:
>>
>>> On 10/26/2021 9:28 AM, Ben Bacarisse wrote:
>>>> olcott <NoOne@NoWhere.com> writes:
>>>>
>>>>> Linz's "proof" that no universal halt decider exists is entirely based
>>>>> on his belief that H cannot decide ⟨Ĥ⟩ ⟨Ĥ⟩ correctly.
>>>> It's a proof, not a "proof". It only becomes a "proof" when someone
>>>> finds a mistake in it. And it does not rely on his beliefs. That's not
>>>> the word to use when talking about simple logic like this:
>>>
>>> Because this is necessarily true (only a fool would disagree)
>>>
>>> Simulating Halt Decider Theorem (Olcott 2021):
>>> Whenever simulating halt decider H correctly determines that input P
>>> never reaches its final state (whether or not its simulation of P is
>>> aborted) then H correctly decides that P never halts.
>> What is H?
> a simulating halt decider
This does not specify H. Is this the behaviour you want or expect H to
exhibit:
Hq0 <M> s ⊦* Hqy if M applied to s halts, and
Hq0 <M> s ⊦* Hqn if M applied to s does not halt?
I doubt it, but please either confirm that when you write H you mean a
TM with the above behaviour, or, if you are using H to mean a TM that
might deviate from this specification, please give the actual
specification. *And stop calling it H!*
>> What is P?
> input to Simulating Halt Decider
>
>> What is a correct decision?
>
> Reports whether or not simulated input reaches its final state
>
>> What does it mean
>> to say an input does or does not halt?
>
> the simulation of the input never reaches its final state
> (whether or not its simulation of P is aborted)
A proper simulation (by a UTM) matches the halting behaviour of the
encoded TM. I.e. this:
Hq0 <M> s ⊦* Hqy if UTM(<M>, s) reaches M's halt state, and
Hq0 <M> s ⊦* Hqn if UTM(<M>, s) does not reach M's halt state
is exactly the same as the much simpler description above so the mystery
word "abort" (TM don't "abort") is hiding those cases where you intend a
decider to be wrong about some inputs.
If there are no inputs your imagined TM gets wrong (i.e. what you call H
really is what Linz calls H) then your "theorem" is just a badly worded
statement about a non-existent class of Turing machines.
>> Linz's proof does not leave any
>> of these things unspecified. He defines what H should do in all cases.
>> He does not introduce new symbols like P without definition. (The last
>> time you used P it as a scrap of C code, not an input string to a
>> Turning machine.)
>>
>>> As soon as I show that Ĥq0 <Ĥ> <Ĥ> meets the above theorem then Linz
>>> has been refuted. Only a fool would disagree.
>> No. You need to point out where Linz goes wrong, not pontificate about
>> an ill-defined "theorem".
>
> Linz is wrong is his final conclusion at the top of page 320.
> https://www.liarparadox.org/Peter_Linz_HP(Pages_315-320).pdf
So the step you think is wrong is to determine what Ĥ applied to <Ĥ>
does by substituting specific values (Ĥ for M and <Ĥ> for <M>) into the
previous, correctly derived equation? That's an odd complaint. What's
wrong with the substitution?
> q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn
Remember to add "if Ĥ applied to ⟨Ĥ⟩ does not halt". Without that
condition, you are not talking about the proof you are trying to
invalidate.
(Aside: this notation is a step backwards. Writing Ĥq0 when you don't
mean state q0 of Ĥ is just silly. You could call in Hq0 or H'q0 because
it is the state that corresponds to the initial states of both H and H',
but Ĥq0 is just an awful name.)
I cut the rest because it does not explain why you reject that final
substitution step. If you think it does, put it back in reply to my
question ("What's wrong with the substitution?") and I'll try to explain
it to you.
--
Ben.
