A Crime by a Board of Old Imposters

Rolf Tueschen wrote:
Sidney Cadot wrote:
Mr. Reul failed to [provide his source code]

Yes, right. Because there was no case.

We don't know whether there was a case or not. It's not for him to
decide, though. Just like, if the police turned up at your door with a
warrant for your arrest, they don't show you the evidence and let you
decide; they just arrest you.


And I gave other reasons why he was in time pressure plus stress.

You're speculating: you don't actually know why he didn't supply his
source code.


Dave.

--
David Richerby Old-Fashioned Dangerous Windows (TM):
www.chiark.greenend.org.uk/~davidr/ it's like a graphical user interface
but it could explode at any minute and
it's perfect for your grandparents!

Sidney Cadot wrote:

Rolf Tueschen wrote:

If you are a tournament director, you have the obligation to make
objective rulings in case of problems. As far as possible, you will have
to rely on the rules as set in the tournament's regulations. This is
what they did with the LIST incident; the tournament regulations
explicitly stipulate that you have to be prepared to show source code to
the tournament directorate in case of dispute.

Pardon me, you failed to show me a good argument since you forgot to
mention
why a tournament director should ask for such a code.

Because he is judging a dispute, and asking to see the code is the best
means of passing judgement; moreover, it is specifically stated in the
rules that this is within his jurisdiction. What's the problem?

What happened before? Know what I mean? And you are calling from a
university? Uhem.

No, I don't know what you mean. Could you perhaps be a bit clearer?

Mr. Reul failed to do so;

Yes, right. Because there was no case.

There was a dispute. As to whether there is a case: You don't know and I
don't know.

And I gave other reasons why he was in time pressure plus stress.

If this is the reason he didn't comply, I'm truly sorry for him.
However, the tournament director is also in time pressu he has a
tournament to direct, after all.

I think the tournament director and Mr. Reul should meet and try to work
out the situation (including a code review) and issue a joint statement,
in a few weeks time, about how sorry they are that things went the way
they did. This would also clear Mr. Reul, assuming he is indeed in the
clear.

this left the directors with no choice other than to disqualify his entry.

Yes, if you define them as dickheads. Please read what a number one expert
from
the USA is telling you about that decision.

Provide some references please.

Knee jerking comes to mind. Of course they had other possibilities.
In special let LIST play to the end. And research later.

Yes, this would also have been an option. However, as TD you have to
weigh: what if LIST must be disqualified afterwards, you'd have to
re-score the tournament, perhaps with consequences. The option chosen
now minimizes the influence of LIST on the end score. It's a though call.

As for the Shredder/Johnny threefold repitition: this case is not
clearly covered by the rules, so the tournament director must use his
best judgement in order to proceed. All this hinges on the fact that the
participants trust the tournament director to make a just and objective
call. If a participant does not subscribe to this, he should pack his
stuff and leave.

Objection, Sid. I would always believe Henk nut Henk had the wrong question
he
answered.

I truly don't understand this sentence. Could you rephrase?

And who is this Henk nut Henk you speak about? I'm very much prepared to
discuss this issue with you, but please abstain from this childishness.

Jonny author wanted to play on although machine said REMIS. Know what
I mean? - You are correct. What Henk decided was ok. But the question was a
different one.
Please read what Hyatt wrote about it. And also Amir. Now Henk looks like a
clown(e).

No, I don't know what you mean. Who is Henk?

I've been a judge at an entirely non-related type of event (a
programming contest, if you want to know), and sometimes things happen
that are not planned. You then consult the rules and do your job, which
is to take hairy decisions in hairy situations. If you cannot do this,
or are not able to do this when push comes to shove, you shouldn't be in
that position.

Right. I agree. But you must also be sure that you have the details of the
problem. That wasn't kosher for both cases. 1. No need to throw out LIST
for
the last 3 rounds.

If a competitor does not comply by the rules, there's little choice. Now
all the exam pressure and stuff, that's all very well, but you have to
be available for discussing issues with the TD, or otherwise appoint a
spokesperson. It's not like the TD didn't try to contact Mr. Reul.

2. T.D. must ask what a player plans to do or did do.
Again read Hyatt, also a professor, Sid.

Please provide a reference.

