Common sense returns to CCRL

Discussion of anything and everything relating to chess playing software and machines.

Moderators: Harvey Williamson, bob, hgm

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Ant_Gugdin
Posts: 60
Joined: Sun Dec 26, 2010 8:13 pm

Re: Common sense returns to CCRL

Post by Ant_Gugdin » Mon Feb 21, 2011 2:22 pm

Kaj Soderberg wrote:
Frank Quisinsky wrote:Hi Gerold,

a rating list have to test engines.
Main order for an engine is to play chess and not which political reasons are in background.

Best
Frank
Possible theft is something different than "political reasons".
For the rest i agree with the purpose of a rating list.

Best regards,
Kaj
I think the issue most people had with CCRL's approach was the lack of consistency in the treatment of Rybka and Fruit.

Personally, I don't think this is such a great outcome. If I were an engine tester, I would include both Rybka and Houdini but I would asterisk both of them, detailing the claims which have been made about each engine.

Kaj Soderberg
Posts: 137
Joined: Sat Jan 01, 2011 6:33 pm

Re: Common sense returns to CCRL

Post by Kaj Soderberg » Mon Feb 21, 2011 2:29 pm

Ant_Gugdin wrote:
Kaj Soderberg wrote:
Frank Quisinsky wrote:Hi Gerold,

a rating list have to test engines.
Main order for an engine is to play chess and not which political reasons are in background.

Best
Frank
Possible theft is something different than "political reasons".
For the rest i agree with the purpose of a rating list.

Best regards,
Kaj
I think the issue most people had with CCRL's approach was the lack of consistency in the treatment of Rybka and Fruit.

Personally, I don't think this is such a great outcome. If I were an engine tester, I would include both Rybka and Houdini but I would asterisk both of them, detailing the claims which have been made about each engine.
As such not a bad idea, giving some the benefit of the doubt.
The risk is though, that there will be lots of (maybe classes of) asterisks in a while. Who will be the upper priest of asterisks? The intention sounds good, but execution could be tricky.

bob
Posts: 20918
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: And where do we draw the line?

Post by bob » Mon Feb 21, 2011 3:09 pm

Tom Barrister wrote:How do you prove that a closed-source engine does or does not violate the rules, unless the author releases the source?

Also, how can you prove an open-source engine does or does not violate the rules by being a clone/whatever of a closed-source engine, unless the author of the closed-source engine reveals the source?

The argument is (apparently) a hopeless one. Somehow stealing someone's source is not the same as stealing their money or their property. Electronic data is sort of "vaporware" and doesn't exist in a physical sense, so it is ok to take what you want and do what you want with it...

rbarreira
Posts: 900
Joined: Tue Apr 27, 2010 1:48 pm

Re: And where do we draw the line?

Post by rbarreira » Mon Feb 21, 2011 3:30 pm

Tom Barrister wrote:How do you prove that a closed-source engine does or does not violate the rules, unless the author releases the source?

Also, how can you prove an open-source engine does or does not violate the rules by being a clone/whatever of a closed-source engine, unless the author of the closed-source engine reveals the source?
There are a few ways to prove an engine is a clone of another one without looking at the code. If PVs match exactly for several depths in test positions, it's extremely likely that both engines share code.

It is harder to prove the opposite, of course...

Post Reply