Here's an attempt at a defence by Nelson Hernandez:Sean Evans wrote:I took a look at the Rybka forum to see if this topic was allowed to be posted and it was there. I was fascinated by the fan-boys trying to defend Rybka as an original program
http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=21059
Cordially,
Sean
In other words: R1 beta was a rip-off but the genius of Vas has since added several hundred ELO points, so it's all OK. Doesn't this contain a rather large and unsubstantiated assumption: that the ELO advancement since R1 beta has been achieved by honest means?The contention here is Rajlich violated the terms of the GPL, or in my real estate analogy he was a squatter. My sense, based on the conviction of accusers, is that he probably did what was alleged. However I also perceive that this event took place over five years ago, which is an eternity in computer chess, and on top of that the passage of time in computer chess is measured not so much in years as in ELO points. If something can be described as "hundreds of ELO points ago" that's a very long time ago. It's practically ancient history. And if Rybka 1.0 Beta was indeed a Fruit derivative, that does not have much bearing at all on the current commercial version of Rybka, which undoubtedly has been overhauled from end to end several times over.
If it is correct that pre-Fruit Rybkas contained stolen Crafty code, then Vas has demonstrated a clear propensity to write his engines by stealing from whatever strong open-source engines were available at the time. It follows that a comparison of later Rybka versions with whatever strong open-source engines were available at the time those engines were written may produce interesting results. Has anyone tried decompiling R3 and R4 and cross-checking against Stockfish?
Of course, other strong open-source engines were available when R4 was written. We know that IPPOLIT and Robbolito were not identical to Rybka 3 and, according to "Head Comrade Yakov," contained "improveds" which might usefully have been added to Rybka. Wouldn't it be ironic if R4 contained plagiarised IPPOLIT code?