• =?utf-8?B?UmU6IExpbnogSCBhcHBsaWVkIHRvIOKfqEjin6kg4p+oSOKfqSBpcyByZXF1a

    From Mikko@21:1/5 to olcott on Sun Feb 18 19:22:17 2024
    On 2024-02-18 15:27:38 +0000, olcott said:

    On 2/18/2024 4:46 AM, Mikko wrote:
    On 2024-02-17 14:59:34 +0000, olcott said:

    On 2/17/2024 4:08 AM, Mikko wrote:
    On 2024-02-16 19:54:06 +0000, Ben Bacarisse said:

    Mikko <[email protected]> writes:

    On 2024-02-16 11:43:15 +0000, Ben Bacarisse said:

    Mikko <[email protected]> writes:

    On 2024-02-16 06:07:04 +0000, olcott said:

    // Linz Turing machine H --- M applied to w
    // --- Does M halts on w?
    H.q0 ⟨M⟩ w ⊢* H.qy // M applied to w halts
    H.q0 ⟨M⟩ w ⊢* Hqn // M applied to w does not halt
    // Linz Turing machine H ---  H applied to ⟨H⟩
    // --- Do you halt on your own Turing Machine description ?
    H.q0 ⟨H⟩ ⟨H⟩ ⊢* H.qy // H applied to ⟨H⟩ halts >>>>>>>>> H.q0 ⟨H⟩ ⟨H⟩ ⊢* H.qn // H applied to ⟨H⟩ does not halt >>>>>>>> What does H applied to <H> mean? H requires two argumlents.
    Do you mean that the unspecified second argument is the empty tape? >>>>>>>> Linz always specifies both arguments of H.
    Turing machines don't have "arguments" -- there is just a tape.
    Personally, I would prefer a bit more rigour with an explicit notation >>>>>>> for the encoding of a pair[1], but Linz is outlining this proof only >>>>>>> because is has some historical interest.

    When discussing a Turing machine it may be practical to call the
    initial tape content a sequence of arguments if the problem
    specification specifies the input as a combination of separately
    defined parts.

    Yes, but that won't help here because you are asking what "H applied to >>>>> <H>" might mean.  All TM's only require a string, even though in some >>>>> cases it's convenient to pretend that there is a sequence of
    "arguments".

    What follows the words "applied to" must specify an input to the
    Turing machine identified before those words. Linz does not say
    what H is expected to do if the input is not a pair of descritions,
    a Turing machine and a tape content. Therefore, whoever says
    "H applied to <H>" must tell what that means, in particular,
    whether H may answer something else that "yes" or "not".


    Maybe you should read Linz
    https://www.liarparadox.org/Peter_Linz_HP_317-320.pdf

    Linz uses Wm as the TM description of M

    // Verbatim Linz Turing machine H --- M applied to w
    // --- Does M halt on w?
    H.q0 Wm w ⊢* H.qy // M applied to w halts
    H.q0 Wm w ⊢* H.qn // M applied to w does not halt

    A particular caseof this is

    H.q0 ⟨H⟩ ⟨H⟩ ⊢* H.qy // H applied to ⟨H⟩ halts
    H.q0 ⟨H⟩ ⟨H⟩ ⊢* H.qn // H applied to ⟨H⟩ does not halt

    which you said above. However, Linz does not specify whether
    H applied to <H> should halt or not, as the input is not in
    the domain of H, so nothing can be inferred from this case,
    which therefore is uninteresting.


    "Turing machine H will halt with either a yes or no answer.
    We achieve this by asking that H halt in one of two
    corresponding final states, say, qy or qn." https://liarparadox.org/Peter_Linz_HP(Pages_318-319).pdf

    Your quote is wrong. A partial sentence should not be quoted
    without a good reason and when there q good reason to omit
    a part of a sentence the omission should be clearly indicated
    (usually with trhree dots).

    --
    Mikko

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