Important announcement: World Computer Chess Championship –

[Ferdy]

For the 2017 World Computer Chess Championship we are changing the format, removing the entry fee, and offering an expenses budget to each of the contestants. Our aim for 2017 will be to attract the strongest participants. There will be only 4 contestants, each of whom will receive 1,000 Euro for their expenses from the ICGA.

Does this mean only 4 participants are accepted? or more can be accepted but they can only offer 1000 euros to each of 4 participants?

Regarding attracting the strongest participants, perhaps ICGA should just invite the Stockfish team.

I want people to know that although I posted the announcement at David Levy's request, I do not agree with:

a. limiting WCCC to just 4 entries (David does say WCSC is not limited)

Perhaps of the thought that only few are capable of using a system like the one used by Jonny.

b. excluding all but the top two programs from this year from entering WCCC next year.

Sort of like this year will be the candidates tournament. The top 2 will be matched next year to determine the world champion.

Both seem unfair to me. Programs change a lot from year to year so who knows which ones will be strongest next year?

Agree.

Restricting them could mean we do not even have the two most qualified programs. to me "World" should mean everyone. I do know space limits will mean we cannot host a huge number, but things like estimated elo could be used to select the top X programs that will "fit" in a playing room.

I thought of creating a qualification system to enter the candidates tournament for 2017.
1. Pick 1 from CCRL rating group
2. Pick 1 from CEGT
3. Pick 1 from Frank
4. The same for other well known rating group
5. If the player is already selected then pick next player

Probably the list could be
1. Komodo
2. Stockfish
3. Houdini
4. Shredder
5. Pick 2 from sponsors
6. Pick 2 from ICGA
7. Perhaps more depending on the budget

These engines then will be provided with some incentives, not just the normal expenses but some compensations.

Can't really believe that human tournaments are getting millions of attention and yet these hardworking programmers have nothing. Most people are now using engines to prepare tournaments.

[syzygy]

Can't really believe that human tournaments are getting millions of attention and yet these hardworking programmers have nothing. Most people are now using engines to prepare tournaments.

The ones that sell their program get paid.
The ones that do it for fun get the fun.

[Ras]

Can't really believe that human tournaments are getting millions of attention and yet these hardworking programmers have nothing.

In a market economy, the price of something is not related to its cost, but to demand and supply. It's just when the price goes below the cost, that thing will not make it if it's commercial. Many chess programmers are not paid at all because it's a passion project.

Even for human chess events, it's tough. Human masters have to capitalise on other things. Carlsen is not just smart over the board, he also used his popularity for some fashion label.

Do you remember the WCC when Carlsen lost a game to Karjakin and rushed out, violating the rules and risking a fine? It looked stupid and emotional, but media covered his behaviour and made gossip. Pure gold and actually smart even if it was dumb. Now engines don't rush out, do they? No story, no money.

[leavenfish]

and in any case, there is no 'soul' in engine vs engine games. Not really...but seeing a human player go all out for a win...saving a desperate position...stalking, then pouncing - it's things like these that make humans enjoy the game, not a bunch of 'ones and zeroe's' coming into conflict.

[hgm]

We may have had this discussion before. It is blatantly not the case that releasing code under the GPL implies permission for the author's name to be listed on the entry form of whatever engine in whatever tournament. What the GPL means is stated in the GPL itself.

You don't need legal permission to write anyones name anywhere. If someone asks me who is author of Shredder, I can write Stephan Meyer-Kahlen, without consulting Stephan in any way. And Shredder is not even open source. The GPL actually states that it is not possible to put restrictions of any kind on the software, other than that modifications must be released under GPL. Not being allowed to mention the author would be such a restriction.

(Not to mention the problem that any author's name can only be listed once in case of ICGA tournaments.)

This doesn't seem to be a problem. Eugene Nalimov, for example, used to be mentioned many times. Many engine's used his EGT probing code. This is a moot point anyway, as I don't think any of the contributors to Stockfish wants to enter an engine of his own. The real problem is exactly the opposite: no one wants to enter an engine at all.

Beside, the ICGA rule that each author can enter only once can be interpreted in a flexible way, after rounding to the nearest integer. That would make it OK for someone that enters an engine that is 100% his to still contribute 5% to each of 10 other engines. This could be supplemented with a condition that he should not have contributed more than 5% to any individual other engine.

