• Re: Defining a halt decider with perfect accuracy

    From olcott@polcott333@gmail.com to comp.theory,sci.logic,comp.ai.philosophy on Sat Dec 13 16:41:47 2025
    From Newsgroup: comp.ai.philosophy

    On 12/13/2025 3:32 PM, olcott wrote:
    All of the textbooks require halt deciders to
    report on the behavior of machine M on input w.
    This may be easy to understand yet not precisely
    accurate.

    Since no Turing machine ever takes any Machine
    M as an input this <is> a category error even
    when this makes no functional difference.

    They simply glossed over this key detail because
    they thought that it made no difference.

    *Defining a halt decider with perfect accuracy*
    Turing machine halt deciders compute the mapping
    from input finite strings to an {accept, reject}
    value on the basis of the behavior that this
    input finite string specifies.


    By simply adding more detail we can make the
    original definition more precise:

    A Turing machine based halt decider reports on the
    behavior of machine M on input w thorough the
    proxy of the finite string machine description of
    ⟨M⟩ on input w.

    The above seems to be more precisely accurate
    than any published proof. It includes a key
    detail that all of them seem to leave out.

    If you know of any published proof that directly
    refers to the idea of a proxy, please let me know.
    --
    Copyright 2025 Olcott<br><br>

    My 28 year goal has been to make <br>
    "true on the basis of meaning expressed in language"<br>
    reliably computable.<br><br>

    This required establishing a new foundation<br>
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Richard Damon@Richard@Damon-Family.org to comp.theory,sci.logic,comp.ai.philosophy on Sat Dec 13 18:02:09 2025
    From Newsgroup: comp.ai.philosophy

    On 12/13/25 5:41 PM, olcott wrote:
    On 12/13/2025 3:32 PM, olcott wrote:
    All of the textbooks require halt deciders to
    report on the behavior of machine M on input w.
    This may be easy to understand yet not precisely
    accurate.

    Since no Turing machine ever takes any Machine
    M as an input this <is> a category error even
    when this makes no functional difference.

    They simply glossed over this key detail because
    they thought that it made no difference.

    *Defining a halt decider with perfect accuracy*
    Turing machine halt deciders compute the mapping
    from input finite strings to an {accept, reject}
    value on the basis of the behavior that this
    input finite string specifies.


    By simply adding more detail we can make the
    original definition more precise:

    A Turing machine based halt decider reports on the
    behavior of machine M on input w thorough the
    proxy of the finite string machine description of
    ⟨M⟩ on input w.

    The above seems to be more precisely accurate
    than any published proof. It includes a key
    detail that all of them seem to leave out.

    If you know of any published proof that directly
    refers to the idea of a proxy, please let me know.


    And the use of a string proxy is just normally assumed by the theory, as
    that is how Turing Machine work.

    They almost ALWAYS work by a string representation proxy, as very few
    real questions are based on the "arbitrary" symbol set of the Turing
    Machines native operation.

    If you had bothered to learn the basics of the field, you would have understood that.

    Most works assume the basic knowledge of the field.

    Note, even the Linz proof you mention explicitly talks about giving the decider a representation of the machine in question, the Wm as the proxy
    for giving it M.

    So, why did you not understand the use of a proxy.

    Sometimes the problem when expressed for lay people will talk about the decider being given a description or representation of the machine.

    You just reject those as you think it too vague, when it is a well
    defined term, and even the general meaning is applicable, you just need
    to remember that it must be a SUFFICIENT description to convey the
    needed details of the machine.
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From olcott@polcott333@gmail.com to comp.theory,sci.logic,comp.ai.philosophy on Sat Dec 13 20:30:05 2025
    From Newsgroup: comp.ai.philosophy

    On 12/13/2025 5:02 PM, Richard Damon wrote:
    On 12/13/25 5:41 PM, olcott wrote:
    On 12/13/2025 3:32 PM, olcott wrote:
    All of the textbooks require halt deciders to
    report on the behavior of machine M on input w.
    This may be easy to understand yet not precisely
    accurate.

    Since no Turing machine ever takes any Machine
    M as an input this <is> a category error even
    when this makes no functional difference.

    They simply glossed over this key detail because
    they thought that it made no difference.

    *Defining a halt decider with perfect accuracy*
    Turing machine halt deciders compute the mapping
    from input finite strings to an {accept, reject}
    value on the basis of the behavior that this
    input finite string specifies.