[toc] | [prev] | [next] | [standalone]
| From | olcott <NoOne@NoWhere.com> |
|---|---|
| Date | 2021-10-27 09:21 -0500 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <O42dnaOGdrvN_eT8nZ2dnUU7-SHNnZ2d@giganews.com> |
| In reply to | #40881 |
On 10/27/2021 8:15 AM, Ben Bacarisse wrote:
> olcott <NoOne@NoWhere.com> writes:
>
>> On 10/26/2021 6:19 PM, Ben Bacarisse wrote:
>>> olcott <NoOne@NoWhere.com> writes:
>>>
>>>> On 10/26/2021 9:28 AM, Ben Bacarisse wrote:
>>>>> olcott <NoOne@NoWhere.com> writes:
>>>>>
>>>>>> Linz's "proof" that no universal halt decider exists is entirely based
>>>>>> on his belief that H cannot decide ⟨Ĥ⟩ ⟨Ĥ⟩ correctly.
>>>>> It's a proof, not a "proof". It only becomes a "proof" when someone
>>>>> finds a mistake in it. And it does not rely on his beliefs. That's not
>>>>> the word to use when talking about simple logic like this:
>>>>
>>>> Because this is necessarily true (only a fool would disagree)
>>>>
>>>> Simulating Halt Decider Theorem (Olcott 2021):
>>>> Whenever simulating halt decider H correctly determines that input P
>>>> never reaches its final state (whether or not its simulation of P is
>>>> aborted) then H correctly decides that P never halts.
>>> What is H?
>> a simulating halt decider
>
> This does not specify H. Is this the behaviour you want or expect H to
> exhibit:
>
> Hq0 <M> s ⊦* Hqy if M applied to s halts, and
> Hq0 <M> s ⊦* Hqn if M applied to s does not halt?
>
> I doubt it, but please either confirm that when you write H you mean a
> TM with the above behaviour, or, if you are using H to mean a TM that
> might deviate from this specification, please give the actual
> specification. *And stop calling it H!*
>
>>> What is P?
>> input to Simulating Halt Decider
>>
>>> What is a correct decision?
>>
>> Reports whether or not simulated input reaches its final state
>>
>>> What does it mean
>>> to say an input does or does not halt?
>>
>> the simulation of the input never reaches its final state
>> (whether or not its simulation of P is aborted)
>
> A proper simulation (by a UTM) matches the halting behaviour of the
> encoded TM. I.e. this:
>
> Hq0 <M> s ⊦* Hqy if UTM(<M>, s) reaches M's halt state, and
> Hq0 <M> s ⊦* Hqn if UTM(<M>, s) does not reach M's halt state
>
> is exactly the same as the much simpler description above so the mystery
> word "abort" (TM don't "abort") is hiding those cases where you intend a
> decider to be wrong about some inputs.
>
> If there are no inputs your imagined TM gets wrong (i.e. what you call H
> really is what Linz calls H) then your "theorem" is just a badly worded
> statement about a non-existent class of Turing machines.
>
>>> Linz's proof does not leave any
>>> of these things unspecified. He defines what H should do in all cases.
>>> He does not introduce new symbols like P without definition. (The last
>>> time you used P it as a scrap of C code, not an input string to a
>>> Turning machine.)
>>>
>>>> As soon as I show that Ĥq0 <Ĥ> <Ĥ> meets the above theorem then Linz
>>>> has been refuted. Only a fool would disagree.
>>> No. You need to point out where Linz goes wrong, not pontificate about
>>> an ill-defined "theorem".
>>
>> Linz is wrong is his final conclusion at the top of page 320.
>> https://www.liarparadox.org/Peter_Linz_HP(Pages_315-320).pdf
>
> So the step you think is wrong is to determine what Ĥ applied to <Ĥ>
> does by substituting specific values (Ĥ for M and <Ĥ> for <M>) into the
> previous, correctly derived equation? That's an odd complaint. What's
> wrong with the substitution?
>
>> q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn
>
> Remember to add "if Ĥ applied to ⟨Ĥ⟩ does not halt". Without that
> condition, you are not talking about the proof you are trying to
> invalidate.
>
> (Aside: this notation is a step backwards. Writing Ĥq0 when you don't
> mean state q0 of Ĥ is just silly. You could call in Hq0 or H'q0 because
> it is the state that corresponds to the initial states of both H and H',
> but Ĥq0 is just an awful name.)
>
> I cut the rest because it does not explain why you reject that final
> substitution step. If you think it does, put it back in reply to my
> question ("What's wrong with the substitution?") and I'll try to explain
> it to you.
