1. Forum
  2. Usenet
  3. COMP.LANG.C++

  • Re: How do you prove Halting Problem in C/C++?

    From Wuns Haerst@21:1/5 to All on Mon Aug 25 17:11:08 2025
    Am 25.08.2025 um 17:07 schrieb wij:
    On Mon, 2025-08-25 at 16:47 +0200, Bonita Montero wrote:
    Am 25.08.2025 um 16:27 schrieb wij:
    This topic relates to how much you know about programs and prove the C/C++ >>> codes is (logically) correct.


    The halting problem is simple to solve:

    jthread( []
    {
    this_thread::sleep_for( 1h );
    abort();
    } ).detach();

    Each program that uses this halts at least in an hour.
    Aren't I a genius ? I'll get the Turing Award for that !

    Halting Problem is about Turing Machine, But you can use, for example,
    a 'function', or an executable,...,others.

    You just prove you are another olcott who deos not know what HP is.
    But in your case, how do you prove 'jthread' is correct?

    I've proven that any program can be modified in a way that it halts.
    This is so genius that I think I'll get not the Turing Award but a
    Nobel Price for that. I'll change the world.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Gerald Tubin@21:1/5 to wij on Mon Aug 25 14:57:15 2025
    On 25/08/2025 15:27, wij wrote:
    This topic relates to how much you know about programs and prove the C/C++ codes is (logically) correct.


    In order to demonstrate your understanding of the halting problem,
    please summarise it for us. If you are unable to do so, please take the discussion elsewhere. So far, this discussion has had nothing to do with
    C/C++.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Gerald Tubin@21:1/5 to wij on Mon Aug 25 15:51:59 2025
    On 25/08/2025 16:10, wij wrote:
    I guess you have no idea!

    We rely on you to explain things to us in a simple way. As Albert
    Einstein once said, 'If you can't explain it simply, you don't
    understand it well enough'.

    Given your extensive knowledge of C/C++ and halting problems, please
    summarise it for us here.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Gerald Tubin@21:1/5 to wij on Mon Aug 25 16:47:49 2025
    On 25/08/2025 17:16, wij wrote:
    On Mon, 2025-08-25 at 15:51 +0000, Gerald Tubin wrote:
    On 25/08/2025 16:10, wij wrote:
    I guess you have no idea!

    We rely on you to explain things to us in a simple way. As Albert
    Einstein once said, 'If you can't explain it simply, you don't
    understand it well enough'.

    Given your extensive knowledge of C/C++ and halting problems, please
    summarise it for us here.

    The idea was just inspired by olcott. Halting Problem should be simple and does not involve too many stuff. Detail of the rule I can think about now is:
    1. The correctness of C/C++ codes is based Turing Machine.
    (otherwise, no way to verify. Or, you porved Church-Turing thesis is wrong)
    2. C++std library does not prove itself correct, the proof generally cannot
    relies on it (except you can prove it. Some primative things should be OK).
    3. xxx are correct iff it can be implemented by Turing Machine. But normally we
    don't need to go that deep. In some cases, specification is fine if it can
    be correctly assumed correct.


    So it's all humbug! Can we please move on to discussing something that
    can be proven.

    Who is Olcott? I can't see any of his posts here. Have I filtered him
    out? Probably. I don't want to read anything he writes, so please don't
    ever quote him here. I might have filtered him for a reason.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)