Path: ...!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: olcott <polcott333@gmail.com>
Newsgroups: comp.theory
Subject: Re: Termination analyzer defined ---RICHARD IS WRONG !!!
Date: Sun, 12 May 2024 12:12:00 -0500
Organization: A noiseless patient Spider
Lines: 74
Message-ID: <v1qt92$2rdui$1@dont-email.me>
References: <v1me7i$1l6ut$1@dont-email.me> <v1nec4$1vb8i$1@dont-email.me>
 <v1o6p5$24f4c$2@dont-email.me> <v1pvj0$2knal$1@dont-email.me>
 <v1qi01$2on4q$2@dont-email.me> <v1qn4o$2pts6$1@dont-email.me>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Injection-Date: Sun, 12 May 2024 19:12:02 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="72ecece1a9bec17b694bf422ef8241f0";
	logging-data="2996178"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX18+BBVH8Co0798QpePI9WqT"
User-Agent: Mozilla Thunderbird
Cancel-Lock: sha1:9RVpagalzGlt3fY3bX4KmtLgvVo=
Content-Language: en-US
In-Reply-To: <v1qn4o$2pts6$1@dont-email.me>
Bytes: 4045

On 5/12/2024 10:27 AM, Mikko wrote:
> On 2024-05-12 13:59:28 +0000, olcott said:
> 
>> On 5/12/2024 3:45 AM, Mikko wrote:
>>> On 2024-05-11 16:35:48 +0000, olcott said:
>>>
>>>> On 5/11/2024 4:39 AM, Mikko wrote:
>>>>> On 2024-05-11 00:30:40 +0000, olcott said:
>>>>>
>>>>>> A termination analyzer is different than a halt decider in that it 
>>>>>> need
>>>>>> not correctly determine the halt status of every input. For the 
>>>>>> purposes
>>>>>> of this paper a termination analyzer only needs to correctly 
>>>>>> determine
>>>>>> the halt status of one terminating input and one non-terminating 
>>>>>> input.
>>>>>> The computer science equivalent would be a halt decider with a 
>>>>>> limited
>>>>>> domain that includes at least one halting and one non-halting input.
>>>>>
>>>>> From https://www.google.fi/search?q=termination+analysis and
>>>>> https://en.wikipedia.org/wiki/Termination_analysis :
>>>>>
>>>>> "In computer science, termination analysis is program analysis 
>>>>> which attempts to determine whether the evaluation of a given 
>>>>> program halts for each input. This means to determine whether the 
>>>>> input program computes a total function."
>>>>>
>>>>> So the term "termination analysis" is already defined. The derived 
>>>>> term
>>>>> "termination analyzer" means a performer of termination analysis. That
>>>>> does not agree with the propsed defintion above so a differnt term
>>>>> should be used.
>>>>>
>>>>> That "termination analysis" is a know term that need not be defined
>>>>> is demostrated e.g. by
>>>>>
>>>>>    https://arxiv.org/pdf/2101.09783
>>>>>
>>>>> which simply assumes that readers know (at least approximately) what
>>>>> the term means.
>>>>>
>>>>
>>>> You are doing a great job performing an honest review!
>>>> So every time that Richard referred to a {termination analyzer} that
>>>> ignores its inputs *Richard was WRONG*
>>>
>>> More important is that you are wrong whenever you use "termination
>>> analyser" for something that by the conventional meaning isn't.
>>>
>>
>> A conventional termination analyzer is free to use any algorithm
>> as long as it analyzes termination.
> 
> It is not sufficient to analyse something about termination. The
> conventional meaning is that a termination analyser does not say
> "yes" unless the analysed program terminates with every possible
> input.
> 

A specific program halts with every input is not at all the same
thing as correctly analyzing every program with every input.

Termination analyzers are allowed to be quite dumb so that they
handle very few programs. One of the best termination analyzers
could not handle any recursive programs at all for quite
a few years.


-- 
Copyright 2024 Olcott "Talent hits a target no one else can hit; Genius
hits a target no one else can see." Arthur Schopenhauer