>
Ĥ.q0 ⟨Ĥ⟩ ⊢* Ĥ.qx ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥ.qn
If the UTM simulation of the input to Ĥqx ⟨Ĥ⟩ ⟨Ĥ⟩ would never reach its
final state (whether or not this simulation is aborted) then it is
necessarily true that Ĥqx transitions to Ĥ.qn correctly.
The mistake that Linz makes on the top of page 320 is that he wrongly
assumes that Ĥ.qx is deciding the halt status of itself.
As long as halt decider H has correctly decided the halt status of its
input P then the halt status of P was correctly decided by H and nothing
in the universe can possibly contradict this.
--
Copyright 2021 Pete Olcott
"Great spirits have always encountered violent opposition from mediocre
minds." Einstein
[toc] | [prev] | [next] | [standalone]
| From | Ben Bacarisse <ben.usenet@bsb.me.uk> |
|---|---|
| Date | 2021-10-27 21:07 +0100 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <87lf2ew7sc.fsf@bsb.me.uk> |
| In reply to | #40883 |
olcott <NoOne@NoWhere.com> writes:
> On 10/27/2021 8:15 AM, Ben Bacarisse wrote:
>> olcott <NoOne@NoWhere.com> writes:
>>
>>> On 10/26/2021 6:19 PM, Ben Bacarisse wrote:
>>>> olcott <NoOne@NoWhere.com> writes:
>>>>
>>>>> On 10/26/2021 9:28 AM, Ben Bacarisse wrote:
>>>>>> olcott <NoOne@NoWhere.com> writes:
>>>>>>
>>>>>>> Linz's "proof" that no universal halt decider exists is entirely based
>>>>>>> on his belief that H cannot decide ⟨Ĥ⟩ ⟨Ĥ⟩ correctly.
>>>>>> It's a proof, not a "proof". It only becomes a "proof" when someone
>>>>>> finds a mistake in it. And it does not rely on his beliefs. That's not
>>>>>> the word to use when talking about simple logic like this:
>>>>>
>>>>> Because this is necessarily true (only a fool would disagree)
>>>>>
>>>>> Simulating Halt Decider Theorem (Olcott 2021):
>>>>> Whenever simulating halt decider H correctly determines that input P
>>>>> never reaches its final state (whether or not its simulation of P is
>>>>> aborted) then H correctly decides that P never halts.
>>>> What is H?
>>> a simulating halt decider
>> This does not specify H. Is this the behaviour you want or expect H to
>> exhibit:
>> Hq0 <M> s ⊦* Hqy if M applied to s halts, and
>> Hq0 <M> s ⊦* Hqn if M applied to s does not halt?
>> I doubt it, but please either confirm that when you write H you mean a
>> TM with the above behaviour, or, if you are using H to mean a TM that
>> might deviate from this specification, please give the actual
>> specification. *And stop calling it H!*
So you won't say? Why not?
>>>> What is P?
>>> input to Simulating Halt Decider
>>>
>>>> What is a correct decision?
>>>
>>> Reports whether or not simulated input reaches its final state
>>>
>>>> What does it mean
>>>> to say an input does or does not halt?
>>>
>>> the simulation of the input never reaches its final state
>>> (whether or not its simulation of P is aborted)
>>
>> A proper simulation (by a UTM) matches the halting behaviour of the
>> encoded TM. I.e. this:
>> Hq0 <M> s ⊦* Hqy if UTM(<M>, s) reaches M's halt state, and
>> Hq0 <M> s ⊦* Hqn if UTM(<M>, s) does not reach M's halt state
>> is exactly the same as the much simpler description above so the mystery
>> word "abort" (TM don't "abort") is hiding those cases where you intend a
>> decider to be wrong about some inputs.
>>
>> If there are no inputs your imagined TM gets wrong (i.e. what you call H
>> really is what Linz calls H) then your "theorem" is just a badly worded
>> statement about a non-existent class of Turing machines.
So you won't say? Why not?
>> I cut the rest because it does not explain why you reject that final
>> substitution step. If you think it does, put it back in reply to my
>> question ("What's wrong with the substitution?") and I'll try to explain
>> it to you.
