Chess One wrote:
There are 2 routes to explore in computerizing chees - one is the current
paradigm which is to make emulation of chess as strong as possible, even if
that means the program contravenes the laws of chess. The other is to
understand something about the evaluation paradigm, a completely untended
sector, which nevertheless is the primum mobile of AI. [which is too
complicated for anyone here, since writers cannot even differentiate
emulation from real]

Since programmers have abandoned this second factor in favor of playing
moves the computer could not achieve, it locks itself into the emulation
paradim, and has nothing whatever to do with learning, which is why
computerization of chess is an abandoned subject in AI.

For computer geeks it is all about strength [commercially to make money],
even if that requires cheating the rules of chess so that it is not chess,
and nothing to do with learning. Review your own writing and tell me in
which camp you are in.

At the end, the initial question remains that the computer plays moves it
does not understand. And maybe you will then consider why this is less than
useful in full expert systems and in AI, since, eg, where does it obtain the
base postions it posits, which it cannot even understand?

That is a systemic approach, or in other words, a scientific one. The
process is all.



From a philosophical point of view, the discussion about computers
"understanding" what they do started with the Turing test and later
Searle's chinese room argument, and the so-called difference between the
syntaxic and semantic level of consciousness.

This is not much discussed today, even at the academic level. The
general view is much more pragmatic, and the general idea is that
"understanding" is a rather difficult thing to define; after all, as
many people have pointed out, the brain itself is just a lot of neurons
with chemical interactions and, from a physical point of view, there is
nothing that can define "understanding"; it is a subjective thinking,
linked only to your own consciousness: you think you are understanding
something, but it doesn't mean much.

"Learning" is also a very difficult thing to define. I wrote a reversi
program a few years ago that reached a quite high ranking on IOS, and it
was a program which fully learned its evaluation function (genetic
algorithm + pattern recognition). Was my program "understanding" Reversi
? Is learning reversi patterns the same thing as learning to speak?

The "expert system" paradigm has been slowly dying for the last 15
years, since the end of the FGCS japanese initiative (1992, Tokyo, 15
years already and I was there...).
Many computer science AI labs have even abandoned the name, and have
become "neural computing lab" or "evolutionary computing lab" instead...

[BTW, I do know a little bit about AI and its philosophical
implications; I even wrote a book about that, and I have been teaching
AI for quite a few years now.. So I hope it is not too complicated for
me at least?]

21.08.2007 17:05, help bot:
On Aug 20, 9:21 pm, Ralf Callenberg wrote:

Those who compete for the championship are likely
among the strongest, not necessarily a representative
sample of the "geeks".
Why?

It's just as with the human world championships: do you
think the two guys playing for the world title are average,
representatives of all chess players?

No. But we have two different characteristics: being good in creating
efficient programs and being good in chess. That the one is correlated
to the other is something I don't see immediately. Therefore I ask: why
should the programmers of strong programs be significantly stronger in
chess than the programmer of a weaker program?

Greetings,
Ralf

On Aug 21, 11:04 am, Guy Macon http://www.guymacon.com/ wrote:

"Arguing with anonymous strangers on the Internet is a sucker's
game because they almost always turn out to be -- or to be
indistinguishable from -- self-righteous sixteen-year-olds
possessing infinite amounts of free time."
-Neil Stephenson

Does this sound to others the way it does to me? Like
a pathetic loser making excuses for losing arguments?
The giveaway is the knee-jerk ad hominem approach,
one which is invariably resorted to after a long series of
ignominious defeats. I especially enjoyed the part where
the loser whines about not having as much free time as
his many child vanquishers! LOL

I think it goes without saying that any *self-respecting*
adult would never admit to such a defeat at the hands
of a 16-year-old punk kid; but this clown appears to be
an adult, so that means... he needs to go and find some,
somewhere. (Self-respect, I mean.)


Please tell me exactly where this alleged not-so-clever
replacement differs from what I posted and quoted.

It differs in that you at first complained of IM Innes'
postings being boring, but in a later post stated that
your original claim was that *he* was boring.

My comment simply pointed out your inconsistency,
while adding that it was ludicrous to "judge" his post
as boring while admitting you had not even read it.