    By simply adding more detail we can make the
    original definition more precise:

    A Turing machine based halt decider reports on the
    behavior of machine M on input w thorough the
    proxy of the finite string machine description of
    ⟨M⟩ on input w.

    The above seems to be more precisely accurate
    than any published proof. It includes a key
    detail that all of them seem to leave out.

    If you know of any published proof that directly
    refers to the idea of a proxy, please let me know.


    And the use of a string proxy is just normally assumed by the theory, as that is how Turing Machine work.


    See that three agreements in one day.
    That may be more than we have ever had.

    Because none of the textbooks ever directly said
    that the finite string input is only a proxy for
    the behavior everyone always took the proxy to be
    exactly one-and-the-same thing as the actual behavior.

    They almost ALWAYS work by a string representation proxy, as very few
    real questions are based on the "arbitrary" symbol set of the Turing Machines native operation.

    If you had bothered to learn the basics of the field, you would have understood that.

    Most works assume the basic knowledge of the field.

    Note, even the Linz proof you mention explicitly talks about giving the decider a representation of the machine in question, the Wm as the proxy
    for giving it M.

    So, why did you not understand the use of a proxy.

    Sometimes the problem when expressed for lay people will talk about the decider being given a description or representation of the machine.

    You just reject those as you think it too vague, when it is a well
    defined term, and even the general meaning is applicable, you just need
    to remember that it must be a SUFFICIENT description to convey the
    needed details of the machine.
    --
    Copyright 2025 Olcott<br><br>

    My 28 year goal has been to make <br>
    "true on the basis of meaning expressed in language"<br>
    reliably computable.<br><br>

    This required establishing a new foundation<br>
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Richard Damon@Richard@Damon-Family.org to comp.theory,sci.logic,comp.ai.philosophy on Sat Dec 13 21:56:58 2025
    From Newsgroup: comp.ai.philosophy

    On 12/13/25 9:30 PM, olcott wrote:
    On 12/13/2025 5:02 PM, Richard Damon wrote:
    On 12/13/25 5:41 PM, olcott wrote:
    On 12/13/2025 3:32 PM, olcott wrote:
    All of the textbooks require halt deciders to
    report on the behavior of machine M on input w.
    This may be easy to understand yet not precisely
    accurate.

    Since no Turing machine ever takes any Machine
    M as an input this <is> a category error even
    when this makes no functional difference.

    They simply glossed over this key detail because
    they thought that it made no difference.

    *Defining a halt decider with perfect accuracy*
    Turing machine halt deciders compute the mapping
    from input finite strings to an {accept, reject}
    value on the basis of the behavior that this
    input finite string specifies.


    By simply adding more detail we can make the
    original definition more precise:

    A Turing machine based halt decider reports on the
    behavior of machine M on input w thorough the
    proxy of the finite string machine description of
    ⟨M⟩ on input w.

    The above seems to be more precisely accurate
    than any published proof. It includes a key
    detail that all of them seem to leave out.

    If you know of any published proof that directly
    refers to the idea of a proxy, please let me know.


    And the use of a string proxy is just normally assumed by the theory,
    as that is how Turing Machine work.


    See that three agreements in one day.
    That may be more than we have ever had.

    Because none of the textbooks ever directly said
    that the finite string input is only a proxy for
    the behavior everyone always took the proxy to be
    exactly one-and-the-same thing as the actual behavior.

    But the behavior represented by the string *IS* exactly the behavior of
    the string, so you attempted point just falls flat.

    And, as I said, even your Linz book made that clear, as H took as it
    input Wm (the string) not M (the machine).

    Also, if you did any real study, you would have learned that the input
    to the machine is almost always just a "represemtation" of the input to
    the function, as we rarely are really interested in computing a result
    on the strings.

    The one exception is the very earliest exercises where you learn basic
    string manipulation with Turing Machines, but you rapidly get to wanting
    to do things like "arithmatic" and then learning you need to REPRESENT
    numbers as something. (and a common method which baffled you as I
    remember was unary, you wanted your Turing Machine to use UNICODE as it
    symbol set.


    They almost ALWAYS work by a string representation proxy, as very few
    real questions are based on the "arbitrary" symbol set of the Turing
    Machines native operation.

    If you had bothered to learn the basics of the field, you would have
    understood that.

    Most works assume the basic knowledge of the field.

    Note, even the Linz proof you mention explicitly talks about giving
    the decider a representation of the machine in question, the Wm as the
    proxy for giving it M.