>
> Ĥ.q0 ⟨Ĥ⟩ ⊢* Ĥ.qx ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥ.qn
>
> If the UTM simulation of the input to Ĥqx ⟨Ĥ⟩ ⟨Ĥ⟩ would never reach
> its final state (whether or not this simulation is aborted) then it is
> necessarily true that Ĥqx transitions to Ĥ.qn correctly.
>
> The mistake that Linz makes on the top of page 320 is that he wrongly
> assumes that Ĥ.qx is deciding the halt status of itself.
>
> As long as halt decider H has correctly decided the halt status of its
> input P then the halt status of P was correctly decided by H and
> nothing in the universe can possibly contradict this.
When Linz writes
Ĥ.q0 ⟨M⟩ ⊢* Ĥ.qx ⟨M⟩ ⟨M⟩ ⊢* oo if M applied to ⟨M⟩ halts
Ĥ.q0 ⟨M⟩ ⊢* Ĥ.qx ⟨M⟩ ⟨M⟩ ⊢* Ĥ.qn if M applied to ⟨M⟩ does not halt
the variable M stands for any Turing machine. Nothing in the argument
limits what Turing machines M might stand for. This generality flows
from the start of the proof to the end. H is defined by what it does
with any encoded M, so H' is also defined by what it does with any
encoded M. Consequently Ĥ's behaviour also applies to any encoded M:
Either Ĥ does not halt when given the encoding of any TM that halts when
given its own encoding, or Ĥ halts when given the encoding of any TM
that does not halt when given its own encoding.
Can you see why what you wrote does not explain your objection? I
accept that you don't like what results from substituting Ĥ for M in
that last step, but nothing you write has anything to do with why you
think it's not a validly logical thing to do.
(As usual, you are wrong about some of the things you say in your
explanation, but none of those misconceptions have anything to do with
why you think you can't substitute Ĥ for M. It would just be a waste of
time to try to correct them, especially as you have refused to do so on
more than one occasion.)
--
Ben.
[toc] | [prev] | [next] | [standalone]
| From | Richard Damon <Richard@Damon-Family.org> |
|---|---|
| Date | 2021-10-27 19:46 -0400 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <DxleJ.5133$SR4.2078@fx43.iad> |
| In reply to | #40883 |
On 10/27/21 10:21 AM, olcott wrote:
> On 10/27/2021 8:15 AM, Ben Bacarisse wrote:
>> olcott <NoOne@NoWhere.com> writes:
>>
>>> On 10/26/2021 6:19 PM, Ben Bacarisse wrote:
>>>> olcott <NoOne@NoWhere.com> writes:
>>>>
>>>>> On 10/26/2021 9:28 AM, Ben Bacarisse wrote:
>>>>>> olcott <NoOne@NoWhere.com> writes:
>>>>>>
>>>>>>> Linz's "proof" that no universal halt decider exists is entirely
>>>>>>> based
>>>>>>> on his belief that H cannot decide ⟨Ĥ⟩ ⟨Ĥ⟩ correctly.
>>>>>> It's a proof, not a "proof". It only becomes a "proof" when someone
>>>>>> finds a mistake in it. And it does not rely on his beliefs.
>>>>>> That's not
>>>>>> the word to use when talking about simple logic like this:
>>>>>
>>>>> Because this is necessarily true (only a fool would disagree)
>>>>>
>>>>> Simulating Halt Decider Theorem (Olcott 2021):
>>>>> Whenever simulating halt decider H correctly determines that input P
>>>>> never reaches its final state (whether or not its simulation of P is
>>>>> aborted) then H correctly decides that P never halts.
>>>> What is H?
>>> a simulating halt decider
>>
>> This does not specify H. Is this the behaviour you want or expect H to
>> exhibit:
>>
>> Hq0 <M> s ⊦* Hqy if M applied to s halts, and
>> Hq0 <M> s ⊦* Hqn if M applied to s does not halt?
>>
>> I doubt it, but please either confirm that when you write H you mean a
>> TM with the above behaviour, or, if you are using H to mean a TM that
>> might deviate from this specification, please give the actual
>> specification. *And stop calling it H!*
>>
>>>> What is P?
>>> input to Simulating Halt Decider
>>>
>>>> What is a correct decision?
