On 2024-09-07 13:06:52 +0000, olcott said:
> On 9/7/2024 3:35 AM, Mikko wrote:
>> On 2024-09-06 12:22:04 +0000, olcott said:
>>
>>> On 9/6/2024 6:55 AM, Mikko wrote:
>>>> On 2024-09-03 12:44:00 +0000, olcott said:
>>>>
>>>>> On 9/3/2024 5:38 AM, Mikko wrote:
>>>>>> On 2024-09-02 13:01:23 +0000, olcott said:
>>>>>>
>>>>>>> On 9/2/2024 2:54 AM, Mikko wrote:
>>>>>>>> On 2024-09-01 13:47:00 +0000, olcott said:
>>>>>>>>
>>>>>>>>> On 9/1/2024 7:52 AM, Mikko wrote:
>>>>>>>>>> On 2024-08-31 18:48:18 +0000, olcott said:
>>>>>>>>>>
>>>>>>>>>>> *This is how I overturn the Tarski Undefinability theorem*
>>>>>>>>>>> An analytic expression of language is any expression of formal or
>>>>>>>>>>> natural language that can be proven true or false entirely on the basis
>>>>>>>>>>> of a connection to its semantic meaning in this same language.
>>>>>>>>>>>
>>>>>>>>>>> This connection must be through a sequence of truth preserving
>>>>>>>>>>> operations from expression x of language L to meaning M in L. A lack of
>>>>>>>>>>> such connection from x or ~x in L is construed as x is not a truth
>>>>>>>>>>> bearer in L.
>>>>>>>>>>>
>>>>>>>>>>> Tarski's Liar Paradox from page 248
>>>>>>>>>>> It would then be possible to reconstruct the antinomy of the liar
>>>>>>>>>>> in the metalanguage, by forming in the language itself a sentence
>>>>>>>>>>> x such that the sentence of the metalanguage which is correlated
>>>>>>>>>>> with x asserts that x is not a true sentence.
>>>>>>>>>>> https://liarparadox.org/Tarski_247_248.pdf
>>>>>>>>>>>
>>>>>>>>>>> Formalized as:
>>>>>>>>>>> x ∉ True if and only if p
>>>>>>>>>>> where the symbol 'p' represents the whole sentence x
>>>>>>>>>>> https://liarparadox.org/Tarski_275_276.pdf
>>>>>>>>>>>
>>>>>>>>>>> *Formalized as Prolog*
>>>>>>>>>>> ?- LP = not(true(LP)).
>>>>>>>>>>> LP = not(true(LP)).
>>>>>>>>>>
>>>>>>>>>> According to Prolog semantics "false" would also be a correct
>>>>>>>>>> response.
>>>>>>>>>>
>>>>>>>>>>> ?- unify_with_occurs_check(LP, not(true(LP))).
>>>>>>>>>>> false.
>>>>>>>>>>
>>>>>>>>>> To the extend Prolog formalizes anything, that only formalizes
>>>>>>>>>> the condept of self-reference. I does not say anything about
>>>>>>>>>> int.
>>>>>>>>>>
>>>>>>>>>>> When formalized as Prolog unify_with_occurs_check()
>>>>>>>>>>> detects a cycle in the directed graph of the evaluation
>>>>>>>>>>> sequence proving the LP is not a truth bearer.
>>>>>>>>>>
>>>>>>>>>> Prolog does not say anything about truth-bearers.
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>> It may seem that way if you have no idea what
>>>>>>>>> (a) a directed is
>>>>>>>>> (b) what cycles in a directed graph are
>>>>>>>>> (c) What an evaluation sequence is
>>>>>>>>
>>>>>>>> More relevanto would be what a "truth-maker" is.
>>>>>>>> Anyway, it seems that Prolog does not say anything about
>>>>>>>> truth-bearers because Prolog does not say anything about
>>>>>>>> truth-bearers.
>>>>>>>>
>>>>>>>
>>>>>>> When Prolog derives expression x from Facts and Rules
>>>>>>> by applying the truth preserving operations of Rules to
>>>>>>> Facts is the truthmaker for truth-bearer x.
>>>>>>
>>>>>> A Prolog impementation applies Prolog operations.
>>>>>
>>>>> Which are (like logic) for the most part truth preserving.
>>>>> If (A & B) then A
>>>>
>>>> Logic where the infoerence rules are for the most part truth prserving
>>>> is not useful. They all must be.
>>>>
>>>>>> For some cases
>>>>>> Prolog offers several operations letting the implementation to
>>>>>> choose which one to apply.
>>>>>
>>>>> I don't thing so. Once the Facts and Rules are specified
>>>>> Prolog chooses whatever Facts and Rules to prove x or not.
>>>>> It is back-chained inference.
>>>>
>>>> Standard Prolog is what the Prolog standard says. Conforming implementations
>>>> may vary if the standard allows. If you think otherwise you are wrong.
>>>> There are also non-starndard Prlongs that vary even more.
>>>>
>>>
>>> The fundamental architectural overview of all Prolog implementations
>>> is the same True(x) means X is derived by applying Rules (AKA truth
>>> preserving operations) to Facts.
>>
>> The details are permitted to differ.
>>
>
> Instead of using any single order of logic we simultaneously
> represent an arbitrary number of orders of logic in a type
> hierarchy knowledge ontology.
The type system of Prolog is different.
--
Mikko
|
|