As for your cheapshots at Prof. van der Herik's alleged "academic
sluggishness": I have met the man, and although I have no particular
like or dislike of him, I feel that he is the perfect man for the job
and would always try to pass objective judgement, within the framework
of the written rules if possible, but based on what is just in other
circumstances. I don't think you have a clue on his standing in the
computer chess community. I have his PhD thesis sitting in my bookshelf,
which is about using advanced search algorithms for chess endgames, he
wrote this when your parents probably didn't even conceive of conceiving
you. Who are you to judge this man?

Actually I would be the best you can get.

What are your qualifications? Academically, and in the chess world?

BTW I met him in person. Very sympathetic man. His science is another
problem.

I take it you mean "issue".

Here we are talking about a
spare time job. And you become old at times, Sid. You know what, you
remember
me in your argumentation of the German Ingo Althöfer. Also a prof. And
because
he's a prof for a field in mathematics he thinks that he can make the best
experiments in Dreihirn and stuff like that. But he forgets about the
documentation... Know what I mean?

No I don't know what you mean. In fact this entire paragraph doesn't
make much sense to me. Could you please be somewhat clearer?

Ingo is as amateurish as he can be. Brightness in maths seldom speaks for
talents in all other fields. But he always behaves as if he had something
important to decide with his vote.

And this Ingo fellow is relevant to our discussion because.... ?

With the same sluggishness Henk overlooked what I saw in seconds. Hyatt
again:
there are no _rules_ for what happens when someone accuses a collegue of
plagiatism. But such rules should exist.

Ah, there's the mysterious "Henk" again.

Could you provide references to "Henk" and Hyatts reaction? The former I
doubt exists, the latter I haven't read.

But if they don't then I have a new freedom to make decisions also in dubio
pro
reo! Know what i mean?

Golly, no. Please work a bit on your communication skills, this is
getting ridiculous. Feels a bit like I'm talking to a 1960's Lisa program.

Until now nobody said exactly what the reasons had been for suspicions. I
must
conclude that there are no reasons.

No, you must not. It is perfectly normal not to publicise suspicions
without due process having taken place. In fact, it would be unethical
not to do so.

BTW if there had been strong reasons then all is fine. But I tend to
believe
that Fritz Reul is kosher.

We can all agree, I think, that we hope this has all been a tragic
misunderstanding. But I have no reason to believe one way or another.

Of course Mr. Reul is innocent until proven guilty. However, given the
nature of the suspicions, he should cooperate to clear his name. I'm
sure the TD would be very, very willing to help in that respect.

If you have doubts with my reasoning then please reflect how Henk would
have
decided if someone had accused SHREDDER of having Crafty code. You believe
Henk
had made the same decision???

Do you mean Prof. van der Herik when you say "Henk" ?


Yes, of course. I'm sorry if his prename isn't Henk. Above the sentence meant
because of a typo:

"Henk *b*ut Henk" - I hope that this helps now.

Further discussion with you is useless because you simply don't answer on the
remarks one important expert had made. Robert Hyatt.

Also I saw that you don't care for the already committed violation of Reul's
integrity.

On the other side it's nice that you also hope that the truth will be found -
also with the help of H.v.d.H.


Rolf





If someone had presented founded suspicions to this effect, then, yes, I
think Prof. van der Herik would act similarly.

Finally, all the best to your own career,

Well, quite. Same to you then.

Sidney

Rolf Tueschen wrote:

Do you mean Prof. van der Herik when you say "Henk" ?

Yes, of course. I'm sorry if his prename isn't Henk. Above the sentence meant
because of a typo:

"Henk *b*ut Henk" - I hope that this helps now.

Fair enough. His first name is "Jaap".

Further discussion with you is useless because you simply don't answer on the
remarks one important expert had made. Robert Hyatt.

I seriously haven't seen his remark. Where can I read it?

Also I saw that you don't care for the already committed violation of Reul's
integrity.

I don't acknowledge such a violation was made. It is my assessment that
the TD did the best he could in this case, but unfortunately things
sometimes don't work out.

On the other side it's nice that you also hope that the truth will be found -
also with the help of H.v.d.H.

Yes. I'm sure Prof. v.d. Herik isn't at all happy with the two
incidents I am sure; I'm confident that the rules will be clearer
regarding both issues next time. And I am sure he will help Mr. Reul to
clear his name, if all this was indeed an unfortunate mistake.