>>>
>>> Reports whether or not simulated input reaches its final state
>>>
>>>> What does it mean
>>>> to say an input does or does not halt?
>>>
>>> the simulation of the input never reaches its final state
>>> (whether or not its simulation of P is aborted)
>>
>> A proper simulation (by a UTM) matches the halting behaviour of the
>> encoded TM. I.e. this:
>>
>> Hq0 <M> s ⊦* Hqy if UTM(<M>, s) reaches M's halt state, and
>> Hq0 <M> s ⊦* Hqn if UTM(<M>, s) does not reach M's halt state
>>
>> is exactly the same as the much simpler description above so the mystery
>> word "abort" (TM don't "abort") is hiding those cases where you intend a
>> decider to be wrong about some inputs.
>>
>> If there are no inputs your imagined TM gets wrong (i.e. what you call H
>> really is what Linz calls H) then your "theorem" is just a badly worded
>> statement about a non-existent class of Turing machines.
>>
>>>> Linz's proof does not leave any
>>>> of these things unspecified. He defines what H should do in all cases.
>>>> He does not introduce new symbols like P without definition. (The last
>>>> time you used P it as a scrap of C code, not an input string to a
>>>> Turning machine.)
>>>>
>>>>> As soon as I show that Ĥq0 <Ĥ> <Ĥ> meets the above theorem then Linz
>>>>> has been refuted. Only a fool would disagree.
>>>> No. You need to point out where Linz goes wrong, not pontificate about
>>>> an ill-defined "theorem".
>>>
>>> Linz is wrong is his final conclusion at the top of page 320.
>>> https://www.liarparadox.org/Peter_Linz_HP(Pages_315-320).pdf
>>
>> So the step you think is wrong is to determine what Ĥ applied to <Ĥ>
>> does by substituting specific values (Ĥ for M and <Ĥ> for <M>) into the
>> previous, correctly derived equation? That's an odd complaint. What's
>> wrong with the substitution?
>>
>>> q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn
>>
>> Remember to add "if Ĥ applied to ⟨Ĥ⟩ does not halt". Without that
>> condition, you are not talking about the proof you are trying to
>> invalidate.
>>
>> (Aside: this notation is a step backwards. Writing Ĥq0 when you don't
>> mean state q0 of Ĥ is just silly. You could call in Hq0 or H'q0 because
>> it is the state that corresponds to the initial states of both H and H',
>> but Ĥq0 is just an awful name.)
>>
>> I cut the rest because it does not explain why you reject that final
>> substitution step. If you think it does, put it back in reply to my
>> question ("What's wrong with the substitution?") and I'll try to explain
>> it to you.
>>
>
> Ĥ.q0 ⟨Ĥ⟩ ⊢* Ĥ.qx ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥ.qn
> If the UTM simulation of the input to Ĥqx ⟨Ĥ⟩ ⟨Ĥ⟩ would never reach its
> final state (whether or not this simulation is aborted) then it is
> necessarily true that Ĥqx transitions to Ĥ.qn correctly.
Finally, you use the right words. We Know that UTM(<H^> <H^>) will act
exactly like H^<H^>).
We have that:
q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn
Where the above initial q0 is that starting state of H^, so this is
showing the behavior of H^ (<H^>) which we see ends up at H^qn which is
defined as a terminal state for H^, so this PROVES that H^ applied to
<H^> is a Halting Computation.
>
> The mistake that Linz makes on the top of page 320 is that he wrongly
> assumes that Ĥ.qx is deciding the halt status of itself.
Since H^.qx is where we placed a copy of the code of H, we know by
definition that the behavior from H^.qx will be identical to the
behavior of the machine H. Since you have stipulated that H(<H^><H^>)
will say halting, that means that H.q0 <H^> <H^> goes to H.qn and thus
H^q0 <H^><H^> also goes to H^qn, which as we noted above halts.
>
> As long as halt decider H has correctly decided the halt status of its
> input P then the halt status of P was correctly decided by H and nothing
> in the universe can possibly contradict this.
>
Except that H has said that H^(<H^>) will never halt, when it does, so H
did NOT correct decide the Halting status of the input <H^> <H^>.
