What is the definition of a clone?

[Guenther Simon]

...

Practically there are many competitions that allow clones

ssdf list ccrl list cegt list give rating to different versions of the same program.

...
Uri

What do you mean with the above sentences?
If it means what I suppose it means, it has nothing to do with cloning,
except e.g. you call Shredder 10 a clone of Shredder 9.
This makes no sense for me for the clone discussion.
AFAIK SSDF never had clones in their list.

Guenther

[Uri Blass]

...

Practically there are many competitions that allow clones

ssdf list ccrl list cegt list give rating to different versions of the same program.

...
Uri

What do you mean with the above sentences?
If it means what I suppose it means, it has nothing to do with cloning,
except e.g. you call Shredder 10 a clone of Shredder 9.
This makes no sense for me for the clone discussion.
AFAIK SSDF never had clones in their list.

Guenther

If Toga is considered to be fruit clone then I see no reason why shredder10 should not be considered as shredder9 clone.

The rules do not allow both
shredder9 and shredder10 in WCCC and they also do not allow both fruit and toga in WCCC.

Only one of them is not a problem(assuming that the authors agree and in case of shredder there is only one author).

Toga does not participate in SSDF but I think that the ssdf will also have no problem to test it assuming that Fabien is going to be interested in it.

Uri

[Roman Hartmann]

If Toga is considered to be fruit clone then I see no reason why shredder10 should not be considered as shredder9 clone.

The rules do not allow both
shredder9 and shredder10 in WCCC and they also do not allow both fruit and toga in WCCC.

Only one of them is not a problem(assuming that the authors agree and in case of shredder there is only one author).

Toga does not participate in SSDF but I think that the ssdf will also have no problem to test it assuming that Fabien is going to be interested in it.

Uri

What do you mean with 'if Toga is considered to be a Fruit clone'?
Toga _is_ a (legal) Fruit clone and that's easy to see as the sources for both Fruit and Toga are released under the GPL. Everybody can see that Toga is a derivative work of Fruit.

But for Shredder we don't have any code and we don't know if version 9 shares any code with version 10. Maybe Shredder 10 was complete rewrite.
We just don't know the details and it hardly matters in the cloning context as the author of version 9 and version 10 of Shredder is the same anyway. I strongly doubt that SMK will try to compete with version 9 and 10 in an event at the same time so there shouldn't be any issue with having several versions of Shredder out there.

Roman

[Uri Blass]

If Toga is considered to be fruit clone then I see no reason why shredder10 should not be considered as shredder9 clone.

The rules do not allow both
shredder9 and shredder10 in WCCC and they also do not allow both fruit and toga in WCCC.

Only one of them is not a problem(assuming that the authors agree and in case of shredder there is only one author).

Toga does not participate in SSDF but I think that the ssdf will also have no problem to test it assuming that Fabien is going to be interested in it.

Uri

What do you mean with 'if Toga is considered to be a Fruit clone'?
Toga _is_ a (legal) Fruit clone and that's easy to see as the sources for both Fruit and Toga are released under the GPL. Everybody can see that Toga is a derivative work of Fruit.

But for Shredder we don't have any code and we don't know if version 9 shares any code with version 10. Maybe Shredder 10 was complete rewrite.
We just don't know the details and it hardly matters in the cloning context as the author of version 9 and version 10 of Shredder is the same anyway. I strongly doubt that SMK will try to compete with version 9 and 10 in an event at the same time so there shouldn't be any issue with having several versions of Shredder out there.

Roman

Same is correct for toga.
I strongly doubt that Fabien will try to compete with Fruit and Toga in an event at the same time.

Maybe Shredder10 is a complete rewrite of shredder9 but I assume for this discussion that shredder10 is not a complete rewrite.

I see no reason for difference between the case of 2 programs of fabien(when in one of them thomas is also involved) and the case of 2 programs of stefan.

Uri

[bob]

...

Practically there are many competitions that allow clones

ssdf list ccrl list cegt list give rating to different versions of the same program.

...
Uri

What do you mean with the above sentences?
If it means what I suppose it means, it has nothing to do with cloning,
except e.g. you call Shredder 10 a clone of Shredder 9.
This makes no sense for me for the clone discussion.
AFAIK SSDF never had clones in their list.

Guenther

If Toga is considered to be fruit clone then I see no reason why shredder10 should not be considered as shredder9 clone.

The rules do not allow both
shredder9 and shredder10 in WCCC and they also do not allow both fruit and toga in WCCC.

Only one of them is not a problem(assuming that the authors agree and in case of shredder there is only one author).

Toga does not participate in SSDF but I think that the ssdf will also have no problem to test it assuming that Fabien is going to be interested in it.

Uri

There is one big reason. The author for both versions of shredder is the _same_ person. In a tournament both versions will _not_ be allowed to play. Rules state that each author can enter exactly _one_ program.

