CCT: Update to rules for future events

[Peter Skinner]

But as I get the rule, you are particularly referring to engines, not original author, in this regards SF is based on Glaurang and it is not an original work itself.

Didn't Glaurung turn into Stockfish, being that Glaurung stopped being actively developed, and the program became what we now know as Stockfish, just from a team?

Correct me if I am wrong... I could be..

Peter

[bob]

Not good enough yet....a lot of holes hangs out there....
Dr.D

Then by all means, provide me with suggestions and let's openly talk about them.

I _encourage_ people to make suggestions. This is after all _your_ tournament. I hold it purely as my contribution to this community for everything it has given me in the way of enjoying chess programs for the last 17 yrs or so.

Do you _really_ think it is fun to have to stay up all night waiting for games to finish, record the games, deal with childish sniping, and above all be bored half to death at times because no one is chatting?

I try to make it as fair as possible for everyone.

Input is not only accepted, it is encouraged.

Peter

My suggestion is simple. ICC/FICS really can't handle disconnects very well. You can either disallow them completely (one disconnect = loss) or you allow an infinite number, which is also bad. I wish there was an option where you set it to N, rather than 0 (infinite disconnects) or 1 (no disconnects) so that it would terminate the game on the nth disconnect. Since this is not possible, I think the rule should simply be purely draconian. 3rd disconnect is a forfeit. Someone informs the TD, the game ends. Don't require it to be a team member, or the operator, every game has observers, let any of them send the TD a tell, the TD can verify without using that person's name, the game ends, life goes on. I don't like the "out" that one _could_ continue to play on. I'd like to play on beyond mate at times. For me, rules without exceptions are far easier to deal with, cause fewer arguments, and keep things totally fair.

I think it needs to be a team member, unless the TD has a way for verifying the disconnect count. We can't trust observers to count. When Telepath is in a match and a disconnect happens for either side, I get email with the pgn and who disconnected. It is easy for me to look at my email and count the disconnects.

For instance, lets say A disconnects twice and then B disconnects. How many spectators are going to think A disconnected 3 times? Some sort of viewable evidence is needed before the call is made or both opponents agreeing on what happened.

I agree that the TD and/or organizer can't keep up with all games accurately in a large tournament.

Is there a way for the TD/Organizer to get the pgn email of each game as they end and/or disconnect? That would make things easier.

I think that anyone should be able to report the problem, then the TD has to verify. Best idea would be to notify the TD for _every_ disconnect since there should not be many per round. Then he can count 'em. What if the operator is away getting something to eat, since these things happen at all sorts of times depending on where you are? Or he's gone to the little boy's room. Or fell asleep if round 1 starts at midnight?

I'd be just as happy to play with noescape=1 and solve this problem once and for all. A disconnect is a loss.

In thinking further, if the TD sets up a second account that auto-observes every game, and sends the chat stuff to a file, he could see every disconnect and verify what has happened easily... I always have Crafty capture its ICC messages and send 'em to a file... I have a file that now goes back several years. Currently 128.8mb of "stuff".

[Look]

But as I get the rule, you are particularly referring to engines, not original author, in this regards SF is based on Glaurang and it is not an original work itself.

Didn't Glaurung turn into Stockfish, being that Glaurung stopped being actively developed, and the program became what we now know as Stockfish, just from a team?

Correct me if I am wrong... I could be..

Peter

You are right in this, but my point is that, the rule you have stated is not clear enough in this situation, and could be subject to personal interpretations, so IMO it makes sense to clarify the rule so what you are saying here could be clearly inferred.

[Peter Skinner]

If you are referring to Rybka, then please show me something to state the _current_ version is a clone of anything. If someone can provide solid reasoning behind the accusation, I would ask for the code to be sent to the team that I retain each CCT.

I am not referring to anything in particular, just trying to see if your rules are fair. Also interesting is the fact that you are interested in the _current_ version of an engine. In this regards soon codes of some new engines accused of being clones, could be different from what they are now. So they could compete in the tournament isn't it?

No what I am saying is while nothing was definitively proved that Rybka was a clone of Fruit (There was some evidence), I am quite sure that Rybka 2,3,4 aren't or have any offending code in it.

I could be wrong, but if a challenge is put forward, then it will be investigated. It has to be credible, with evidence, and from a participant.

I should have made that clear with the rule. It has to be a participant, not a spectator making the claim. If I tried to hunt down every claim at every CCT from a spectator I would never get anything done.

If someone wants to put forth a valid claim, so be it, but I have a feeling this would have been done at tournaments like Leiden, WCCC, IPCCC, Dutch open... you get the idea.

Peter

[Peter Skinner]

I think that anyone should be able to report the problem, then the TD has to verify. Best idea would be to notify the TD for _every_ disconnect since there should not be many per round. Then he can count 'em. What if the operator is away getting something to eat, since these things happen at all sorts of times depending on where you are? Or he's gone to the little boy's room. Or fell asleep if round 1 starts at midnight?

I'd be just as happy to play with noescape=1 and solve this problem once and for all. A disconnect is a loss.

In thinking further, if the TD sets up a second account that auto-observes every game, and sends the chat stuff to a file, he could see every disconnect and verify what has happened easily... I always have Crafty capture its ICC messages and send 'em to a file... I have a file that now goes back several years. Currently 128.8mb of "stuff".

If we use the mamer bot to run the tournament, I am notified of all game interactions. I can easily count the number of disconnections, but when manually pairing rounds, I don't see every little disconnection.

