open source winis-prolog project: example and question
ozah...@yahoo.com
Dear Mr Dobrev,
If it is occur to be of You interest and enjoyment, let
us consider a standard Prolog computational semantic
where a term, a clause, a strict ordered set of clauses
and a question to this set of clauses are well defined notions.
Let us observe the following example of strict ordered set of clauses:
a(x) ;- b(x,y), c(y,x).
a(1).
b(1,2).
c(2,x) :- a(x)
and the question to it in the form a(1).
How, according to Your papers:
- "An Abstract Approach to Some Loop Detection Problems"
http://www.fmi.uni-sofia.bg/fmi/logic/skordev/aa.htm
- "Periodical Cyclings in Prolog"
http://dobrev.com
one can describe to himself :
- the answer to the question to this strict ordered set of clauses;
- is there a cycling (loop gap) herein;
- is there a need to accordingly look over, situate, the notion
of complexity to the example;
- might You suggest a handling of this "unpleasant situation",
at this obvious for the computational semantics situation, but
the handling must neither change the strict
ordered set of clauses neither change the question to it.
Kind Regards,
Ognyan Zahariev
winis
P.S. My handling of the "unpleasant situation" come from/with the
stated at:
- "The LISP&Prolog Or For The Promised............ Page IX"
- at my blog in the book "missleading of misstaking":
http://360.yahoo.com/ozahariev;
- with some print error in the book here corrected (Note:
copy/paste authors/editors applied techniques
sometimes makes headaches), part of which is:
"For to evaluate the written here, You need knowledge of LISP
and Prolog programing languages and also of the Mats Carlson
specification, I also has read a MIT textbook with close
spec's, Henkin's theorems are as a might. So here also
are some simple assumptions:
- a Prolog clause is a list of the from (head.tail), where head
is a LISP list and tail is one of NULL or a [?nonrecusive] list
having at its leafs instances of native LISP types;
- Prolog clauses is of the from Null or (clause.clauses)
- a question to the Prolog engine is to be considered as an
evaluation of the LISP form (query clause)
Following the established requirements, I will mark the case for
the (query (x ?)) question to Prolog machine at the example Prolog
clauses (((x y) (x z))), so this case will never fulfill to get an
answer, according to the gaps of noncycling at Prolog engine.
The handlings for the gaps of noncycling are precheckings for the
special regardings of repeatable pattern-matching of a clause with
the same head at every case of the clause head pattern-matching,
also pattern-matchable for the tail at the history, but totally or
partially unbound at the variables pattern-matching level, having
to check for the repeatables starting always at the beginning of
pattern-matching history, of initial question at the [?constant]
initial Prolog clauses, therefore at an occurrence of a regardings
the unobtainable answer for initial question to the Prolog engine
is to be set accordingly to a NULL.
My Artificial Intelligence Hypothesis: Prolog machine extended with
natural handlings of the gaps of non-cycling is a sort of
artificial intelligence gain tool [?nonbreakable theorem prover].
One might wonder and say that I was not practical, at the summer
of 1987 or latter or now, by not talking to the public of what a
great gain tool and hypothesis I have found. I beg your pardon,
read what exactly is written at this page and at the book as all."
P.S. A Carbon Copy of the letter is sent to ma...@sicse.se - if I
am lucky this has to be the email of Mats Carlson, the person
mentioned at previous P.S.. I only have such a suggestion -
according to the relations of mentioned specification and the
long history of Quintus Prolog development, nevertheless that
the specification and the read at 1986-87 Carlson's master
thesis consider LISP environment for the Prolog implementation
P.S. A Carbon Copy of the letter is sent to war...@cs.sunysb.edu -
if I am lucky this has to be the email of David Warren author
of
Warren's Prolog Abstract Machine very well described at: "New
Generation Computing 2 (1984) OHMSHA LTD And Springer-
Verlag". Where also one can find out a lot of sensable and
valuable things about Prolog and etc., actually this book drive
me
to chose the technique and the tactic to implement Prolog,
handle Prolog cycling (loop gap) only and only as a pure release
of a simple but quite long logical theorem (or if You prefer set
of
mathematical lemas). I simply decline to consider "other types
of logic programing" as far as, under VMS and INEX project
circumstances my implementation also become to be a virtual
Prolog Machine - very very very unbelievable indeed.