In the case it wasn't a mistake, and there was indeed a genuine problem
with Mr. Reul's program, it would be best if he owned up to it. He's a
young guy, Dr. Hyatt's freely available Crafty source may have just been
too seductive... I'm sure most people would be willing to forgive him.

What I am afraid of is that this will keep on lingering if Mr. Reul is
not prepared to give access to his source code for inspection. To most
people (including me), that would be suspicious. As it would be in the
case of Fritz or Shredder, by the way. For code inspection, the TD and
helpers should be under a stict non-disclosure agreement, and it should
be agreed that their judgement will be final.

As to everything else, well, we can just agree to disagree.

Regards,

Sidney Cadot

ssince you say you are an attorney, and i have no reason to assume
differently, why do you insist the the central question is 'copyright law?'
this is a misguided premise.
it is misguided because the author of crafty is NOT the plaintiff and there
are no economic damages at issue, mr attorney.
the programmer of list broke a contract. he refused to provide information
persuant to a claim of an ethics violation from one of his competitors in a
contest where he contractually agreed to a set of rules and a method of
arbitration.
"Znarf" wrote in message
...
I never called the ICGA imposters; it was the subject to which I replied.
Please double check your facts. Thus, it appears the premise of your
retort
is faulted. Additionally, read again my original post; there are no
denigrating states of the ICGA per se.

As for the losing side of a debate, I suggest you again look at the
merits
of your position. I am an attorney and know about the law (at least in
the
US) of copyright infringment and the burden of proof in both civil and
criminal court. While the ICGA is not a judicial body, it was acting in a
quasi-judicial role. My point is some procedural process should be in
place
rather than an ad-hoc demand for source code within a couple of days. The
culpability of the accused is not my concern, it is the process by which
the
accused was *ASSUMED* culpable and had the *BURDEN OF PROOF* to prove he
was
not culpable that I found to be troublesome, and the willingness to
publish
an adjudication without giving the accused a *REASONALBE AMOUNT OF TIME*
to
respond.




"marc margolies" wrote in message
...
lemme see, you called the tournament commitee of the igca 'a bunch of
imposters' because you don't like their decision, but of course my tone
offends you.
i suggest you resort to critiquing the manners of others and
disregarding
your own because you are on the losing side of a stupid debate than
cannot
be won on the basis of facts and content.
"Znarf" wrote in message
...
It is not BS. My subjective belief is irrelevant; I do not know if he
copied any code or not. My point is that giving a person roughly 48
hours
to answer allegations of copyright infringement and then banning him
for
a
couple of years is not a fair procedural process. Perhaps the later
poster's recommendation could be used - source code is provided under
a
non-disclosure agreement and may be examined ONLY to evaluate copying
allegation.

As for your offensive tone, it is not warranted, and only illustrates
the
"guilty until proven innocent" ignorance of unfair people.


"marc margolies" wrote in message
...
since you defend the programmer, then you don't think his
'reputation
is
damaged beyond repair" either or you wouldn't bother trying to do
just
that.
cut the bull****.

On Fri, 28 Nov 2003 02:17:00 GMT, "Znarf" wrote:

What I find interesting is the following quote at
http://chessbase.com/newsdetail.asp?newsid=1330 :

"The program List is suspected to be a clone of the program Crafty. Autor
Fritz Reul failed to prove otherwise and allowed a final deadline to pass."

So apparently an author's reputation and integrity can be maligned on
"suspicion." What is truly tragic is the "assumed guilty" posture of the
accused. LIST was *SUSPECTED* to be a clone, and was disqualified because
the author *FAILED TO PROVE* otherwise. How incredibly unjust.

Furthermore, it does not appear that the ICGA followed thier own rule:

"Each program must be the original work of the entering developers.
Programming teams whose code is derived from or including game-playing code
written by others must name all other authors, or the source of such code,
in their application details. PROGRAMS WHICH ARE DISCOVERED TO BE CLOSE
DERIVATIVES OF OTHERS (e.g., by playing nearly all moves the same), may be
declared invalid by the Tournament Director after seeking expert advice. For
this purpose a listing of all game-related code running on the system must
be available on demand to the Tournament Director."

LIST was ONLY SUSPECTED of being a clone; it was NOT DISCOVERED TO BE A
CLONE. The rule, as written, places the BURDEN on the ICGA to prove it is a
close derivate of another before disqualification; it does not place the
burden on the accused to prove that it is not a derivative. Thus, the rule
is inapplicable to the present situation.