The problem is that it made its decision based on UNSOUND logic, by
assuming that the copy of H that it sees inside H^ would NOT abort its
simulation, but, since H is a Computation, we know that this copy of H
inside H^ will behave exactly like the copy of H doing the decision,
since they both have the exact same input.
Since the deciding H will abort its simulation, so will the copy in the
simulation, but the point in time when that will happen will be after
the point in time when the simulating H 'gives up', makes the wrong
decision, and aborts it simulation, so it will never see that happen.
When we simulate that exact same input with a UTM, as your stated
condition above says, THAT will continue to see that simulated H also
abort is simulation and return the non-halting answer (making exactly
the same error as the original H).
[toc] | [prev] | [next] | [standalone]
| From | olcott <NoOne@NoWhere.com> |
|---|---|
| Date | 2021-10-27 20:39 -0500 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <sO2dnUgyToP3YuT8nZ2dnUU7-UXNnZ2d@giganews.com> |
| In reply to | #40892 |
On 10/27/2021 6:46 PM, Richard Damon wrote:
>
> On 10/27/21 10:21 AM, olcott wrote:
>> On 10/27/2021 8:15 AM, Ben Bacarisse wrote:
>>> olcott <NoOne@NoWhere.com> writes:
>>>
>>>> On 10/26/2021 6:19 PM, Ben Bacarisse wrote:
>>>>> olcott <NoOne@NoWhere.com> writes:
>>>>>
>>>>>> On 10/26/2021 9:28 AM, Ben Bacarisse wrote:
>>>>>>> olcott <NoOne@NoWhere.com> writes:
>>>>>>>
>>>>>>>> Linz's "proof" that no universal halt decider exists is entirely
>>>>>>>> based
>>>>>>>> on his belief that H cannot decide ⟨Ĥ⟩ ⟨Ĥ⟩ correctly.
>>>>>>> It's a proof, not a "proof". It only becomes a "proof" when someone
>>>>>>> finds a mistake in it. And it does not rely on his beliefs.
>>>>>>> That's not
>>>>>>> the word to use when talking about simple logic like this:
>>>>>>
>>>>>> Because this is necessarily true (only a fool would disagree)
>>>>>>
>>>>>> Simulating Halt Decider Theorem (Olcott 2021):
>>>>>> Whenever simulating halt decider H correctly determines that input P
>>>>>> never reaches its final state (whether or not its simulation of P is
>>>>>> aborted) then H correctly decides that P never halts.
>>>>> What is H?
>>>> a simulating halt decider
>>>
>>> This does not specify H. Is this the behaviour you want or expect H to
>>> exhibit:
>>>
>>> Hq0 <M> s ⊦* Hqy if M applied to s halts, and
>>> Hq0 <M> s ⊦* Hqn if M applied to s does not halt?
>>>
>>> I doubt it, but please either confirm that when you write H you mean a
>>> TM with the above behaviour, or, if you are using H to mean a TM that
>>> might deviate from this specification, please give the actual
>>> specification. *And stop calling it H!*
>>>
>>>>> What is P?
>>>> input to Simulating Halt Decider
>>>>
>>>>> What is a correct decision?
>>>>
>>>> Reports whether or not simulated input reaches its final state
>>>>
>>>>> What does it mean
>>>>> to say an input does or does not halt?
>>>>
>>>> the simulation of the input never reaches its final state
>>>> (whether or not its simulation of P is aborted)
>>>
>>> A proper simulation (by a UTM) matches the halting behaviour of the
>>> encoded TM. I.e. this:
>>>
>>> Hq0 <M> s ⊦* Hqy if UTM(<M>, s) reaches M's halt state, and
>>> Hq0 <M> s ⊦* Hqn if UTM(<M>, s) does not reach M's halt state
>>>
>>> is exactly the same as the much simpler description above so the mystery
>>> word "abort" (TM don't "abort") is hiding those cases where you intend a
>>> decider to be wrong about some inputs.
>>>
>>> If there are no inputs your imagined TM gets wrong (i.e. what you call H
>>> really is what Linz calls H) then your "theorem" is just a badly worded
>>> statement about a non-existent class of Turing machines.
>>>
>>>>> Linz's proof does not leave any
>>>>> of these things unspecified. He defines what H should do in all
>>>>> cases.
