swami wrote:
I don't think Toga developers only altered a few lines of code, I'd think they added more lines to the code. I recall someone (Uri? Dann?) saying Toga had had over 100 lines of code added to Fruit 2.1 and more what's more.. the alteration had been done to the existing code.
It'd be good If Toga is still being developed and imrpoved upon further.
Hi Swami,
how can adding more lines to the code not be interpreted as adding any further ideas as suggested by some?
I think that the anti-Toga argument is basically an anti-open source argument. Either you agree with open source or you don't, and those who are anti seem to have the most to say.
I'm not getting at you here.
Cheers, Graham.
I'm obviously not "anti-open-source" yet I am against more than one copy of a program entering. Tracy has recently modified over a hundred lines of Crafty code. Ted just sent me some changes to test. I have modified thousands of lines of code. Can we / should we enter 4 different versions?
That's the problem. If you start this, there is no ending it. We now have three versions of Glaurung 2 including the original by Tord. We have versions of fruit. We have had many versions of Crafty in the past. And I have consistently been against more than one entry from the same author, whether the final modifications are from that author or not. It just doesn't make sense for tournaments.
swami wrote:
I don't think Toga developers only altered a few lines of code, I'd think they added more lines to the code. I recall someone (Uri? Dann?) saying Toga had had over 100 lines of code added to Fruit 2.1 and more what's more.. the alteration had been done to the existing code.
It'd be good If Toga is still being developed and imrpoved upon further.
Hi Swami,
how can adding more lines to the code not be interpreted as adding any further ideas as suggested by some?
I think that the anti-Toga argument is basically an anti-open source argument. Either you agree with open source or you don't, and those who are anti seem to have the most to say.
I'm not getting at you here.
Cheers, Graham.
I'm obviously not "anti-open-source" yet I am against more than one copy of a program entering. Tracy has recently modified over a hundred lines of Crafty code. Ted just sent me some changes to test. I have modified thousands of lines of code. Can we / should we enter 4 different versions?
That's the problem. If you start this, there is no ending it. We now have three versions of Glaurung 2 including the original by Tord. We have versions of fruit. We have had many versions of Crafty in the past. And I have consistently been against more than one entry from the same author, whether the final modifications are from that author or not. It just doesn't make sense for tournaments.
I have received 215 betas from Mark since H12 - maybe we should enter all of these also,
CRoberson wrote:
As far as tournaments go, all of the rules prohibited clones prior to
the release of Fruit much less Toga. The only reason I could see the
ICGA allowing cluster Toga in was to allow a scientific experiment on
the clustering of a chess program.
This is completely false! Clones have always been allowed, as they should be. Cluster Toga was allowed because there was permission from all of its authors, including Fabien.
Requiring permissions from the author(s) and not allowing one author in multiple teams solves the problem completely.
Yes, that is exactly what I proposed. Very simple.
Every entry should list the name of people that contribute it with chess playing code (authors) and their permission.
Miguel
PS: I also believe that this should apply to Nalimov EGTBs. The EGTB code in most engines was cut an pasted from him. is that correct? I did not get an answer to this question yet.
If the original author allows a clone to enter, what possible objection could you have to a clone entering? How is this different from a program which is a team effort?
If there are many clones of a single program, the decision might be difficult for the original author and he might have to disappoint many people. Well, though luck! Go write your own program, lazy bastards
The original code was developed inside Crafty. Once we decided on how to compress, the compression blocksize, etc, Eugene started to let others use the code, which was always his original intent. He needed a guinea pig and I volunteered to help. But it was never "just for Crafty" even though he contributed several bits and pieces of assembly code before I started doing it as inline gcc rather than as separate asm files.
swami wrote:
I don't think Toga developers only altered a few lines of code, I'd think they added more lines to the code. I recall someone (Uri? Dann?) saying Toga had had over 100 lines of code added to Fruit 2.1 and more what's more.. the alteration had been done to the existing code.