The "proving" is done via inspection of the source code, among other
things.

The ICGA needs a procedure to follow in resolving these disputes. Apparently
it has none, so it made a procedure up at the expense of an author's
reputation.

It has a procedure to follow. Mr. Reul failed to allow that procedure
to take place, as required by the rules of the ICGA.

Furthermore, the ICGA has now possibly damaged the author's
reputation beyond repair. Allegations of copyright infringement are serious
concerns in the software community. Finally, the ICGA should have
accommodated the accused author's schedule - is it too much to ask to give a
person a small reprieve while he tends to examinations rather than publicly
call into question the author's integrity in a worldwide publication on the
Internet.

Such was offered to Mr. Reul, but never accepted.


A public retraction is in order, and an apology.

Certainly not! Mr. Reul abused the ICGA rules, which he knew before
the tournament began. It is he who should apologize to the ICGA, not
the other way around.

Fritz Reul did not FAIL to prove he was innocent - he failed to agree
to supply his source code (in a confidential manner if he likes, with
him present at the examination of the code), to the ICGA expert
official.

The ICGA was willing to do this after his exams, in nearby Frankfurt.

Important to remember that the code examination itself would be the
means the ICGA would use to discover a derivative program.

The ICGA has no crystal ball to detect cheats, that's why they have
the examination of source code provision in the first place.

What "procedure", other than an examination by an expert, would you
prefer the ICGA to use?

Unfortunately, Mr. Reul was too busy to call the ICGA back, even after
they spoke to his operator, and even called his Mom.

Given the circumstances, if Mr. Reul's program was indeed original, he
should have taken 5 minutes to call them back and set up the meeting
in Frankfurt, after his exams.

These are the rules which List's author knew before the tournament
started. It's not a simple match with the chess program from across
town. This is the World Championships for computer chess, and Fritz
Reul, and everybody else, should be expected to follow them.

The ICGA needs to change it's rules, and they have already stated that
will indeed be the case for subsequent tournaments, to minimize this
type of problem.

They never called Mr. Reul a cheater, that was the "Sun" (aka "the
wapping liar").

I believe they did as good a job as they could under the
circumstances, and should be applauded for going the extra mile to
arrange for a closer, entirely confidential, and later meeting with
Fritz Reul.

Darrz

Sidney Cadot wrote:
Rolf Tueschen wrote:

Yes, that is correctly described because there are no rules for what happens if
someone just takes his insinuations out of the blue.

Somehow the LIST affair and then later the Jonny present to SHREDDER and SMK
which was agreed by the tournament director Jaap vdH - all that will hopefully
lead to the retreat of these imposters. They simply have no longer the spirit
and sophistication to make correct decisions. They are - so to speak - blinded
by a whole army of blind spots resulting of their academic sluggishness.

I think you have no idea what you are talking about.

If you are a tournament director, you have the obligation to make
objective rulings in case of problems. As far as possible, you will have
to rely on the rules as set in the tournament's regulations. This is
what they did with the LIST incident; the tournament regulations
explicitly stipulate that you have to be prepared to show source code to
the tournament directorate in case of dispute. Mr. Reul failed to do so;
this left the directors with no choice other than to disqualify his entry.

As for the Shredder/Johnny threefold repitition: this case is not
clearly covered by the rules, so the tournament director must use his
best judgement in order to proceed. All this hinges on the fact that the
participants trust the tournament director to make a just and objective
call. If a participant does not subscribe to this, he should pack his
stuff and leave.

Sorry, but this _was_ covered by the rules. The human operator is
"passive". If, at any time during a game, it is discovered that the
operator played a move different than the move the program displayed,
or that the operator entered a move different than what the opponent
played, then the game backs up to that point _immediately_ and the
correct move is entered/played and the game proceeeds from that point.

The point has _always_ been that this is computer vs computer. The
human is there to assist the two players, just like the assistant a
blind player is allowed to use. But this "assistant" may _not_
participate in the game in any way. Many programs can't accept draws
nor offer them. So unlike FIDE events, the computer chess TDs have
_always_ had absolute say on draws. If I want to offer my opponent
a draw, or if I want to resign, I have to first ask the TD. He will
look at the board and say "OK or play on". I have had _both_ happen
to me. If the program offers a draw, and the opponent program can
accept draws, then the two programs "agree" and the result is final.
If either opponent requires a human to make the decision to accept or
decline a draw, then the TD has final say. No such rule in FIDE. But
there is in CC. And here it was _clearly_ violated. The human saw
the claim of "3-fold repetition". He should have claimed a draw and
called the TD over to verify it. The TD doesn't get to say "Hey the
GUI is claiming draw but the engine is not." The TD only gets to
say "the claim is valid or it is not." Here the TD and the operator
went _way_ outside the rules, clearly.