You appear to be in a state of denial, where this
error refuses to register. My advice is to get over
yourself, and face up to your (apparent) strong dislike
of IM Innes, which has you moaning and groaning
about postings being boring when you never even got
the chance to be bored by them, since you failed to
read them, or at least so you say.

Please allow those of us who were and are stupid
enough to *actually read* IM Innes' postings to take
the credit for having suffered all the boredom; thank
you very much. Fair is fair.



One doesn't cover things up by quoting them four times in a row.

That depends on *where* and how you place the
quotations, doesn't it? Nitwit.



In the original post I also wrote "You bore me" just to make
sure that my meaning was clear.

No, you didn't. In your original post you wrote exactly
what I have quoted, again and again: that IM Innes'
*postings* were what you found boring.

These attempts to cover up your little blunder are just
making you look stupid. I already carefully explained
the difference between a man being boring and his
postings being boring, so hopefully we can move along
now.


You appear to believe that personal attacks and namecalling
are an acceptable substitute for rational arguments.

Really? To me, you appear to be a complete and
utter imbecile, but that is before taking into account
your emotional problem which seems to be clouding
your mind. Once you account for that, it becomes
clear that you simply have a powerful dislike of IM
Innes, and this keeps you in a dense mental fog
here.


------------

The real problem with IM Innes' postings on the
issue of "computer cheating", if you will, is that he
long ago grew tired of repeating himself, and so now
he often seems barely comprehensible in that he
just assumes everyone knows what he means; he's
gotten lazy, and just repeats, like a stuck recording.

But is he really being "boring"? To the contrary,
he seems to be taking a position directly opposite
to that of most other posters, who simply accept
the status quo as inevitable; in a sense, it it those
posters who come off as boring, mindless sheep,
who all follow one another down the worn-down path,
simply because it is the path of least resistance;
the path where all the other sheep go.


-- help bot

On Aug 21, 11:32 am, Guy Macon http://www.guymacon.com/ wrote:

[3] You keep repeating your assertions, which is boring.

This reminds me of a fellow who insists that he never
wrote something, but rather that he wrote something
else altogether; he, too, just repeats his assertion,
which as you observe, is very boring.


If you had written something that you haven't written again and
again, it is possible that you could have generated a non-boring
post. Alas, you did not. You did, however, manage to avoid boring
personal attacks this time, so there exists slight variations in
your boringness. Or was your calling me a chess patzer and a
geek an attempt at an insult? They are accurate descriptions.

Look, you may indeed be a patzer, but not a first
class one; patzers are a dime a dozen these days,
but if you want to see what a REAL patzer looks
like, go to chessworld.net and play a man called
Rob Mitchell! Let me see... in one game, I moved
my Queen out threatening several things, and he
left a full piece en prise. Okay, everybody makes
mistakes. But in another game, I did the same
thing, moving my Queen out to Queen's-Rook-four,
and he did it again! Then, as if to prove the point,
he did it yet again in another game. (I hope this
doesn't mess up my thinking to where I start moving
my Queen there even when that square is attacked
by a pawn or something.)

Just remember: there are patzers, and then there
are PATZERS.


The rest of your overly-long post (*must* you quote everything
including the .sig?) was boring for the same reason; endless
repeats of things that you have written again and again.


Worse things exist; have you ever seen postings
by a fellow called Parser Blair? He quoted everyone
but Hellen Keller, as I seem to recall.



BTW, the following assertion;

AI is differentiated from emulation by the Turing test.

...is particularly boring because it is even more wrong
than usual, and is something that you have asserted again
and again with a more-than-usual frequency.

Okay, but this disagreement would have been less
boring had you inserted a "refutation" here.


(Big Yawn) I don't know whether I can stay awake through
your next boring post.

People who seem to always be bored may in fact be
suffering from perpetual "bad company", as a result of
being around themselves practically all the time -- even
when they are asleep. Food for thought.


-- help bot

21.08.2007 15:44, Chess One:

This is too boring for the chess computing community, who simply specualte
on what would happen, and call that science.

Who calls it science?

No reference material of games are permitted.

Yes, and it is as well not permitted to keep any kind of notes, for
instance in order to keep track of the variations a player has
evaluated. Hence, hashtables are as illegal as opening books. That's my
point in this "legal" issue: a chess program in itself is against the
rules, that it uses opening books is a minor thing. If you'd switch them
off it would still be against the rules.


