[Strachey 1965]'s "Impossible Program" and associated Halting
Problem proofs tell us that halt deciders cannot exist as you can
create a program that does the opposite of what any halt decider
thinks that program will do (halt or run forever).

This is a fiction.

The ONLY valid halt decider that doesn't fall foul of the "Impossible
Program" contradiction is a halt decider that can recognise that a
program is attempting to defeat the decider. The ONLY halt decider that
can achieve this is the Flibble Signaling Halt Decider (tm):

https://github.com/i42output/halting-problem#readme

This is groundbreaking research in the field of computer science.

Message ends.

/Flibble

--
N.B. it is unknown whether or not the above post is a shitpost
as it appears to be legit at least at a superficial level and the
author was unavailable for comment.

On Friday, 2 September 2022 at 03:39:49 UTC+2, Mr Flibble wrote:
> [Strachey 1965]'s "Impossible Program" and associated Halting
> Problem proofs tell us that halt deciders cannot exist as you can
> create a program that does the opposite of what any halt decider
> thinks that program will do (halt or run forever).
>
> This is a fiction.
>
> The ONLY valid halt decider that doesn't fall foul of the "Impossible
> Program" contradiction is a halt decider that can recognise that a
> program is attempting to defeat the decider. The ONLY halt decider that
> can achieve this is the Flibble Signaling Halt Decider (tm):
>
> https://github.com/i42output/halting-problem#readme
>
> This is groundbreaking research in the field of computer science.
>
> Message ends.
>
> /Flibble
>
> --
> N.B. it is unknown whether or not the above post is a shitpost
> as it appears to be legit at least at a superficial level and the
> author was unavailable for comment.

Tell us you don't understand (un)decidability without telling us you don't understand (un)decidability.

Instead of trying to implement a decider H which attempts to determine whether a particular program P will; or won't halt you've now tasked yourself with attempting to implement decider D which attempts to decide whether a particular program P is; or isn't trying to defeat it.

Might as well say it out loud: You are trying to decide whether any particular program P should be treated as code; or data.
You are trying to decide whether any particular input P will exploit your system and hijack its control-flow.

Good luck! When you solve it - you'll solve the problem of 0-day exploits in the entire information security industry!

On Fri, 2 Sep 2022 05:54:23 -0700 (PDT)
Skep Dick <skepdick22@gmail.com> wrote:

> On Friday, 2 September 2022 at 03:39:49 UTC+2, Mr Flibble wrote:
> > [Strachey 1965]'s "Impossible Program" and associated Halting
> > Problem proofs tell us that halt deciders cannot exist as you can
> > create a program that does the opposite of what any halt decider
> > thinks that program will do (halt or run forever).
> >
> > This is a fiction.
> >
> > The ONLY valid halt decider that doesn't fall foul of the
> > "Impossible Program" contradiction is a halt decider that can
> > recognise that a program is attempting to defeat the decider. The
> > ONLY halt decider that can achieve this is the Flibble Signaling
> > Halt Decider (tm):
> >
> > https://github.com/i42output/halting-problem#readme
> >
> > This is groundbreaking research in the field of computer science.
> >
> > Message ends.
> >
> > /Flibble
> >
> > --
> > N.B. it is unknown whether or not the above post is a shitpost
> > as it appears to be legit at least at a superficial level and the
> > author was unavailable for comment.
>
> Tell us you don't understand (un)decidability without telling us you
> don't understand (un)decidability.
>
> Instead of trying to implement a decider H which attempts to
> determine whether a particular program P will; or won't halt you've
> now tasked yourself with attempting to implement decider D which
> attempts to decide whether a particular program P is; or isn't trying
> to defeat it.
>
> Might as well say it out loud: You are trying to decide whether any
> particular program P should be treated as code; or data. You are
> trying to decide whether any particular input P will exploit your
> system and hijack its control-flow.
>
> Good luck! When you solve it - you'll solve the problem of 0-day
> exploits in the entire information security industry!

One has nothing to do with the other.

/Flibble