Ken Thompson and I have been harping on eliminating the operators since
the late 1970's. I wrote software to take a microcomputer with a bunch
of serial ports, and use that to connect multiple programs so that they
could communicate directly. The software I wrote managed the clock,
the programs could query the clock and send moves back and forth (somewhat
like auto232 but much more reliably). The ICCA said "sounds great, but we
have some commercial programs that have dedicated hardware and they don't
have a serial port to use, so we aren't going to use this approach." Next
I suggested installing a FICS-type server on a laptop, having a LAN at
the tournament hall, and requiring everyone to use an FICS-capable
interface (we have this requirement for the CCT event we hold every 6
months on ICC). The ICCA again said "some commercial programs don't have
a GUI that supports internet chess so we won't use this idea." In the CCT
events we _require_ this for a participant. Our last event had 3x _more_
entries than the just-completed WCCC. Why they can't move to the 21st
century is beyond me.

This could have been avoided with a LAN and no operators. It wasn't. It
could have been avoided by the TD making a rational decision. It wasn't.

The final result is a joke, the title is tainted, and the after-smell will
haunt the ICGA for years. Shredder may very well be the best program there.
I don't question that at all. But more than once I had the best program at
a CC event and lost due to a bug or whatever. I simply looked my opponent
in the eye and said "good game" and moved on.

This was a blunder of ridiculous proportions. It could have been avoided.
It will only serve to make the CCT (held on ICC) the main computer chess
event of the year from here on.

How anyone could reach a decision allowing a program operator to influence
not only the outcome of a game, but the outcome of the tournament defies
any logical explanation.



I've been a judge at an entirely non-related type of event (a
programming contest, if you want to know), and sometimes things happen
that are not planned. You then consult the rules and do your job, which
is to take hairy decisions in hairy situations. If you cannot do this,
or are not able to do this when push comes to shove, you shouldn't be in
that position.

yes, but here there _were_ rules. The spirit of the WCCC/WMCCC events
has _always_ been that this is a contest between two programs, and that
the humans have very limited influence on the games. Just read the rules
about incorrect moves, changing times, changing engine parameters, etc.
The human is _very_ limited in what he can/can't do, and this decision let
one go _far_ beyond what was allowed.


As for your cheapshots at Prof. van der Herik's alleged "academic
sluggishness": I have met the man, and although I have no particular
like or dislike of him, I feel that he is the perfect man for the job
and would always try to pass objective judgement, within the framework
of the written rules if possible, but based on what is just in other
circumstances. I don't think you have a clue on his standing in the
computer chess community. I have his PhD thesis sitting in my bookshelf,
which is about using advanced search algorithms for chess endgames, he
wrote this when your parents probably didn't even conceive of conceiving
you. Who are you to judge this man?

I think he is simply "in over his head." I participated in CC events
from 1976 to 1994 when the last ACM event was held. For most of those
IM Mike Valvo was the T.D. and we _never_ had such irrational decisions.
This is _not_ the first gaffe Jaap has made. I happen to like him myself,
I've known him for years. But as a TD, he is simply in over his head.

One simple example. I think it was in 1995 that a program played a move
and displayed it on the GUI chessboard. In looking at the analysis window
the engine said "move X is best". But somehow the GUI displayed the wrong
move (it was a pawn promotion and the engine wanted to promote to queen
and the GUI said knight, or vice-versa). Jaap said "we use the GUI move,
not what the engine is showing." Reasonable? Flash forward to 2003. The
GUI says 3-fold repetition. The engine doesn't (apparently) understand
3-fold repetition. Japp says "we use the engine's analysis, not the GUI."

Does that seem rational or logical? Not to this long-time CC
competitor. If you look at Mike Valvo's decisions, they were consistent
from the first tournament he directed to the last one. That is vital.