Are you asking me to say something unpopular to the chess computing
community, that you cannot answer for yourself? If I say it is or is not
illegal, is that because you need me to tell you?

No, you were complaining that nobody but you understands what is "real"
and what is "emulated". Therefore I asked you to explain it.


You are not addressing the rules of chess. You are excusing the activity of
a chess player [the computer].

You didn't understand my point. I said repeatedly (a few lines above for
intance) the opposite: the program itself is against the rules from
square one, whether you use opening books or not.


An emulation is different than real AI, since it only presents the
appearance of learning. [see Turing test]

Modern programs do not learn, whether they use opening books or not. I
don't understand why you are so fixed on those stupid opening books.
Todays programs fundamentally defy any possibility to learn. Again:
opening books are just a side issue of nearly no importance compared to
the fundamental way modern chess programs work.


If "real" means similar to how humans play, chess computers don't really
play chess anyhow, whether they use opening books or not. But what is the
purpose of making the distinction between "real" and "emulated"?

Because it is not real! It is not even the computer. It is the programmer.
The computing engine offers the appearance, but cannot learn, by design of
the programmer who controls everything - especially the crucial heart of the
engine which is the evaluation matrix. This is entirely unlike how human
beings behave - and though terms we use for both the program and human
behavior are made deliberately similar, they are not the same.

And? If a computer program stands the Turing tests, nobody denies its
intelligence based on how it works internally. If a program achieves to
convince intelligent people that it is intelligent - it is intelligent.
This looking inside how programs work does not say anything about
whether it is "real" or not.


Does it make sense? In the much more ambitious goal to create intelligent
programs, the AI relies on the Turing test.

AI is differentiated from emulation by the Turing test.

Not every topic of AI is headed at standing the Turing test. AI is the
field where very complex tasks, usually only humans are able to perform,
shall be addressed by computers. Face recognition for instance belongs
to the field of AI, but nobody ever expects that such programs will
stand the Turing test.

The real problem was, that the more ambitious approaches turned out to be
inferior to the "dumb" approach.


[,,,] The 'problem' as you put it, is no problem at all!
People deliberately chose emulation over AI, and strength of the engine to
perform over learning.

Many guys in the beginning tried using learning etc. But those programs
didn't improve considerably over the time in the playing strength, while
the non-learning programs did become very strong.

We forget that the chess engine does nothing that its programmers
didn't intend it to do [even if they can't understand the chess it plays].

Who forgets? Did anybody deny that?

If you made a learning engine to play chess which evolved its own evaluation
matrix in real time, and it was rated initially 800, you would win the Nobel
Prize.

Yes, but how is this correlated to those damn opening books? Again, you
are addressing here the fundamental principle of how chess programs
work. But if they are replacing the opening book just by another
evaluation matrix better suited for the opening phase, nothing would
fundamentally change. That was my point, when saying that replacing the
opening books in itself is not the interesting question when it comes
to more sophisticated chess programs.


I think your assertion lacks two things; any evidence whatever, but also an
appreciation of what AI could be.

You obviously simply didn't understand what my assertion meant.

As explained above, I don't see, why using opening books is more
cheating
than the way computers find their moves in the middlegame.

I know you don't see it, because you do not talk about what the rules say.

You are so convinced that you see things so much clearer than I do, that
you didn't even bother to try to understand what I was saying. Using an
opening book is cheating as well as just relying on the standard
algorithm. *Therefore* it doesn't make any difference. It's cheating anyway.

And I have to tell you,

No, you don't have to tell me, thank you very much.


Beside what point? That there is nothing learned about the worth of opening
books, or that the rules are avoided?

It is besides the point, because it's just a minor flaw compared to the
fundamental flaws of current computer programs when judged by your
principles: they are cheating, they are not learning, it is no "real" chess.


If you don't want to know anything

Did I say something like that? Show me the quote, where I said something
like that.

- then I suppose you will continue to say
that knowledge is beside the point - for you.

Please, spare me this nonsense.


It is a much more fundamental question about current chess programs.
Indeed the principal approach of current chess programming stepped away
from the goal to somehow simulate the human thought process, to create
systems using knowledge and possibly being able to learn.

While that is the hoped for goal, actual develop of computer engines has
done the complete opposite in reducing learning and increasing copy-cat
emulation.