>>>>> He does not introduce new symbols like P without definition. (The
>>>>> last
>>>>> time you used P it as a scrap of C code, not an input string to a
>>>>> Turning machine.)
>>>>>
>>>>>> As soon as I show that Ĥq0 <Ĥ> <Ĥ> meets the above theorem then Linz
>>>>>> has been refuted. Only a fool would disagree.
>>>>> No. You need to point out where Linz goes wrong, not pontificate
>>>>> about
>>>>> an ill-defined "theorem".
>>>>
>>>> Linz is wrong is his final conclusion at the top of page 320.
>>>> https://www.liarparadox.org/Peter_Linz_HP(Pages_315-320).pdf
>>>
>>> So the step you think is wrong is to determine what Ĥ applied to <Ĥ>
>>> does by substituting specific values (Ĥ for M and <Ĥ> for <M>) into the
>>> previous, correctly derived equation? That's an odd complaint. What's
>>> wrong with the substitution?
>>>
>>>> q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn
>>>
>>> Remember to add "if Ĥ applied to ⟨Ĥ⟩ does not halt". Without that
>>> condition, you are not talking about the proof you are trying to
>>> invalidate.
>>>
>>> (Aside: this notation is a step backwards. Writing Ĥq0 when you don't
>>> mean state q0 of Ĥ is just silly. You could call in Hq0 or H'q0 because
>>> it is the state that corresponds to the initial states of both H and H',
>>> but Ĥq0 is just an awful name.)
>>>
>>> I cut the rest because it does not explain why you reject that final
>>> substitution step. If you think it does, put it back in reply to my
>>> question ("What's wrong with the substitution?") and I'll try to explain
>>> it to you.
>>>
>>
>> Ĥ.q0 ⟨Ĥ⟩ ⊢* Ĥ.qx ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥ.qn
>> If the UTM simulation of the input to Ĥqx ⟨Ĥ⟩ ⟨Ĥ⟩ would never reach
>> its final state (whether or not this simulation is aborted) then it is
>> necessarily true that Ĥqx transitions to Ĥ.qn correctly.
>
> Finally, you use the right words. We Know that UTM(<H^> <H^>) will act
> exactly like H^<H^>).
>
> We have that:
> q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn
>
> Where the above initial q0 is that starting state of H^, so this is
> showing the behavior of H^ (<H^>) which we see ends up at H^qn which is
> defined as a terminal state for H^, so this PROVES that H^ applied to
> <H^> is a Halting Computation.
That Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ halts is not the question.
The simulation of ⟨Ĥ⟩ ⟨Ĥ⟩ by Ĥq0 never halts is the answer.
If the halt decider correctly decides the halt status of its input then
it is correct no matter what else:
If when we directly execute Ĥ applied to ⟨Ĥ⟩ and it:
(a) Halts.
(b) Destroys the whole universe.
(c) Converts all fascists to true Christianity.
(d) Circumnavigates the globe.
(e) Cures cancer.
(f) Raises the dead.
It has still correctly decided the halt status of its input.