P.S. A Carbon Copy of the letter is sent to other members of
Department of Mathematical Logic and Applications at Sofia
University "St. Kliment Ohridski"
P.S. A Carbon Copy of the letter is sent to icl...@dcc.fc.up.pt wich
has to be the program committee of the 23rd International
Conference on Logic Programming
P.S. A copy of this letter is published at sci.math , sci.logic,
comp.lang.prolog, prolog.general, alt.philosophy,
comp.ai.philosophy, comp.ai
P.S. A copy of this letter is published at sourceforge discutions
forum, regarding the project:
https://sourceforge.net/projects/winis-prolog/
P.S. A Carbon Copy of the letter is also sent to Daniel Zahariev, for
whom one can read at:
http://www.fmi.uni-sofia.bg/microsoft/contacts.html, or
http://daniel-zahariev.hit.bg/myCV.html
For to opinion your opinion at my [mis]understandings:
MSN,Y! : ozahariev ; ICQ# : 267774623
Mobile: +359899483922 ; or
Blog: http://360.yahoo.com/ozahariev
Projects: https://sourceforge.net/projects/winis-prolog/
A.L.
>Dear prof.Skordev,
>
>Dear Mr Dobrev,
e-mail was invented to send private letters to private people.
A.L.
ozah...@yahoo.com
As far as public forum stands for public opinion I am quite not agree
that the open letter written herein is private.
By the way and for example: garbage collectors, not only in LISP, are
for garbadge collecting dear A.L. and if A.L. has experiance with this
please ask him to email me: ozah...@yahoo.com.
Ognyan Zahariev
A.L.
<ozah...@yahoo.com> wrote:
>A.L. ??????:
I believe that garbage collector is needed to collect garbage you post
to this group
A.L.
ozah...@yahoo.com
> On 1 Apr 2007 04:12:40 -0700, "ozahar...@yahoo.com"
>
>
>
>
>
> <ozahar...@yahoo.com> wrote:
> >A.L. ??????:
> >> On 30 Mar 2007 22:43:32 -0700, ozahar...@yahoo.com wrote:
>
> >> >Dear prof.Skordev,
>
> >> >Dear Mr Dobrev,
>
> >> e-mail was invented to send private letters to private people.
>
> >> A.L.
>
> >As far as public forum stands for public opinion I am quite not agree
> >that the open letter written herein is private.
>
> >By the way and for example: garbage collectors, not only in LISP, are
> >for garbadge collecting dear A.L. and if A.L. has experiance with this
>
> I believe that garbage collector is needed to collect garbage you post
> to this group
>
>
> - Show quoted text -
perhaps if You are the moderator of the forum what are You waiting for
- do it.
I will go to the proper uncensured poke place.
Or if You are interested might be will explain to the public and me
this citate from an answer
letter from prof.D.Skordev:
"Of course, a(1) is true in all models of the program, but the
incompleteness of the depth-first
strategy prevents ordinary Prolog compilers from answering by
"Yes"."
Regards
Ognyan Zahariev
winis
P.S. For the intersted of this example and open question:
1. Prof.Skordev is quite right that clause a(x):-b(x,y),c(y,x) is only
misprinted. The prolog interpreter/compiler, by default, has to signal
this error in the syntax.
2. After some observation one might agree that the example is tricky,
since "a(1) is true in all models of the program".
3. I have to comment the open example, as I was all ready done, this
but in Bulgarian, and this comment is not of mathematical nature, but,
on my risk, of notion in concern with practical programming
principles:
a) from the point of view named "normal programming" one might expect
that so given Prolog knowledge base will "throw expected answer" as
an answer to the question a(1) or a(?) overcoming cycling - some how;
(I do not suggest throw build-in predicate)
b) from the point of view named "observed prolog program logic" it
will be pretty nice feature of the prolog machine "to see and inform
about cycling just in time";
c) from the point of view named "programming causes" this simple
example is not a case since, for example under UNIX OS wrongly written
programs doesn't "inform about that" and also might crush the system,
therefore we have at programming what is named best current practice
of programming style - this prolog program is a contra-example for the
named mentioned with concern to prolog programming;
d) :) from the point of view named "science fiction" the possibility
that on Mars the problem is unhandled is close to zero, and perhaps
this is Mars we know that either the martial law is enforced or new
taxes imposition is on, both because of some current programming in
prolog about ...., but nevertheless and immediately some where at one
of thousand of programmers galaxies a crew with rebels will be formed.