Well, that doesn't contradict the slightest what I am saying all the time.


But it is possible to look at this stress on strength not only on the
commercial aspect. Free programs without commercial directions are also
aimed at being as strong as possible. Computer chess has become its own
competion, like creating faster cars.

That's right! Its like entering a Nascar in the 400 metres to learn about
how people run, no? What a farce!

That's not what I meant. Computer programs are competing against each
other now, more than they compete against human players. In the 80s you
saw very often Mephisto engines taking part in open tournaments for
instance. This kind of competition has completely vanished. There are a
few exhibition matches between GMs and computers here and there. Compare
this with dozens of rating lists for engines, of ongoing tournaments
between engines, not the least an official World Championship. There
most of todays computer programs compete - not with humans. That's over.

AI abandoned chess since chess abandoned learning.

But you can not blame the programmers of Fritz & Co for that. If the
researchers at the universities and research institutes have stopped
trying to write learning chess programs (did they do this, by the way?)
- what has this to do with the commercial programs or those hobby
programs trying to become as strong as possible?


If it plays a move that it cannot evaluate, or algoritmically evaluates
negatively, but is out of the book, then exactly! it literally does not
understand what it is doing. This relegates emulation engines to
simulators - since no one at all can think of real world applications they
are going to trust with anything important.

You can regard the opening book as part of the evaluation matrix. If
this position appears: it gets this or that value. That it evaluates a
doubled pawn less than two connected pawns is also brought to the
program from the outside. So, does a program "understand", that a
doubled pawn is weaker?


I am interested scientifically and philosophically interested in /why/ the
engine cannot cope with the early complexity.

The techniques used for pruning and following forced lines to increase
the depth of search, does not work in the opening so well, when there is
only very limited interaction between the pieces. This is similar to
the reason, that chess programs are also not so good in endgames as
strong GMs.


Greetings,
Ralf

"help bot" wrote in message
ps.com...
On Aug 20, 8:37 pm, Guy Macon http://www.guymacon.com/ wrote:


I sometimes wonder whether he actually engages his brain
while posting. Perhaps he is on autopilot when generating
boring and repetitive flames.

Perhaps IM Innes' brain is a low horsepower model,
which even at full throttle, burns little fuel.

how would a 1600 know this? - the same person who happily contradicts GMs,
even world champions, but only recently decided he should actually play
chess with us ordinary folk, rather than mouth off about it?

computer geeks explain away why they are geeks, which has nothing to do with
chess or with science, and seems to me, something urgent in them to do with
a subject they cannot themselves express

my 'complaint' is that computer geeks are more want to speculate than test
their hypothesies. rather like corn-bot

if there is any actual science in any of these remonstrations, where is it?
the simple question is about the value of the books in terms of ratings
points, and when i challenged Crafty to play without book 6 years ago,
considering it an 1800 performer, the challenge was not accepted

no one has apparently tested an unaided prgroam for 10 years to achieve a
real rating - interesting, no? no one, for any program

if people such as corn-fed here, have no curiosity, they still need to
pretend something, that is, that they can exist via computer proxy to defeat
strong players [stronger than they are] and it is rare to encounter any
psychology which rings true, which goes beyond this motive.

phil innes



and thus that Phil Innes himself is boring. (They were
and he is.) Based upon past performance, I predicted that the
post before me would also be boring and so decided to delete it
unread.

Fair enough. Just don't try and tell us that your
"evaluation" was that his post was boring, while at
the same time admitting you didn't even read it;

I would never think of trying to tell you that, and indeed
I *didn't* tell you that. I am aware that you have spun a
fantasy where I have, but it is not based on anything I wrote
-- just your misinterpretation of same.

Put it another way: just shut up if you didn't read the
post, and let those of us who were stupid enough to
do so make the evaluation. Obviously, if you were too
smart to read the post, you know nothing about it
(except that it probably was not worth reading). We,
the really stupid people who actually read IM Innes'
postings, are the real experts here. Get over yourself.



but it is difficult to understand how you could possibly read
"You bore me" and conclude that I was writing that Phil Innes'
posting was boring, not that Phil Innes himself is boring. Who
do you imagine I was addressing? His post?