Sidney Cadot
The Netherlands


--
Robert M. Hyatt, Ph.D. Computer and Information Sciences
University of Alabama at Birmingham
(205) 934-2213 136A Campbell Hall
(205) 934-5473 FAX Birmingham, AL 35294-1170

Rolf wrote:

Yes, if you define them as dickheads. Please read what a number one expert
the USA is telling you about that decision. Knee jerking comes to mind. Of
course they had other possibilities. In special let LIST play to the end. And
research later.

There were three possible actions for the list problem:

1. Research this before the event started and not allow it to play
if he couldn't convince them it was not Crafty.

2. Kick list out in the middle of the event, with all the bad publicity
and harm to the programmer's reputation that would incur, not to mention
the unfairness to the tournament standings. List was pretty strong. Some
had already played it. After it was taken out, everyone beyond that
poing got a free point. Those before that point lost or drew with it
and got less, simply because they were unlucky in the pairings.

3. Investigate and make the decision _after_ the tournament ended. Then
the result could stand with List taken out of the standings, which let the
tournament have a reasonable result with no bad luck / good luck stuff.

ICGA took the _worst_ possible case of the three, #2.

They've ruined the programmer's reputation. The tournament standands are
somewhat skewed. And nobody is really happy. (Of course they have much
bigger problems in the terrible draw decision that robbed Fritz of the
WCCC title it actually earned.)



Objection, Sid. I would always believe Henk nut Henk had the wrong question he
answered. Jonny author wanted to play on although machine said REMIS. Know what
I mean? - You are correct. What Henk decided was ok. But the question was a
different one.
Please read what Hyatt wrote about it. And also Amir. Now Henk looks like a
clown.

The operator can't make a decision. He is simply a human "I/O device"
that is responsible only for relaying moves between the program and the
game board. This operator went beyond that. And the TD allowed it to
stand. All of which is contrary to published rules for ICGA computer
chess tournaments.



Right. I agree. But you must also be sure that you have the details of the
problem. That wasn't kosher for both cases. 1. No need to throw out LIST for
the last 3 rounds. 2. T.D. must ask what a player plans to do or did do.
Again read Hyatt, also a professor, Sid.

I have been on both sides of this. I lost a game in 1977 when my program
was the one guilty of screwing up and displaying one move while really
playing another. I simply went home and fixed it so that that would not
happen again. I didn't complain. I didn't try to nit-pick and get into
this "the GUI said this, the engine said this, the GUI is not really the
engine" nonsense. This game has two players. Each player is comprised
of hardware and software. It is a _single_ entity. Not multiple pieces
where one can be eliminated from the discussion.



Actually I would be the best you can get. BTW I met him in person. Very
sympathetic man. His science is another problem. Here we are talking about a
spare time job. And you become old at times, Sid. You know what, you remember
me in your argumentation of the German Ingo Althofer. Also a prof. And because
he's a prof for a field in mathematics he thinks that he can make the best
experiments in Dreihirn and stuff like that. But he forgets about the
documentation... Know what I mean?

I've known Jaap for years. I consider him a friend. But he totally _blew_
this decision. He has blown others. He's a good person, but he is not a
good TD, IMHO. IM Mike Valvo ran ACM and WCCC events for years. These
things did not come up. When problems arose, he made decisions based on
the rules. Not upon witchcraft and superstition and a ouiji board.

snip

Until now nobody said exactly what the reasons had been for suspicions. I must
conclude that there are no reasons. BTW if there had been strong reasons then
all is fine. But I tend to believe that Fritz Reul is kosher.

I have no opinion whatsoever, so my gut feeling says "innocent until proven
guilty" since we have absolutely _no_ information of any kind. That makes
this even worse. They should at _least_ release what led to their decision,
to support their decision in public light. Now all we have is the "whisper".






--
Robert M. Hyatt, Ph.D. Computer and Information Sciences
University of Alabama at Birmingham
(205) 934-2213 136A Campbell Hall
(205) 934-5473 FAX Birmingham, AL 35294-1170

Robert Hyatt wrote:


2. Kick list out in the middle of the event, with all the bad publicity
and harm to the programmer's reputation that would incur, not to mention
the unfairness to the tournament standings. List was pretty strong. Some
had already played it. After it was taken out, everyone beyond that
poing got a free point. Those before that point lost or drew with it
and got less, simply because they were unlucky in the pairings.

I don't know how the pairing and scoring went in the tournament but would
it not have been fairer to declare all games void and award everyone the
same score against List?

