1. Forum
  2. Usenet
  3. COMP.THEORY

  • Re: Defining problems to make solutions impossible --- DOS detector tha

    From joes@21:1/5 to All on Sun Mar 16 20:25:49 2025
    Am Sun, 16 Mar 2025 14:54:29 -0500 schrieb olcott:
    On 3/16/2025 2:18 PM, dbush wrote:
    On 3/16/2025 2:47 PM, olcott wrote:
    On 3/16/2025 1:41 PM, dbush wrote:
    On 3/16/2025 2:17 PM, olcott wrote:
    On 3/16/2025 1:11 PM, dbush wrote:
    On 3/16/2025 2:05 PM, olcott wrote:
    On 3/16/2025 12:38 PM, dbush wrote:
    On 3/16/2025 1:18 PM, olcott wrote:
    On 3/16/2025 11:54 AM, Richard Heathfield wrote:
    On 16/03/2025 16:46, dbush wrote:

    A solution to the halting problem is an algorithm H

    And therefore, according to Knuth, the solution has the
    following properties:

    Finiteness - An algorithm must start and stop. The rules an >>>>>>>>>> algorithm applies must also conclude in a reasonable amount of >>>>>>>>>> time. What “reasonable” is depends on the nature of the >>>>>>>>>> algorithm, but in no case can an algorithm take an infinite >>>>>>>>>> amount of time to complete its task. Knuth calls this property >>>>>>>>>> the finiteness of an algorithm.

    Definiteness - The actions that an algorithm performs cannot be >>>>>>>>>> open to multiple interpretations; each step must be precise and >>>>>>>>>> unambiguous. Knuth terms this quality definiteness. An
    algorithm cannot iterate a “bunch” of times. The number of >>>>>>>>>> times must be precisely expressed, for example 2, 1000000, or a >>>>>>>>>> randomly chosen number.

    Inputs - An algorithm starts its computation from an initial >>>>>>>>>> state. This state may be expressed as input values given to the >>>>>>>>>> algorithm before it starts.


    The direct execution of DD() IS NOT AN INPUT VALUE TO HHH.
    Yes it is.

    It is when the requirements say so:

    That is not the way that reality works.
    Dogma that disagrees with truth is i ncorrect.
    A static finite string <is not> a dynamic process.
    Whatever. The input to HHH does not include that HHH simulates it,
    HHH will do that anyway.

    But it's stipulated to be a correct description of a dynamic
    process, as per the requirements:

    Yet the semantics of the x86 language conclusively proves that

    When those actual instructions are executed on an actual x86
    processor that they will halt.

    False. It is possible that an actual variation of the x86 processor
    could have HHH encoded in its micro-code then we get the same result.

    In other words, not an actual x86 processor, and therefore irrelevant


    The issue is that PATHOLOGICAL SELF-REFERENCE CANNOT BE CORRECTLY
    IGNORED NO MATTER WHAT YOUR INDOCTRINATION TELLS YOU.
    It can be ignored incorrectly :D
    But what do you even mean by that?

    That doesn't change the fact that I want to know if any arbitrary
    algorithm X with input Y will halt when executed directly.

    That is a stupid and arbitrary requirement
    That is exactly the formulation of the halting problem.

    What we really need to know is this:
    Would input D to denial-of-service-attack detector H cause a denial-of-service-attack to succeed if not aborted?
    DOS != nontermination.

    Does an H exist that can tell me that or not?
    This is now the 6th time I've asked this question.  What are you afraid
    will happen if your answer it?
    --
    Am Sat, 20 Jul 2024 12:35:31 +0000 schrieb WM in sci.math:
    It is not guaranteed that n+1 exists for every n.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to olcott on Sun Mar 16 22:52:14 2025
    On 3/16/25 8:14 PM, olcott wrote:
    On 3/16/2025 3:25 PM, joes wrote:
    Am Sun, 16 Mar 2025 14:54:29 -0500 schrieb olcott:

    What we really need to know is this:
    Would input D to denial-of-service-attack detector H cause a
    denial-of-service-attack to succeed if not aborted?

    DOS != nontermination.


    My termination analyzer does correctly report the
    non-termination of its inputs.


    Only if you consider wrong answer acceptable, because you believe that
    lying is ok.

    Sorry, but you have the problem that as long as you imply you are in the standard system, you are actually stuck with the standard meaning of words.

    Since the direct execution of D will halt, since H(D) returns 0, your H
    is just wrong as a halt decider.

    Now, if you really mean by "termination analyser" a DOS detector for the Termination Analyzer, you might be able to make a point, except that the
    input isn't a DOS for it, since it know about it and aborts it.

    It is only a DOS attack for a less smart DOS detector.

    Sorry, you are just digging a deeper hole for your dead repuation.

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