Idiot. Here is the exact quotation I was referring to,
and which you -- a complete idiot -- have attempted to
cover up:

Not true. One doesn't cover things up by quoting them.

Idiot. You have once again attempted to erase the
quote to which I replied and not-so-cleverly replace it
with another of your own choosing. Imbecile.

Here is the original quote again, the one where I
noted that you had not said that IM Innes was
boring, but rather that his *postings* we

"Until that day, I have better things to do that to read boring
posts."

Obviously, my commentary pointed out that
you erred by insisting that your original comment
had been that IM Innes himself was boring; as a first-
rate pedant, I couldn't help bot leap on this blunder.

IMO, in order to assess whether or not IM Innes, the
man, is boring, one would need to know him personally.
But in order to assess his postings, one would only
need to read them here, as we stupid people often do.
Frankly, you're simply not qualified to judge, because
you're much too smart to even read his postings.



I say you are a complete idiot only because

because you know that you are wrong and are trying to cover
up that fact by slinging childish insults? That's the usual
reason people resort to personal attacks.

My attacks are hardly personal. All I have done is
point out where you obviously blundered. Now, a
smart person learns from such mistakes, tries to
clarify their mind and root out the fogginess which
led to the problem in the first place. My guess is that
you have allowed emotions to gum up the works; that
you strongly dislike IM Innes for some reason.

Let me give you an example of how one can try to
remain objective: empty claims have been made with
regard to IM Innes' playing strength, and it would be
rather easy to imagine that he is a complete patzer,
and that his claims to nearly-2450-IMdom are utterly
baseless; but rather than do that, I went to the USCF
Web site and looked him up, to try and get a feel for
the reality of things. What I found was that the nearly-
an-IM is actually a very good player, just not at quite
the level which he had claimed.



it is so dreadfully obvious that you cannot erase google
posts;

Really? [http://groups.google.com/groups/msgs_remove]

That is not even a valid command. A valid command
looks more like this:

ERASE C:*.*

Don't just take my word for it, try it yourself.


-- help bot

On Aug 21, 5:18 pm, "Chess One" wrote:
"help bot" wrote in message

ps.com...

On Aug 20, 8:37 pm, Guy Macon http://www.guymacon.com/ wrote:

I sometimes wonder whether he actually engages his brain
while posting. Perhaps he is on autopilot when generating
boring and repetitive flames.

Perhaps IM Innes' brain is a low horsepower model,
which even at full throttle, burns little fuel.

how would a 1600 know this?

I would say by close observation, only he claims to
not even read your postings because, as he tells us,
they are too boring.


- the same person who happily contradicts GMs,

Better was: 1.e4!, as discovered by non-GM Fred
Reinfeld. (Okay, I admit that one famous GM liked
this move, but that was *after* FR had already
convincingly proved his case.)


even world champions

Rybka blocks in her Queen in the opening, if you
disable her openings book. What a patzer! In fact,
so does Fritz. (Look, if you want to know why it is
so important to not block in one's own Queen, look
no further than my games with Rob Mitchell. LOL)


but only recently decided he should actually play
chess with us ordinary folk, rather than mouth off about it?

A false dichotomy; it is quite easy to both mouth
off about it, AND play patzers. (Check your logic
chip.)


computer geeks explain away why they are geeks, which has nothing to do with
chess or with science, and seems to me, something urgent in them to do with
a subject they cannot themselves express

my 'complaint' is that computer geeks are more want to speculate than test
their hypothesies. rather like corn-bot

Not so; I only speculate because it is nearly-
an-im-possible task to check my speculations
via scientific experiment.

You see, testing would require a battery of
tests against a wide variety of opponents, with
known, reliable ratings, etc. -- things I have no
access to. Let me just say that IMO, for
whatever it may be worth, computers seem to
flub the openings positionally, but not so much
tactically, and this *could* translate into a drop
of, say, 200 rating points against strong opposition.
Against humans, it might not matter as much
because they are effectively taken out of book
and cast upon the rocks of their own ineptitude.


if there is any actual science in any of these remonstrations, where is it?