Regards Charley

Robert Hyatt wrote:

Sidney Cadot wrote:

As for the Shredder/Johnny threefold repitition: this case is not
clearly covered by the rules, so the tournament director must use his
best judgement in order to proceed. All this hinges on the fact that the
participants trust the tournament director to make a just and objective
call. If a participant does not subscribe to this, he should pack his
stuff and leave.

Sorry, but this _was_ covered by the rules. The human operator is
"passive". If, at any time during a game, it is discovered that the
operator played a move different than the move the program displayed,
or that the operator entered a move different than what the opponent
played, then the game backs up to that point _immediately_ and the
correct move is entered/played and the game proceeeds from that point.

My point it that "making a claim" is a vague wording.

It is entirely possible for a chess program GUI to list all available
moves it has at its disposal at any given time. Likewise, it could also
provide status information on the castling status, en passent status,
50-move rule status, and threefold-repetition status.

If the GUI is decoupled from the engine, it could even make a repitition
claim without being told to do so by the engine. That would not be
acceptable to me. Such a GUI should be explicitly abolished, if only to
prevent misunderdstandings.

Like a human player, a chess program is free not to claim a draw based
on the threefold repetition rule, even though it could. It could still
hold some internal status information: "ok, I could take the draw at
this point, but let's see". It could also show the "draw-ability" in the
GUI.

Now if the Johnny program (prompted by the engine) popped up a dialog
saying "threefold repitition - draw", I'd be 100% with you on this.

However, if it just showed an internal flag, I wouldn't be.

I don't know the Johnny GUI and I wasn't there, so I don't know. All I'm
saying is that it would be wise if the rules explicitly stated what a
"claim of draw", as instigated by the engine, looks like.


The point has _always_ ben that this is computer vs computer. The
human is there to assist the two players, just like the assistant a
blind player is allowed to use. But this "assistant" may _not_
participate in the game in any way. Many programs can't accept draws
nor offer them. So unlike FIDE events, the computer chess TDs have
_always_ had absolute say on draws. If I want to offer my opponent
a draw, or if I want to resign, I have to first ask the TD. He will
look at the board and say "OK or play on". I have had _both_ happen
to me. If the program offers a draw, and the opponent program can
accept draws, then the two programs "agree" and the result is final.
If either opponent requires a human to make the decision to accept or
decline a draw, then the TD has final say. No such rule in FIDE. But
there is in CC. And here it was _clearly_ violated. The human saw
the claim of "3-fold repetition".

I'm with you all the way. If this constituted a "claim" by Johnny,
something went horribly wrong. Both the operator and (especially) the TD
should know better.

He should have claimed a draw and
called the TD over to verify it. The TD doesn't get to say "Hey the
GUI is claiming draw but the engine is not." The TD only gets to
say "the claim is valid or it is not." Here the TD and the operator
went _way_ outside the rules, clearly.

If this is the case, I'm surprised that the Fritz operators didn't
object. Could be commercial interest: it's better not to be seen as
"sore losers".

Ken Thompson and I have been harping on eliminating the operators since
the late 1970's. I wrote software to take a microcomputer with a bunch
of serial ports, and use that to connect multiple programs so that they
could communicate directly. The software I wrote managed the clock,
the programs could query the clock and send moves back and forth (somewhat
like auto232 but much more reliably). The ICCA said "sounds great, but we
have some commercial programs that have dedicated hardware and they don't
have a serial port to use, so we aren't going to use this approach." Next
I suggested installing a FICS-type server on a laptop, having a LAN at
the tournament hall, and requiring everyone to use an FICS-capable
interface (we have this requirement for the CCT event we hold every 6
months on ICC). The ICCA again said "some commercial programs don't have
a GUI that supports internet chess so we won't use this idea." In the CCT
events we _require_ this for a participant. Our last event had 3x _more_
entries than the just-completed WCCC. Why they can't move to the 21st
century is beyond me.

On this, I couldn't agree more.

This could have been avoided with a LAN and no operators. It wasn't. It
could have been avoided by the TD making a rational decision. It wasn't.

The final result is a joke, the title is tainted, and the after-smell will
haunt the ICGA for years. Shredder may very well be the best program there.
I don't question that at all. But more than once I had the best program at
a CC event and lost due to a bug or whatever. I simply looked my opponent
in the eye and said "good game" and moved on.

