Re: The actual truth is that ... industry standard stipulative definitions

Am Tue, 15 Oct 2024 14:56:01 -0500 schrieb olcott:

On 10/15/2024 2:29 PM, joes wrote:

Am Tue, 15 Oct 2024 14:18:52 -0500 schrieb olcott:

On 10/15/2024 10:32 AM, joes wrote:

Am Tue, 15 Oct 2024 07:33:47 -0500 schrieb olcott:

On 10/15/2024 3:54 AM, Mikko wrote:

On 2024-10-14 16:05:20 +0000, olcott said:
>

A stipulative definition is a type of definition in which a new or
currently existing term is given a new specific meaning for the
purposes of argument or discussion in a given context.
*Disagreeing with a stipulative definition is incorrect*

The Wikipedia page does not say that. It only says that a
stipulative definition itself cannot be correct.

If X cannot be incorrect then disagreeing that X is correct is
incorrect.

Stipulative definitions can also not be correct. Correctness is
simply out of scope. It can be rejected though. Is your best defense
really "it has no truth value"?

It is the same as verifying that a conclusion logically follows form
its premises when hypothesizing that the premises are true.

What is the same?

The article also says that the scope of a stipulative definition is
restricted to an argument or discussion in given context.

Once a stipulated definition is provided by its author it continues
to apply to every use of this term when properly qualified.
A *non_terminating_C_function* is C a function that cannot possibly
reach its own "return" instruction  (final state) thus never
terminates.

And not a function that can't be simulated by HHH.

Meaning, DDD is terminating function, because it reaches its return,
even though HHH can't simulate the call to itself (because a simulator
terminates only when its input does, so it can't halt simulating
itself).

In other words you insist on failing to understand that the behavior of
DDD after HHH aborts its emulation is different than the behavior that
requires HHH to abort its emulation.

WDYM "after"? DDD has time-invariant behaviour may or may not involve
HHH aborting at the programmers discretion. But there is only a single
HHH (even though it may be called in multiple stack frames) which
either does or doesn't. The sequencing you are alluding to only
occurs in the reasoning: if HHH were to abort, DDD would be halting
and it wouldn't have needed aborting: contradiction.

A *correct_x86_emulation* of non-terminating inputs includes at
least N steps of *correct_x86_emulation*.

This qualifies only as a partial simulation. A correct simulation may
not terminate.

A full emulation of a non-terminating input is logically impossible.
Do you not know this?

Of course. The simulation does not terminate.

Then you don't understand that the emulation of DDD by HHH does not
reach its own "return" instruction BECAUSE DDD calld HHH in recursive
emulation?

It has nothing to do with DDD - it halts exactly iff HHH does.
No simulator can completely simulate itself, like Fred said. Although
it could detect it is simulating itself, abort, and return that it halts.

DDD *correctly_emulated_by* HHH refers to a *correct_x86_emulation*.
This also adds that HHH is emulating itself emulating DDD at least
once.
When HHH is an x86 emulation based termination analyzer then each
DDD *correctly_emulated_by* any HHH that it calls never returns.

And HHH is not a decider.

Where in my stipulated definitions did I ever refer to a decider?

What else interesting is there about this?

Termination analyzer is the term that I have been using for many months.

So what? We are still discussing the same DDD, which we want the
correct termination status of, and we want HHH to halt. Nothing changed.

Each of the directly executed HHH emulator/analyzers that returns 0
correctly reports the above *non_terminating _behavior* of its
input. When evaluating the external truth of my stipulated
definition premises and thus the soundness of my reasoning

Aha! Your premises *can* be false.

Vert unlikely because they do conform to software engineering and
termination analysis standard definitions.

Just noting that your past dozen or so posts were useless and wrong.

It seems dishonest of you yo refer to what I said in the past as the
basis of your rebuttal to what I am saying now. At the very best it is
the systematic error of bias.

You could be honest and admit that you were wrong about premises not
being able to be wrong.

one cannot change the subject away from the termination analysis of
C functions to the halt deciders of the theory of computation this
too is the strawman deception.

Not happening. You are the one claiming to have implemented a halting
decider. Your work is related more to the HP than to the termination
analysis of general functions.

At least everyone will know that you are using the strawman deception
in your rebuttal.

What even IS your claim at this point?

When HHH is an x86 emulation based termination analyzer then each DDD
*correctly_emulated_by* any HHH that it calls never returns.
Each of the directly executed HHH emulator/analyzers that returns 0
correctly reports the above *non_terminating _behavior* of its input.

Ah, that is still wrong, because the input DDD halts.

--
Am Sat, 20 Jul 2024 12:35:31 +0000 schrieb WM in sci.math:
It is not guaranteed that n+1 exists for every n.