It'd be good If Toga is still being developed and imrpoved upon further.
Hi Swami,
how can adding more lines to the code not be interpreted as adding any further ideas as suggested by some?
I think that the anti-Toga argument is basically an anti-open source argument. Either you agree with open source or you don't, and those who are anti seem to have the most to say.
I'm not getting at you here.
Cheers, Graham.
I'm obviously not "anti-open-source" yet I am against more than one copy of a program entering. Tracy has recently modified over a hundred lines of Crafty code. Ted just sent me some changes to test. I have modified thousands of lines of code. Can we / should we enter 4 different versions?
That's the problem. If you start this, there is no ending it. We now have three versions of Glaurung 2 including the original by Tord. We have versions of fruit. We have had many versions of Crafty in the past. And I have consistently been against more than one entry from the same author, whether the final modifications are from that author or not. It just doesn't make sense for tournaments.
I have received 215 betas from Mark since H12 - maybe we should enter all of these also,
That's what I am thinking. We could (between the two of us) come up with at least 512 entrants and make this the biggest CCT ever.
CRoberson wrote:
As far as tournaments go, all of the rules prohibited clones prior to
the release of Fruit much less Toga. The only reason I could see the
ICGA allowing cluster Toga in was to allow a scientific experiment on
the clustering of a chess program.
This is completely false! Clones have always been allowed, as they should be. Cluster Toga was allowed because there was permission from all of its authors, including Fabien.
Requiring permissions from the author(s) and not allowing one author in multiple teams solves the problem completely.
Yes, that is exactly what I proposed. Very simple.
Every entry should list the name of people that contribute it with chess playing code (authors) and their permission.
Miguel
PS: I also believe that this should apply to Nalimov EGTBs. The EGTB code in most engines was cut an pasted from him. is that correct? I did not get an answer to this question yet.
If the original author allows a clone to enter, what possible objection could you have to a clone entering? How is this different from a program which is a team effort?
If there are many clones of a single program, the decision might be difficult for the original author and he might have to disappoint many people. Well, though luck! Go write your own program, lazy bastards
The original code was developed inside Crafty. Once we decided on how to compress, the compression blocksize, etc, Eugene started to let others use the code, which was always his original intent. He needed a guinea pig and I volunteered to help. But it was never "just for Crafty" even though he contributed several bits and pieces of assembly code before I started doing it as inline gcc rather than as separate asm files.
So, in other engines, most of the probing code will be a cut and paste from the one you describe. Is that correct?
bob wrote:
I'm obviously not "anti-open-source" yet I am against more than one copy of a program entering. Tracy has recently modified over a hundred lines of Crafty code. Ted just sent me some changes to test. I have modified thousands of lines of code. Can we / should we enter 4 different versions?
That's the problem. If you start this, there is no ending it. We now have three versions of Glaurung 2 including the original by Tord. We have versions of fruit. We have had many versions of Crafty in the past. And I have consistently been against more than one entry from the same author, whether the final modifications are from that author or not. It just doesn't make sense for tournaments.
For what it's worth, I agree that this policy is correct for the big tournaments.
CRoberson wrote:
As far as tournaments go, all of the rules prohibited clones prior to
the release of Fruit much less Toga. The only reason I could see the
ICGA allowing cluster Toga in was to allow a scientific experiment on
the clustering of a chess program.
This is completely false! Clones have always been allowed, as they should be. Cluster Toga was allowed because there was permission from all of its authors, including Fabien.
Requiring permissions from the author(s) and not allowing one author in multiple teams solves the problem completely.
Yes, that is exactly what I proposed. Very simple.
Every entry should list the name of people that contribute it with chess playing code (authors) and their permission.
Miguel
PS: I also believe that this should apply to Nalimov EGTBs. The EGTB code in most engines was cut an pasted from him. is that correct? I did not get an answer to this question yet.
If the original author allows a clone to enter, what possible objection could you have to a clone entering? How is this different from a program which is a team effort?
