Author: Dann Corbit
Date: 19:44:51 05/14/04
Go up one level in this thread
On May 14, 2004 at 22:31:04, Russell Reagan wrote: >On May 14, 2004 at 21:27:09, Dann Corbit wrote: > >>Hence, to simply use someone else's >>chess opening book would be illegal. But to have some of the same lines is >>totally without problems. > >Where do you draw the line? I imagine that when a book maker sees a new good >line from another book in a tournament game that he adds it to his own book. On >the other end of the spectrum, someone could reverse engineer any opening book >format, or play many games and get many of the lines from a book. The only >differences I see are: > >1. intent >2. manual vs. automated milking of another book > >Manual milking of a book seems to be acceptable because it can't occur at a fast >enough rate to copy a large portion of another person's book. Or maybe it is >possible if someone worked very hard, but then you get into the intent of that >person. I think any sort of "book milking" would be illegal, if that were the intent. It would not fall under fair use. However, if you were to observe games between top engines and discover some interesting novelties, that is another story. I think most good lines are not going to come from computer games anyway. They will come from the human analysis and to some small degree from computer discovered novelties. Another good way to develop lines is to import a huge collection of PGN data (along with analysis) and backsolve it. Bookup and Chess Assistant will both do that sort of thing. That is the only sort of computer generated data I would trust. It would be aware not only of the centipawn evaluation and wins/losses/draws but also of the future outcomes.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.