SSDF/etc have nothing to do with tournaments. Anybody can test whatever they want in their own private rating lists. But sanctioned tournaments do not allow such... for good reasons.

[bob]

except that you are mixing apples and oranges. We are talking about _tournaments_. Not private rating lists. In tournaments you will _never_ find two shredder versions playing because the rules do not allow it. Ditto for clones. If you want to call shredder 10 a clone of shredder 9, that's fine. Because both can't participate in the same WCCC or CCT type tournament anyway, so it is just a semantics point...

[Tony Thomas]

except that you are mixing apples and oranges. We are talking about _tournaments_. Not private rating lists. In tournaments you will _never_ find two shredder versions playing because the rules do not allow it. Ditto for clones. If you want to call shredder 10 a clone of shredder 9, that's fine. Because both can't participate in the same WCCC or CCT type tournament anyway, so it is just a semantics point...

In my opinion calling two different versions of the same program, written by the same programmer, is a stretch of the generally accepted definition of a clone. If that is the case, Robert Hyatt is the biggest cloner I have ever known over the net. He relesed so many versions of his crafty engine, I am not even going to start counting.

[Uri Blass]

except that you are mixing apples and oranges. We are talking about _tournaments_. Not private rating lists. In tournaments you will _never_ find two shredder versions playing because the rules do not allow it. Ditto for clones. If you want to call shredder 10 a clone of shredder 9, that's fine. Because both can't participate in the same WCCC or CCT type tournament anyway, so it is just a semantics point...

In my opinion calling two different versions of the same program, written by the same programmer, is a stretch of the generally accepted definition of a clone. If that is the case, Robert Hyatt is the biggest cloner I have ever known over the net. He relesed so many versions of his crafty engine, I am not even going to start counting.

I did not claim that cloning is a bad thing.
I see the word clone to describe something that is done and not about who do it.

Toga is defined to be a fruit clone.
This means that cloning is defined as taking source of program X and changing it.
This mean that every program is a clone of the previous version assuming that the programmer did not start from scratch in order to get the next version.

Uri

[Roger Brown]

I did not claim that cloning is a bad thing.
I see the word clone to describe something that is done and not about who do it.

Toga is defined to be a fruit clone.
This means that cloning is defined as taking source of program X and changing it.
This mean that every program is a clone of the previous version assuming that the programmer did not start from scratch in order to get the next version.

Uri

Hello Uri,

I vowed never to enter a thread where you were one of the chief participants but are you aware that what you are writing is lacking in your trademark logic and good sense?

You are saying that the author who improves his engine version by version - by adding personalities, multi-cpu ability etc - is cloning it? That the author who is originator of the ideas contained in the source clones it whenever he uses it as a base to continue developing it?

That definition will not fly except in your version of the universe...

Time for me to go.

Later.

[YvesLejeail]

Uri seems to be right about this point; if we look at the definition ; a new version of a engine is a clone (even if it is the same author)!
It is a legal one though
Hopefully may be allowed to participate to a competition, if there are no other engine of the same family (depends of the competition as said above).
The author wish to know what clone version is the best.
Thanks to have pointed out this new fact Uri !
Yves

[Uri Blass]

I did not claim that cloning is a bad thing.
I see the word clone to describe something that is done and not about who do it.

Toga is defined to be a fruit clone.
This means that cloning is defined as taking source of program X and changing it.
This mean that every program is a clone of the previous version assuming that the programmer did not start from scratch in order to get the next version.

Uri

Hello Uri,

I vowed never to enter a thread where you were one of the chief participants but are you aware that what you are writing is lacking in your trademark logic and good sense?

You are saying that the author who improves his engine version by version - by adding personalities, multi-cpu ability etc - is cloning it? That the author who is originator of the ideas contained in the source clones it whenever he uses it as a base to continue developing it?

That definition will not fly except in your version of the universe...

Time for me to go.

Later.

I did not claim that cloning is always a bad thing.

My point is that if you decide that Y is a clone of X then it is a property of X and Y and not a property that is dependent on the people who created X and Y.

Fabien created fruit and thomas changed it to toga.
Fabien is the author and I think that fabien is allowed to use toga in competitions in case that he is interested to do it.

It means that if toga does not compete the problem is not with the fact that toga is a clone but with the fact that the programmer(fabien) does not want it to compete.

If programs participate without the agreement of the author or if
more than one program of the same programmer participate in the same competition then it can create problems in part of the cases(like in WCCC) but not in another part of the cases like CEGT or CCRL competitions.

Uri

[Roman Hartmann]

I guess the meaning of the word clone related to chess programming needs to be defined somehow otherwise people talk of clones meaning different things.

My understanding of a clone in the chess programming context is that a clone is a derivative work based on the sources of an existing chess engine. That means that the code of another chess engine is taken by someone else than the original author, modified and that the program is renamed as well.