If there are many clones of a single program, the decision might be difficult for the original author and he might have to disappoint many people. Well, though luck! Go write your own program, lazy bastards
The original code was developed inside Crafty. Once we decided on how to compress, the compression blocksize, etc, Eugene started to let others use the code, which was always his original intent. He needed a guinea pig and I volunteered to help. But it was never "just for Crafty" even though he contributed several bits and pieces of assembly code before I started doing it as inline gcc rather than as separate asm files.
So, in other engines, most of the probing code will be a cut and paste from the one you describe. Is that correct?
Miguel
I don't even think it is cut and paste. All the work is in egtb.cpp. One has to write (rewrite) probe.c from Crafty to take the other program's board structure and convert it to when egtb.cpp needs to compute the Godel number and do the probe.
It would really be a shame if Toga Cluster could not participate as it is an extremely interesting project. But I can understand why people would think the participation of both Fruit and Toga would be a problem. Imagine Rybka 3 Default, Rybka 3 Human and Rybka 3 Dynamic all participating at the same time. That would cause an uproar I think, as the most likely outcome would be first 3 places for Rybka 3's.
Hopefully the Fruit and Toga team will get together and come out with a strong combined effort to participate the best of both worlds.
CRoberson wrote:
As far as tournaments go, all of the rules prohibited clones prior to
the release of Fruit much less Toga. The only reason I could see the
ICGA allowing cluster Toga in was to allow a scientific experiment on
the clustering of a chess program.
This is completely false! Clones have always been allowed, as they should be. Cluster Toga was allowed because there was permission from all of its authors, including Fabien.
Requiring permissions from the author(s) and not allowing one author in multiple teams solves the problem completely.
Yes, that is exactly what I proposed. Very simple.
Every entry should list the name of people that contribute it with chess playing code (authors) and their permission.
Miguel
PS: I also believe that this should apply to Nalimov EGTBs. The EGTB code in most engines was cut an pasted from him. is that correct? I did not get an answer to this question yet.
If the original author allows a clone to enter, what possible objection could you have to a clone entering? How is this different from a program which is a team effort?
If there are many clones of a single program, the decision might be difficult for the original author and he might have to disappoint many people. Well, though luck! Go write your own program, lazy bastards
The original code was developed inside Crafty. Once we decided on how to compress, the compression blocksize, etc, Eugene started to let others use the code, which was always his original intent. He needed a guinea pig and I volunteered to help. But it was never "just for Crafty" even though he contributed several bits and pieces of assembly code before I started doing it as inline gcc rather than as separate asm files.
So, in other engines, most of the probing code will be a cut and paste from the one you describe. Is that correct?
Miguel
I don't even think it is cut and paste. All the work is in egtb.cpp. One has to write (rewrite) probe.c from Crafty to take the other program's board structure and convert it to when egtb.cpp needs to compute the Godel number and do the probe.
So, every program that supports Nalimov EGTBs has the module egtb.cpp exactly as it is in crafty. Right?
Tord Romstad wrote: When my program ships with a license that says that you are free to do whatever you like with it (apart from making it proprietary), what right do I have to tell you later that you are not allowed to use it in some tournament?
The license governs copyright and distribution.
And also the terms of use, which is the most important thing in this case. The license permits users to use the program to do anything they like with the program. It does not say "anything except playing in a tournament".
A third party (tournament organizer) can put up the (additional) rules he wants and no license you write can change that.
Yes, they can -- for instance by requiring that only the author operates the engine. Such a rule would make a lot of sense.
However, in the CCT, there is no such rule. Operators are allowed, as long as the author(s) accept it. In the case of free programs like mine, this permission is provided by the license. Therefore, anyone can enter with Glaurung -- or a program based on Glaurung -- in a tournament like the CCT (as has already happend this year, as you can see on the participant list). Of course only one version can be allowed to participate, and that will probably have to be the first version that registers.