Not here. Here there exists only idle speculation,
yours and mine. Idle talk, backed by empty claims.
No science in that.


the simple question is about the value of the books in terms of ratings
points, and when i challenged Crafty to play without book 6 years ago,
considering it an 1800 performer, the challenge was not accepted

I expect that any player with the right connections
could have set this up, but still, you are only getting
ICC ratings, which then lose value when translated
into OTB ratings such as FIDE or USCF. In other
words, Bob Hyatt was not the only possibility, just
the most obvious one.


no one has apparently tested an unaided prgroam for 10 years to achieve a
real rating - interesting, no? no one, for any program

Very interesting. It seems difficult to face
the cold, hard fact of poor programming which
has been patched over by inserting rote moves
in a look-up book. Were it not for the vast
speedup in processors and increase in memory,
along with a titanic gap in tactical strength
between humans and computers, this problem
would still plague even the best programs. As
it is, they get by and as in the recent odds
match, strut their stuff in mid game.


if people such as corn-fed here, have no curiosity, they still need to
pretend something, that is, that they can exist via computer proxy to defeat
strong players [stronger than they are]

Can such players exist? At GetClub I was the
top-rated player in the entire world, by a good
margin. Still am, I suppose. At any rate, you
are beginning to bore me... .


-- help bot

Ralf Callenberg wrote:

If a computer program stands the Turing tests, nobody denies
its intelligence based on how it works internally. If a program
achieves to convince intelligent people that it is intelligent
- it is intelligent.

The claim that "If a computer program stands the Turing tests,
nobody denies its intelligence based on how it works internally."
is factually incorrect. Counterexamples abound:

http://en.wikipedia.org/wiki/Turing_test
"The term "Turing Test" is usually taken to indicate a test in
which a human judge converses with a human and a computer
without knowing which is which. It has been argued that the
Turing test is so defined that it cannot serve as a valid
definition of machine intelligence or "machine thinking"..."

http://www.acm.org/crossroads/xrds4-4/turing.html
"the Turing test is not an adequate test of intelligence." and
"I think that the process of arriving at a result is part of
intelligence"

http://www.cs.nott.ac.uk/~gxk/course...%201999-00.doc
"I would expect the student to categorically state if they
agree with Searle that passing the Turing Test does not
prove intelligence"

http://users.ecs.soton.ac.uk/harnad/...2001/0031.html
"In my opinion the Turing test does not prove a system's
intelligence"

http://www.seelrc.org/glossos/issues/8/janda.pdf
"passing the Turing Test does not prove that a computer can think."

--
Guy Macon
http://www.guymacon.com/

James wrote:

From a philosophical point of view, the discussion about computers
"understanding" what they do started with the Turing test and later
Searle's chinese room argument, and the so-called difference between the
syntaxic and semantic level of consciousness.

This is not much discussed today, even at the academic level. The
general view is much more pragmatic, and the general idea is that
"understanding" is a rather difficult thing to define; after all, as
many people have pointed out, the brain itself is just a lot of neurons
with chemical interactions and, from a physical point of view, there is
nothing that can define "understanding"; it is a subjective thinking,
linked only to your own consciousness: you think you are understanding
something, but it doesn't mean much.

"Learning" is also a very difficult thing to define. I wrote a reversi
program a few years ago that reached a quite high ranking on IOS, and it
was a program which fully learned its evaluation function (genetic
algorithm + pattern recognition). Was my program "understanding" Reversi
? Is learning reversi patterns the same thing as learning to speak?

The "expert system" paradigm has been slowly dying for the last 15
years, since the end of the FGCS japanese initiative (1992, Tokyo, 15
years already and I was there...).
Many computer science AI labs have even abandoned the name, and have
become "neural computing lab" or "evolutionary computing lab" instead...

I have been watching this field as an interested observer for years.
So many promising techniques. So few that end up with results
anywhere close to what was hoped. It is nice to see computers do
though-like things like playing chess or checking prescriptions
against medical histories looking fr interactions, but I hoped for
so much more.

--
Guy Macon
http://www.guymacon.com/

help bot writes:

On Aug 21, 11:04 am, Guy Macon http://www.guymacon.com/ wrote:

"Arguing with anonymous strangers on the Internet is a sucker's
game because they almost always turn out to be -- or to be
indistinguishable from -- self-righteous sixteen-year-olds
possessing infinite amounts of free time."
-Neil Stephenson

Does this sound to others the way it does to me? Like
a pathetic loser making excuses for losing arguments?

No. Bye bye.

--
Niko Wellingk