    So, why did you not understand the use of a proxy.

    Sometimes the problem when expressed for lay people will talk about
    the decider being given a description or representation of the machine.

    You just reject those as you think it too vague, when it is a well
    defined term, and even the general meaning is applicable, you just
    need to remember that it must be a SUFFICIENT description to convey
    the needed details of the machine.



    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From olcott@polcott333@gmail.com to comp.theory,sci.logic,comp.ai.philosophy on Sat Dec 13 21:39:48 2025
    From Newsgroup: comp.ai.philosophy

    On 12/13/2025 8:56 PM, Richard Damon wrote:
    On 12/13/25 9:30 PM, olcott wrote:
    On 12/13/2025 5:02 PM, Richard Damon wrote:
    On 12/13/25 5:41 PM, olcott wrote:
    On 12/13/2025 3:32 PM, olcott wrote:
    All of the textbooks require halt deciders to
    report on the behavior of machine M on input w.
    This may be easy to understand yet not precisely
    accurate.

    Since no Turing machine ever takes any Machine
    M as an input this <is> a category error even
    when this makes no functional difference.

    They simply glossed over this key detail because
    they thought that it made no difference.

    *Defining a halt decider with perfect accuracy*
    Turing machine halt deciders compute the mapping
    from input finite strings to an {accept, reject}
    value on the basis of the behavior that this
    input finite string specifies.


    By simply adding more detail we can make the
    original definition more precise:

    A Turing machine based halt decider reports on the
    behavior of machine M on input w thorough the
    proxy of the finite string machine description of
    ⟨M⟩ on input w.

    The above seems to be more precisely accurate
    than any published proof. It includes a key
    detail that all of them seem to leave out.

    If you know of any published proof that directly
    refers to the idea of a proxy, please let me know.


    And the use of a string proxy is just normally assumed by the theory,
    as that is how Turing Machine work.


    See that three agreements in one day.
    That may be more than we have ever had.

    Because none of the textbooks ever directly said
    that the finite string input is only a proxy for
    the behavior everyone always took the proxy to be
    exactly one-and-the-same thing as the actual behavior.

    But the behavior represented by the string *IS* exactly the behavior of
    the string, so you attempted point just falls flat.


    Do you really think that I will keep going
    on this for 22 years if it just falls flat?

    Google Groups has a much better search so
    you can see the 40,000 messages that I posted
    in comp.theory since 2004.

    My very first Halting Problem post Jun 6, 2004, 9:11:19 AM
    Alan Turing's Halting Problem is incorrectly formed
    It has lots and lots of replies. https://groups.google.com/g/sci.logic/c/V7wzVvx8IMw/m/ggPE6a-60cUJ

    And, as I said, even your Linz book made that clear, as H took as it
    input Wm (the string) not M (the machine).


    That is not the issue. All the textbooks say that.

    The issue is that this finite string AS AN INPUT
    is the ultimate basis of the halt decision even
    when it is not a good proxy for the behavior of
    the executed machine.

    Also, if you did any real study, you would have learned that the input
    to the machine is almost always just a "represemtation" of the input to
    the function, as we rarely are really interested in computing a result
    on the strings.

    The one exception is the very earliest exercises where you learn basic string manipulation with Turing Machines, but you rapidly get to wanting
    to do things like "arithmatic" and then learning you need to REPRESENT numbers as something. (and a common method which baffled you as I
    remember was unary, you wanted your Turing Machine to use UNICODE as it symbol set.


    They almost ALWAYS work by a string representation proxy, as very few
    real questions are based on the "arbitrary" symbol set of the Turing
    Machines native operation.

    If you had bothered to learn the basics of the field, you would have
    understood that.

    Most works assume the basic knowledge of the field.

    Note, even the Linz proof you mention explicitly talks about giving
    the decider a representation of the machine in question, the Wm as
    the proxy for giving it M.

    So, why did you not understand the use of a proxy.

    Sometimes the problem when expressed for lay people will talk about
    the decider being given a description or representation of the machine.

    You just reject those as you think it too vague, when it is a well
    defined term, and even the general meaning is applicable, you just
    need to remember that it must be a SUFFICIENT description to convey
    the needed details of the machine.



    --
    Copyright 2025 Olcott<br><br>

    My 28 year goal has been to make <br>
    "true on the basis of meaning expressed in language"<br>
    reliably computable.<br><br>

    This required establishing a new foundation<br>
    --- Synchronet 3.21a-Linux NewsLink 1.2