This coming CCT, I will be logging the screen with Winboard from a second account so that I can capture everything.

Peter

[bob]

Clone/Derivatives Rules

1. Each participant (engine) must an original work. No entry can contain code from another program, or be a "clone" of another program. This includes any "personality" settings of an originating program. This includes opening books.

Hi,

Does it include StockFish too?

I believe that Tord is still a contributing author of Stockfish, so I see no reason why it wouldn't be allowed to enter.

If someone wanted to challenge the code; well it is available for download, so it wouldn't be that hard...

But as I get the rule, you are particularly referring to engines, not original author, in this regards SF is based on Glaurang and it is not an original work itself.

Why would that be? Tord is one of the authors and was the sole author of Glaurung. I don't follow this reasoning. Crafty followed Cray Blitz. Is it not unique and legal? The "name" is irrelevant. The code base is the issue.

[Cubeman]

3rd dissconnect = forfeit.No ability for the players to carry on.FIDE have similar rule for cell phone during game, except 1 time is limit, and players can't decide themselves too carry on regardless.This way no-one can accuse anyone of bad sportmanship and people just will have to move on and get over it.

[Look]

Clone/Derivatives Rules

1. Each participant (engine) must an original work. No entry can contain code from another program, or be a "clone" of another program. This includes any "personality" settings of an originating program. This includes opening books.

Hi,

Does it include StockFish too?

I believe that Tord is still a contributing author of Stockfish, so I see no reason why it wouldn't be allowed to enter.

If someone wanted to challenge the code; well it is available for download, so it wouldn't be that hard...

But as I get the rule, you are particularly referring to engines, not original author, in this regards SF is based on Glaurang and it is not an original work itself.

Why would that be? Tord is one of the authors and was the sole author of Glaurung. I don't follow this reasoning. Crafty followed Cray Blitz. Is it not unique and legal? The "name" is irrelevant. The code base is the issue.

But I am saying something else. AFAIK GPL rights in fact belongs to Free Software Foundation not original author specifically. Now if tournament directors want to have specific rules about their event, that's their decision. To make myself clear, I specifically have a suggestion for GPL code, IMO it is better to be able to use one derivative of each GPL program for each tournament, and decision would be made by original authors of the GPL engine which derivative that would be. Otherwise, I assume it (though maybe unintentionally) would be encouraging closed code and possibility of breaking GPL license. Specially because you consider only _current_ version of an engine on the basis of clone suspicion. Let me ask a more specific question on this, if someone gets source of GPL code , then the person make drastic changes to code so that there could not be a well founded claim that it is based on GPL code, this could be allowed in if I understand you correctly. Note that I am not accusing anyone of doing this.

But overall conclusion for me is, CC guys are (knowingly or not) discouraging free and open source code while allowing "alternative" methods. Maybe I'd better spend my time on other areas instead of CC.

[michiguel]

Clone/Derivatives Rules

1. Each participant (engine) must an original work. No entry can contain code from another program, or be a "clone" of another program. This includes any "personality" settings of an originating program. This includes opening books.

Hi,

Does it include StockFish too?

I believe that Tord is still a contributing author of Stockfish, so I see no reason why it wouldn't be allowed to enter.

If someone wanted to challenge the code; well it is available for download, so it wouldn't be that hard...

But as I get the rule, you are particularly referring to engines, not original author, in this regards SF is based on Glaurang and it is not an original work itself.

Why would that be? Tord is one of the authors and was the sole author of Glaurung. I don't follow this reasoning. Crafty followed Cray Blitz. Is it not unique and legal? The "name" is irrelevant. The code base is the issue.

But I am saying something else. AFAIK GPL rights in fact belongs to Free Software Foundation not original author specifically. Now if tournament directors want to have specific rules about their event, that's their decision. To make myself clear, I specifically have a suggestion for GPL code, IMO it is better to be able to use one derivative of each GPL program for each tournament, and decision would be made by original authors of the GPL engine which derivative that would be. Otherwise, I assume it (though maybe unintentionally) would be encouraging closed code and possibility of breaking GPL license. Specially because you consider only _current_ version of an engine on the basis of clone suspicion. Let me ask a more specific question on this, if someone gets source of GPL code , then the person make drastic changes to code so that there could not be a well founded claim that it is based on GPL code, this could be allowed in if I understand you correctly. Note that I am not accusing anyone of doing this.

But overall conclusion for me is, CC guys are (knowingly or not) discouraging free and open source code while allowing "alternative" methods. Maybe I'd better spend my time on other areas instead of CC.

GPL has nothing to do with this. GPL deals with how you distribute the program and the source code. CC rules do not discourage distributing the source of your program. What do you mean?

The rules are clear when you combine it with the other rule that says that there should be no more than one entry per author.

SF has three authors (AFAIK) and it is not derived from any other engine from any other author.

Miguel

[michiguel]

I second Bob's suggestion, but I would not call it draconian if N is reasonable. If is unfortunate, but the connection is part of the "chess entity" that is participating.

In fact, it follows the concept that the operator should have the minimum interference in the game. If the operator "decides" not to claim the win, it is in fact a huge interference, not only in that particular game, but in the whole tournament. It could easily benefit or harm a third party.

Miguel

I have changed the rule to:

Disconnection/Forfeit Rules

1. In the event of a disconnection, the party will be given 10 minutes to return to complete the game; and no more than 2 disconnections per game will be allowed. On the third time, the game will be a forfeit. This is absolute. Under no circumstances will the game continue.

Peter

Excellent.

Miguel