Rolf wrote:bob wrote:
Your last two paragraphs are a crock. It is ok to copy 99% of a program, and modify 1%, and if that makes it _much_ stronger then plagiarism laws are suspended? For a 30,000 line chess program, one needs to add 300 lines and then they can declare this program to be their own? Can we apply the same standard to books? My absolute favorite author is Matthew Reilly. Can I take one of his books, such as "Area 7" and then change say 300 words assuming his book has 30,000 words, and then call that my own work? This is nonsense.
If you copy a program, and claim the result as your own work, you are guilty of plagiarism. Plain and simple.
Excuse me, Bob, I didnt say that for the case if someone only added 1% in length, I said, that if someone could get a program 20% better than the original (with copied 99%) THEN IMO that guy has attained a class of its own with his (theoretical) 1% addition. It was more a _question_ by me as you know, so I ask again, is such a thing theoretically do-able? Or is it impossible from the design of such a multi-lines-code. But guess it were possible, would it then be forbidden to do, if the "new" thing is 20% better, the more so if it is better than the existing top leader?
Couldnt you please state such a thing as follows, that e.g. TOGA is a clone of FRUIT and with 20% new additions so that it's accepted and tested, or is TOGA something else. Could you tell this for the more uninformed lays? Also, why TOGA isnt as strong as RYBKA? Thanks so much. But without such infos the lay cant understand why TOGA isnt participating at Championships. Rolf
There are programs designed to be used as a programming base. They are intended to be cloned. In such a situation, as long as proper credit is given, it is not wrong to make these clones. For instance, Fruit has spawned an army of clones such as the Toga crew. Now, as to whether these clone programs can compete in tournaments or be analyzed by rating systems is another matter. If (for instance) I run a contest and decide that only one engine from a particular base (e.g. Fruit) can compete, then that would be up to me as the tournament organizer.
As to wheter it is good or bad to take a strong engine and fiddle with it is another matter which I am not fit to decide.
Personally, I like to read the source code of any engine I can get my hands on. Sometimes, they are hard for me to understand and sometimes easy. At any rate, I want to learn the ideas of the authors because they are interesting to me. In a similar way, I like to read all the chess papers I can get my hands on. Some are hard to understand and some are easy to understand. The general idea of search is a very interesting place of study for me. Because I work in the database field, search is something I ponder over all the time.
If someone makes slanderous statements (e.g. "Crafty is a clone" is pure slander) I think we should feel utter contempt towards them.
The notion of cloning something is not inherently bad. But the end product is different than a tool that was written without a single line from another program.
I also believe that reading technical papers and transplanting those ideas into chess programs is another kind of cloning. After all, we are moving good ideas that someone else made into something we want to call our own.
Is someone who creates and posts a strong open source chess program doing something wrong? Is someone who writes a paper that explains how to make a chess program stronger doing something wrong?
In both cases, I would say 'No.' but others may disagree. Personally, I think that knowlege should be public and not hidden whenever possible and also that the competition in chess programming has gone a little bit too far because many people want to hide their ideas instead of publishing them. Once again, I need to qualify that. It is *not* wrong to hide your ideas. But it is better if you do not hide them.
IMO-YMMV