The world clone itself does not contain any information regarding its legal status.

If the original author modifies his own code and creates a new version the word cloning is -while technically maybe not completely wrong- imo at least inappropriate.

Roman

[YvesLejeail]

If the original author modifies his own code and creates a new version the word cloning is -while technically maybe not completely wrong- imo at least inappropriate.
Roman

Why ?
It seems inappropriate if you have in mind the association "clone=bad thing".
But from the technical definition of cloning, it looks fine (not completely wrong as you said). The fact is that when we are talking about cloning in chess programming, we are thinking about illegal cloning implicitly. We should thus always add a "legal" or "illegal " adjective to the word clone, to clarify some things.
Though there are some remaining questions even for legal clones as we have discussed.
Yves

[bob]

except that you are mixing apples and oranges. We are talking about _tournaments_. Not private rating lists. In tournaments you will _never_ find two shredder versions playing because the rules do not allow it. Ditto for clones. If you want to call shredder 10 a clone of shredder 9, that's fine. Because both can't participate in the same WCCC or CCT type tournament anyway, so it is just a semantics point...

In my opinion calling two different versions of the same program, written by the same programmer, is a stretch of the generally accepted definition of a clone. If that is the case, Robert Hyatt is the biggest cloner I have ever known over the net. He relesed so many versions of his crafty engine, I am not even going to start counting.

I completely agree.

We know what "clone" means, it doesn't take an obtuse definition to confuse things.

[bob]

except that you are mixing apples and oranges. We are talking about _tournaments_. Not private rating lists. In tournaments you will _never_ find two shredder versions playing because the rules do not allow it. Ditto for clones. If you want to call shredder 10 a clone of shredder 9, that's fine. Because both can't participate in the same WCCC or CCT type tournament anyway, so it is just a semantics point...

In my opinion calling two different versions of the same program, written by the same programmer, is a stretch of the generally accepted definition of a clone. If that is the case, Robert Hyatt is the biggest cloner I have ever known over the net. He relesed so many versions of his crafty engine, I am not even going to start counting.

I did not claim that cloning is a bad thing.
I see the word clone to describe something that is done and not about who do it.

Toga is defined to be a fruit clone.
This means that cloning is defined as taking source of program X and changing it.
This mean that every program is a clone of the previous version assuming that the programmer did not start from scratch in order to get the next version.

Uri

Now you are pulling a Vincent. "clone" has _always_ meant (within the domain of computer chess) a program derived from another player, but which is claimed to be an original work of a different author. Le Petite. Voyager. Gunda-1. To name just a few "crafty clones". The same author can't produce a clone of his own program by any definition we have been using....

[Wander Ferreira Jr]

A new version of a program can never be a clone of a previous version of it.
It is like me being a clone of myself at age 10!

The engine just "got old", has gained more "experience", more code enhancements... It's author learned from his errors, and made some changes in the program.

He did not created a new engine to compete with the other. The versions just coexist for testing (and for historical reasons), but, as far as I know, programmers wish to make just one version, that will be in the future the new engine.

[Uri Blass]

A new version of a program can never be a clone of a previous version of it.
It is like me being a clone of myself at age 10!

The engine just "got old", has gained more "experience", more code enhancements... It's author learned from his errors, and made some changes in the program.

He did not created a new engine to compete with the other. The versions just coexist for testing (and for historical reasons), but, as far as I know, programmers wish to make just one version, that will be in the future the new engine.

My simple logic tell me that a new version of a program is always a clone of previous versions unless the programmer started from scratch.

Here is the logic that I use to get this conclusion:

1)The claim A is a prime factor of B is only a property of A and B and not of something else.
The claim that A is the mother of B is only a property of A and B and not of something else.

2)Generalization of the same logic
The claim A is *** of B should be only property of A and B and not of other things regardless of ***

3)Private case of this generalization.
The claim that source A is a clone of source B is only property of A and B
and is not dependent on the question if the author of A and B is the same or not the same.

4)If programmer X started from program A (generated by programmer Y) and modified it then B is considered to be a clone of A(see fruit-toga case)

I already explained that the term clone between A and B cannot be dependent on X and Y but only on A and B so even if X and Y are the same programmer then based on simple logic B is considered to be a clone of A.

Uri

[Volker Pittlik]

...

1)The claim A is a prime factor of B is only a property of A and B and not of something else.

...

2)Generalization of the same logic
The claim A is *** of B should be only property of A and B and not of other things regardless of ***

3)Private case of this generalization.

...
4)If programmer X started from program A (generated by programmer Y) and modified it then B is considered to be a clone of A(see fruit-toga case)...

... "clone" has _always_ meant (within the domain of computer chess) a program derived from another player, but which is claimed to be an original work of a different author...

I tend to apply Occam's Razor here.

Volker