--
Copyright 2021 Pete Olcott
"Great spirits have always encountered violent opposition from mediocre
minds." Einstein
[toc] | [prev] | [next] | [standalone]
| From | Richard Damon <Richard@Damon-Family.org> |
|---|---|
| Date | 2021-10-27 21:55 -0400 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <3qneJ.3747$VB5.2631@fx02.iad> |
| In reply to | #40893 |
On 10/27/21 9:39 PM, olcott wrote: > On 10/27/2021 6:46 PM, Richard Damon wrote: >> >> On 10/27/21 10:21 AM, olcott wrote: >>> Ĥ.q0 ⟨Ĥ⟩ ⊢* Ĥ.qx ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥ.qn >>> If the UTM simulation of the input to Ĥqx ⟨Ĥ⟩ ⟨Ĥ⟩ would never reach >>> its final state (whether or not this simulation is aborted) then it >>> is necessarily true that Ĥqx transitions to Ĥ.qn correctly. ^^^^^^^^^^^^^^^^^ NOTE WHAT YOU SAID HERE ^^^^^^^^^^^^^^^^^^^^^^^^ >> >> Finally, you use the right words. We Know that UTM(<H^> <H^>) will act >> exactly like H^<H^>). >> >> We have that: >> q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn >> >> Where the above initial q0 is that starting state of H^, so this is >> showing the behavior of H^ (<H^>) which we see ends up at H^qn which >> is defined as a terminal state for H^, so this PROVES that H^ applied >> to <H^> is a Halting Computation. > That Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ halts is not the question. > The simulation of ⟨Ĥ⟩ ⟨Ĥ⟩ by Ĥq0 never halts is the answer. No, as you said earlier, it is what the input to H (or H^q0) would do if it was made the input to a UTM, NOT your 'non-utm' partial simulation H. The right answer to the Halting Problem is ALWAYS about the actual machine who has been converted into a representation to give to the decider, never about what happens in the decider. > > If the halt decider correctly decides the halt status of its input then > it is correct no matter what else: But it DOESN'T decide correctly. So, yes, if it decided correctly, it is correct, but it it decides incorrectly, it is incorrect. > > If when we directly execute Ĥ applied to ⟨Ĥ⟩ and it: > (a) Halts. > (b) Destroys the whole universe. > (c) Converts all fascists to true Christianity. > (d) Circumnavigates the globe. > (e) Cures cancer. > (f) Raises the dead. > > It has still correctly decided the halt status of its input. > Why does the thing DEFINED to be the source of the right answer not matter, but something that isn't defined that we do? You are just showing how ignorant of the problem you are. Occationally, you 'goof' and give the right words, which shows you may actually know the truth, and are just a pathological liar, doomed forever to have to live with your lies.
[toc] | [prev] | [next] | [standalone]
| From | olcott <NoOne@NoWhere.com> |
|---|---|
| Date | 2021-10-27 21:18 -0500 |
| Subject | Re: Scientific reasoning [ if H is correct then it is not incorrect ] |
| Message-ID | <5Nmdna2nv5z_lef8nZ2dnUU7-fudnZ2d@giganews.com> |
| In reply to | #40894 |
On 10/27/2021 8:55 PM, Richard Damon wrote: > On 10/27/21 9:39 PM, olcott wrote: >> On 10/27/2021 6:46 PM, Richard Damon wrote: >>> >>> On 10/27/21 10:21 AM, olcott wrote: >>>> Ĥ.q0 ⟨Ĥ⟩ ⊢* Ĥ.qx ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥ.qn >>>> If the UTM simulation of the input to Ĥqx ⟨Ĥ⟩ ⟨Ĥ⟩ would never reach >>>> its final state (whether or not this simulation is aborted) then it >>>> is necessarily true that Ĥqx transitions to Ĥ.qn correctly. > > ^^^^^^^^^^^^^^^^^ NOTE WHAT YOU SAID HERE ^^^^^^^^^^^^^^^^^^^^^^^^ > >>> >>> Finally, you use the right words. We Know that UTM(<H^> <H^>) will >>> act exactly like H^<H^>). >>> >>> We have that: >>> q0 ⟨Ĥ⟩ ⊢* Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ ⊢* Ĥqn >>> >>> Where the above initial q0 is that starting state of H^, so this is >>> showing the behavior of H^ (<H^>) which we see ends up at H^qn which >>> is defined as a terminal state for H^, so this PROVES that H^ applied >>> to <H^> is a Halting Computation. >> That Ĥq0 ⟨Ĥ⟩ ⟨Ĥ⟩ halts is not the question. >> The simulation of ⟨Ĥ⟩ ⟨Ĥ⟩ by Ĥq0 never halts is the answer. > > No, as you said earlier, it is what the input to H (or H^q0) would do if > it was made the input to a UTM, NOT your 'non-utm' partial simulation H. > > The right answer to the Halting Problem is ALWAYS about the actual > machine who has been converted into a representation to give to the > decider, never about what happens in the decider. So then you are saying that when the halt decider does correctly decide that its input never halts that it can still be wrong? That is irrational. -- Copyright 2021 Pete Olcott "Great spirits have always encountered violent opposition from mediocre minds." Einstein
[toc] | [prev] | [next] | [standalone]
Page 2 of 5 — ← Prev page 1 [2] 3 4 5 Next page →
Back to top | Article view | comp.theory
csiph-web