Author: Will Singleton
Date: 22:59:29 10/29/99
Go up one level in this thread
On October 30, 1999 at 00:52:35, Dann Corbit wrote: >On October 30, 1999 at 00:40:42, Will Singleton wrote: >[snip] >>Thanks for the thoughtful response. I'd like to say that I learned more about >>king-safety from our brief discussions over a year ago than I would have from >>examining your code. That's because I was energized by the *ideas* you >>imparted, in conversation; I would never feel the same way about digging in >>code. >> >>Some might ask, what the difference? For me, it's the challenge of taking an >>idea and melding it into my previous ideas. Or replacing them, for that matter. >> But one must make his creation his own, or there *is* no creation, just >>duplication, and no real advancement. >> >>So, what is the point of publishing source code? Is it to advance the science? >>What can possibly be gained from complete source code that could not be obtained >>from the ideas contained therein? (I am talking about advanced programs like >>EXchess and above, not tscp.) The only thing to be gained is the ability to >>copy and cheat, nothing more. >> >>This argument assumes a viable forum to present ideas. I think that exists here >>in CCC, in the ICCAJ, and other places. I would be interested to hear from >>others who publish their source to comment. Again, I am not against sharing >>information, quite the contrary. I only advocate providing the means to >>advance, instead of dumping the advancement as a realized whole. > >I think your position is very, very strange. > >If you divulge your ideas in excellent clarity, then it is trivial to create the >code. So where is the harm in showing the code? It is only harmful if you >can't do a good job explaining or don't bother to explain. Otherwise there is >no difference whatsoever that I can see. I also don't believe that you could >cut and paste from a complex system like crafty into existing code and make it >work well. The only way to extract ideas is either to be a big sneak and steal >the whole ball of wax, or to study it, understand it, and use the ideas it >contains. > >Yes, people can copy and cheat. I think they probably get caught every time >they do it. And if not, what have they won? The hollowest imaginable victory. >In time, they will get left in the dust because there is no way that they can >compete without understanding the fundamental mechanisms behind the code. So >the best they can do is manage par out of the gate, and then they will rapidly >get run over. > >I have learned a tremendous amount from the algorithms published in the ACM. >They are not written in any language that I normally program, but it is trivial >to translate them in general. With just a description rather than an algorithm >not only is the translation more difficult, but the understanding as well. > >In short, I disagree very, very strongly that anything is wrong with releasing >source code. > >However (once again) I respect anyone's right to keep their code private. Dann, I don't think *your* position is strange at all. :) I understand your arguments, and I think they have a lot of merit. So, try to understand mine. Programming can be considered an art as well as a science. While the basic ideas are not too hard to learn, to do it very well takes an exceptional person. When you publish the source to an exceptionally fine work, you can short-circuit the creative process in others. Not always, and not with all people, but it happens. You present the opportunity to avoid the work that otherwise would have led to real understanding, and perhaps independent advancement. When you say that it is trivial to translate pseudo-code, that may be more true for you than me. Nonetheless, doing so assumes that you have something of your own to translate it into. Why would you spend the time writing your own program when you could just study Crafty and be happy? Will
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.