I am pretty sure the ICGA would be willing to use this interpretation, should the need arise.

Were there anything in your argument (and a tournament organiser could put something in it by stating in the rules that the author's permission is not required for code released under GPL), then anyone could take/fork/whatever Stockfish and submit it.

It would certainly not be illegal to do that. (I think it is actually what TCEC does...) It is purely up to ICGA to decide what they will accept. They are not likely to accept a registration anyone who has not made a real contribution. If I were to enter Stockfish, they would most certainly refuse it. If a major contributor or even a non-contributor with a prominent position in the project would enter it, they would most certainly accept it. They are not seeking to exclude anyone.

I don't think there is less public interest in computer chess than in the Rybka days.

It is declining all the time. In the Rybka days there were also only 10 participants. Last year there were 9. In "Rybka days" the WCCC was held in Amsterdam, Beijing, Pamplona, Kanazawa and Tilburg. Amsterdam and Tilburg are in the Netherlands, indicative of the fact that no real sponsor could be found. Beijing and Kanazawa sponsored because of their interest in Go / Xiangqi / Shogi, and did not care much about the Chess that came with it. That only leaves Pamplona, which I think was mainly government sponsored because of some cultural festivities there that they felt also needed some scientific / technological dressing. Not really an encouraging record.

[syzygy]

We may have had this discussion before. It is blatantly not the case that releasing code under the GPL implies permission for the author's name to be listed on the entry form of whatever engine in whatever tournament. What the GPL means is stated in the GPL itself.

You don't need legal permission to write anyones name anywhere.

Oh come on. Fraud is a crime. The ICGA entry form asks you to declare that the persons whose names are listed have given permission.

If someone asks me who is author of Shredder, I can write Stephan Meyer-Kahlen, without consulting Stephan in any way. And Shredder is not even open source. The GPL actually states that it is not possible to put restrictions of any kind on the software, other than that modifications must be released under GPL. Not being allowed to mention the author would be such a restriction.

GPL is about copyright. Entering a rightfully owned program in an ICGA tournament is not an infringement of copyright. But this is separate from the act of fraud just described.

(Not to mention the problem that any author's name can only be listed once in case of ICGA tournaments.)

This doesn't seem to be a problem. Eugene Nalimov, for example, used to be mentioned many times. Many engine's used his EGT probing code. This is a moot point anyway, as I don't think any of the contributors to Stockfish wants to enter an engine of his own. The real problem is exactly the opposite: no one wants to enter an engine at all.

The Hyatt-trick, but that trick cannot be stretched forever.

It is not a moot point, as many engines exist that are very much based on SF.

[syzygy]

The ICGA entry form asks you to declare that the persons whose names are listed have given permission.

OK, this may not be accurate.

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 the details of their submission form. 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.

So the "author" of a blatant SF clone can enter SF by stating on the entry form that "his" program includes SF code.

I thought there was a rule stating any author may only be involved in one entry, but I cannot find that rule now.

edit: old threads seem to confirm that Rule 2 has always been interpreted as requiring permission. (Such an interpretation seems unlikely to hold up in court, but still.)

[Ras]

(Such an interpretation seems unlikely to hold up in court, but still.)

Under domestic authority, the ICGA can do whatever they want - unless it's about specific discrimination items mentioned by anti-discrimination laws.

They can always say "not original enough", but they can't say "too much code by developers from Christmas Islands".

[Dann Corbit]

(Such an interpretation seems unlikely to hold up in court, but still.)

Under domestic authority, the ICGA can do whatever they want - unless it's about specific discrimination items mentioned by anti-discrimination laws.

They can always say "not original enough", but they can't say "too much code by developers from Christmas Islands".

They can do whatever they want so long as the actions are contained within the club boundaries.

They cannot legally slander an entrant, for example.

They could throw him out of their contest for wearing green socks, if they liked.

[syzygy]

(Such an interpretation seems unlikely to hold up in court, but still.)

Under domestic authority, the ICGA can do whatever they want - unless it's about specific discrimination items mentioned by anti-discrimination laws.

They can always say "not original enough", but they can't say "too much code by developers from Christmas Islands".

Well, it is rather doubtful if they can interpret their own rules in a way that would lead to a breach of contract as interpreted by a court. But we're not really talking about such a situation here.