Hearing in the SF vs chessbase case begins tomorrow, and let us hope SF wins. My question is how does one prove that there has been license violation? Anybody has more details?
I thought it was about match between them!
Well, what's the story about case?
Well, what's the story about case?
stockfishchess.org/blog/2021/our-lawsuit-against-chessbase/
The hearing begins next Monday.
stockfishchess.org/blog/2022/public-court-hearing-soon/
The hearing begins next Monday.
stockfishchess.org/blog/2022/public-court-hearing-soon/
This is like stealing melodies from songs. Stairway to Heaven case - www.bbc.com/news/entertainment-arts-54423922
They will have to show substantial coding has been used unchanged. Couldn't someone, like that Lucas chess program, create puzzles and different ways to use Stockfish and sell that under the premise it is a new creation? How much of a new creation are chessbase products? Do they offer new features that are significantly unique?
They will have to show substantial coding has been used unchanged. Couldn't someone, like that Lucas chess program, create puzzles and different ways to use Stockfish and sell that under the premise it is a new creation? How much of a new creation are chessbase products? Do they offer new features that are significantly unique?
Creating puzzles using SF or some other open source tool is different from using SF code and selling it under different name, with very minimal changes, without publishing the source code. From what I have read, chessbase only changed some neural network parameters and sold it as fatfritz, without making the source code and the changes available under the same license. But how would they prove and how did they find out? Would the court appoint experts to whom chessbase will have to submit the code and the experts will compare?
@WorldRenownPatzer said in #4:
> This is like stealing melodies from songs. Stairway to Heaven case - www.bbc.com/news/entertainment-arts-54423922
>
> They will have to show substantial coding has been used unchanged. Couldn't someone, like that Lucas chess program, create puzzles and different ways to use Stockfish and sell that under the premise it is a new creation? How much of a new creation are chessbase products? Do they offer new features that are significantly unique?
An engine like Stockfish is extremely hard to change without a lot of time and effort (how do you beat the top engine in the world?) so my guess is they copy/pasted the code or just used the executable.
Edit : My bad I read too fast, indeed you make a point.
> This is like stealing melodies from songs. Stairway to Heaven case - www.bbc.com/news/entertainment-arts-54423922
>
> They will have to show substantial coding has been used unchanged. Couldn't someone, like that Lucas chess program, create puzzles and different ways to use Stockfish and sell that under the premise it is a new creation? How much of a new creation are chessbase products? Do they offer new features that are significantly unique?
An engine like Stockfish is extremely hard to change without a lot of time and effort (how do you beat the top engine in the world?) so my guess is they copy/pasted the code or just used the executable.
Edit : My bad I read too fast, indeed you make a point.
@kajalmaya said in #5:
> But how would they prove and how did they find out? Would the court appoint experts to whom chessbase will have to submit the code and the experts will compare?
This tricky in some cases BUT if it was compiled with same compiler then the resulting binary would carry would be mostly the same with few offset that need to be observed. Changes done were really small so any binary diffing tool will tell that that files are very similar.
Had they been smart enough to use different tool chain in building would have been lot harder for this way.
Next is searching for string from binary, like log prints error messages etc.
if you really interested
edolstra.github.io/pubs/clones-msr2011-final.pdf
is one example for of available . Meat for detecting agains several GPLd products instead of just one at the time
> But how would they prove and how did they find out? Would the court appoint experts to whom chessbase will have to submit the code and the experts will compare?
This tricky in some cases BUT if it was compiled with same compiler then the resulting binary would carry would be mostly the same with few offset that need to be observed. Changes done were really small so any binary diffing tool will tell that that files are very similar.
Had they been smart enough to use different tool chain in building would have been lot harder for this way.
Next is searching for string from binary, like log prints error messages etc.
if you really interested
edolstra.github.io/pubs/clones-msr2011-final.pdf
is one example for of available . Meat for detecting agains several GPLd products instead of just one at the time
@petri999
Thanks for the reference. I will look at the paper. Good to know the techniques both from software tech perspective and from the legal perspective.
Thanks for the reference. I will look at the paper. Good to know the techniques both from software tech perspective and from the legal perspective.
@kajalmaya said in #1:
> Hearing in the SF vs chessbase case begins tomorrow, and let us hope SF wins. My question is how does one prove that there has been license violation? Anybody has more details?
Similarities in coding styles/testimonies I guess
> Hearing in the SF vs chessbase case begins tomorrow, and let us hope SF wins. My question is how does one prove that there has been license violation? Anybody has more details?
Similarities in coding styles/testimonies I guess
This topic has been archived and can no longer be replied to.