This was a blunder of ridiculous proportions. It could have been avoided.
It will only serve to make the CCT (held on ICC) the main computer chess
event of the year from here on.

How anyone could reach a decision allowing a program operator to influence
not only the outcome of a game, but the outcome of the tournament defies
any logical explanation.

I wish I had been there to see it happen. If the Johnny program's
threefold repetition message constituted a "claim" and not just a status
message, I do agree that the TD was indeed at fault here.

I've been a judge at an entirely non-related type of event (a
programming contest, if you want to know), and sometimes things happen
that are not planned. You then consult the rules and do your job, which
is to take hairy decisions in hairy situations. If you cannot do this,
or are not able to do this when push comes to shove, you shouldn't be in
that position.

yes, but here there _were_ rules. The spirit of the WCCC/WMCCC events
has _always_ been that this is a contest between two programs, and that
the humans have very limited influence on the games. Just read the rules
about incorrect moves, changing times, changing engine parameters, etc.
The human is _very_ limited in what he can/can't do, and this decision let
one go _far_ beyond what was allowed.

Ok.

As for your cheapshots at Prof. van der Herik's alleged "academic
sluggishness": I have met the man, and although I have no particular
like or dislike of him, I feel that he is the perfect man for the job
and would always try to pass objective judgement, within the framework
of the written rules if possible, but based on what is just in other
circumstances. I don't think you have a clue on his standing in the
computer chess community. I have his PhD thesis sitting in my bookshelf,
which is about using advanced search algorithms for chess endgames, he
wrote this when your parents probably didn't even conceive of conceiving
you. Who are you to judge this man?

I think he is simply "in over his head." I participated in CC events
from 1976 to 1994 when the last ACM event was held. For most of those
IM Mike Valvo was the T.D. and we _never_ had such irrational decisions.
This is _not_ the first gaffe Jaap has made. I happen to like him myself,
I've known him for years. But as a TD, he is simply in over his head.

That could be. My main reason for reacting to Mr. Tueschen's message was
the foul tone this thread had evolved to. I am sure, as I think you
are, that Prof. van der Herik had no malintent. Of course it is fair
game to discuss his decisions as TD, but I'd prefer to keep this civil.

One simple example. I think it was in 1995 that a program played a move
and displayed it on the GUI chessboard. In looking at the analysis window
the engine said "move X is best". But somehow the GUI displayed the wrong
move (it was a pawn promotion and the engine wanted to promote to queen
and the GUI said knight, or vice-versa). Jaap said "we use the GUI move,
not what the engine is showing." Reasonable? Flash forward to 2003. The
GUI says 3-fold repetition. The engine doesn't (apparently) understand
3-fold repetition. Japp says "we use the engine's analysis, not the GUI."

If it was the GUI claiming 3-fold repetition, this would count as
"status info" in my view. If the engine cannot claim 3-fold repetition,
I would say that a draw cannot be claimed. A "claim" to me is almost
like a move: it's an active action that the "thinker" (human or
computer) must make. To me, an engine that doesn't know how to claim a
draw is only partially a chess-playing engine; I don't see a whole lot
of difference to it and an engine that can't make Knight's moves.

I think the 1995 incident was completely mishandled by Mr. van der
Herik, assuming your recollection is good.

Does that seem rational or logical? Not to this long-time CC
competitor. If you look at Mike Valvo's decisions, they were consistent
from the first tournament he directed to the last one. That is vital.

Yes.

Best regards,

Sidney Cadot

Robert Hyatt wrote:

[snip] I didn't try to nit-pick and get into
this "the GUI said this, the engine said this, the GUI is not really the
engine" nonsense. This game has two players. Each player is comprised
of hardware and software. It is a _single_ entity. Not multiple pieces
where one can be eliminated from the discussion.

So, if I understand correctly: it is your opinion that the human player
is just an I/O device, but the GUI isn't (it's "part of the same single
entity").

I think that is a strange position to take. I feel that the GUI is also
just an external I/O device, like the human operator; For most bare
engines, the GUI is driven by a protocol and runs as a separate process.
To me, it's pretty clear who's in charge.

You pointed out some examples from your own experience where the engine
and the GUI disagreed. What should be the course of action in cases like
that?

Should we follow the engine or the GUI? Or should we ask the TD (or toss
a coin)?

Best regards,

Sidney