Re: Can D simulated by H terminate normally?

On 5/2/24 5:55 PM, olcott wrote:

On 5/2/2024 4:07 AM, Alan Mackenzie wrote:

olcott <polcott333@gmail.com> wrote:

On 4/29/2024 1:19 PM, Alan Mackenzie wrote:

olcott <polcott333@gmail.com> wrote:

On 4/29/2024 11:17 AM, Alan Mackenzie wrote:

olcott <polcott333@gmail.com> wrote:

On 4/29/2024 10:23 AM, Alan Mackenzie wrote:

olcott <polcott333@gmail.com> wrote:

On 4/29/2024 9:37 AM, Alan Mackenzie wrote:

In comp.theory olcott <polcott333@gmail.com> wrote:

On 4/28/2024 1:39 PM, Richard Damon wrote:

On 4/28/24 2:19 PM, olcott wrote:

>
[ .... ]
>

A simulating termination analyzer is not the same thing as a UTM.

>

At the moment, on this newsgroup, it's a meaningless term.  You seem to
have used it merely as a synonym for halting decider in the past.  Yet
you fail to say what you mean by it, if you mean something different.

>

Within my brand new idea of a {simulating termination analyzer}
there is the idea of abnormal termination.

>

It's unlikely to be a brand new idea, whatever it might be, since you
aren't familiar with the literature and so most likely have come up with
somebody else's old idea.

>
[ .... ]
>

You've failed, repeatedly, to address the points I've been making in my
last few posts, so it seems that you have accepted them.  In particular,
you have accepted that "having been aborted" is indeed a final state for
a turing machine or a program.

>

All of the "points" that you have been making were entirely anchored in
your ignorance about what "simulating termination analyzer" are and how
they work.

>
I think it much more likely that there's no such thing as a "simulating
termination analyzer".  I.e. there's no theory about it, no interesting
results, no use for it, or anything like that.  I've asked you several
times to define this object, other people have asked you too, yet you
fail to do so.  Producing a few lines of scrappy C code is not anything
like producing a definition.
>

 *It has been a fully operational software system for a few years now*

No, as it gives the wrong answer to the problem that it was written to solve.
Since that that time, your claim WAS that you were working on the actual Halting Problem, but claimed that H(D,D) could be correct saying its input represents a non-halting pattern, even though, by DEFINITION, since D(D) halt it is an incorret answer.
You gave all sorts of LIES about why a wrong answer could be right.
Now, you just obfuscate what you are actually doing by changing some of the terms, without actually defining them, but still making you proven incorrect claims.

 Universal Turing Machine (UTM) having the x86 language as its Machine description language. https://github.com/plolcott/x86utm

And thus you are STILL indicting that this is about the actual Halting Problem and claims that these are "Turing Equivalents" (except they aren't) so your claims are still proven to be lies.
The biggest part is the code you say is the description of the machine D, isn't, as just that piece of code isn't a "program" but a "program fragment" and thus everything after that is just a lie.

 

"Simulating termination analyzer" probably just means halt decider.  We
all know there's no such thing.
>

 *This is probably the best simulating termination analyzer available*
 *AProVE: Non-Termination Witnesses for C Programs*
To prove (non-)termination of a C program, AProVE
uses the Clang compiler [7] to translate it to the
intermediate representation of the LLVM framework [15].
Then AProVE symbolically executes the LLVM program ...
https://verify.rwth-aachen.de/giesl/papers/TACAS22.pdf

Right, and look at what they actually claim.

 

-- Copyright 2024 Olcott "Talent hits a target no one else can hit; Genius
hits a target no one else can see." Arthur Schopenhauer

>