Is this statement correct?

[tano-urayoan]

* there are still a number of computer chess experts that claim Strelka is not a clone


This is not correct.

Vas

This was posted here: http://rybkaforum.net/cgi-bin/rybkaforu ... =#pid35078

My question is Mr Rajlich statement true? Are computers experts that claim that Strelka is not a clone?[/url]

[Dann Corbit]

* there are still a number of computer chess experts that claim Strelka is not a clone


This is not correct.

Vas

This was posted here: http://rybkaforum.net/cgi-bin/rybkaforu ... =#pid35078

My question is Mr Rajlich statement true? Are computers experts that claim that Strelka is not a clone?[/url]

I am not sure if wrongdoing has been committed or not.
I am not sure if the Strelka author has confessed to wrongdoing.
If the Rybka author has examined the code and knows that something is used inappropriately, then I am not aware of it.

Now, it is illegal to use GPL code in a new project that is not GPL. Perhaps this has been done. Fabian was going to look at it, but I have not heard anything from him about it, so I am not totally clear about it.

Reverse engineering a binary is not illegal, provided certain rules are followed. I do not think we can seriously consider the supposed claims that the Strelka author confessed to reverse engineering Rybka because in the same message he claimed Rybka was a clone. Can we trust one statement? Neither? Both? I do not think there is enough evidence to come to any clear conclusion and I do not know if it was really the Strelka author who made such claims.

I think that somehow the Strelka author has copied facets of the Rybka evaluation or search. Whether this was illegal or not I do not know (though it seems *morally* painful in any case). The unusual similarity of Rybka 1.0 and Strelka evaluating certain positions indicates this in a way that seems extremely unlikely.

I guess that Vas knows more than he is saying but the information that he has on the subject is not as clear to me as it is to him.

For my part, I will not say that Strelka is not a clone or Strelka is a clone. I think that there are many open questions and the three people most qualified to answer them (Fabien, Yusipov, Vas) have not said very much.

These are the things that are clear to me:
1. Strelka uses Fruit ideas (there is no doubt about it). Whether this came from reading the source and writing his own version or by cloning I am not able to say.
2. Strelka uses large tables for evaluation. Whether or not these came from Rybka I do not know for sure. Definitely Vas could tell us. If they did come from Rybka, I am not sure if there is a violation of the law, but something is definitely amiss. I can say for sure that I do not like it because in his release notes he never mentioned anything about it. To me, that is intellectual dishonesty at the least (akin to plagiarism).
3. In the actual implementation, Strelka is not much like Fruit. The fundamental data structures are different along with many other things that are simply done another way. If the similarities warrant Streka becoming also GPL, I think Fabian is most qualified to say. I gave the source to Fabian (with permission from Yuri Yusipov) quite a while ago, but I have not heard anything certain from him about it.

I think it is clear that there are some things about Strelka which are disturbing. I do not think it is clear that copyright laws or GPL status have (or have not) been violated.