Author: Robert Hyatt
Date: 19:38:01 01/26/99
Go up one level in this thread
On January 26, 1999 at 13:08:46, Don Dailey wrote: >On January 26, 1999 at 12:50:54, Albrecht Heeffer wrote: > >>On January 25, 1999 at 21:58:46, Don Dailey wrote: >> >>> >>>The most scientific way is to send us a copy of the program. It is >>>not commercial (it's illegal to be commercial) and you should send >>>us the source code. But I will be satisfied if you just send the >>>executable. I have a quad pentium and a dual here at work to run my >>>own tests on. Bob and others may also want to see the executable. >>>We don't expect your logs to match 100% but do expect a high >>>corelation. Post this executable with support files on the same >>>web site so we can clear this up for you please. Post all the >>>source code if you want to remove any doubt. >>> >> >>No problem at all! The program can be downloaded from the >>website at http://www.impakt.be/bionic/download. The version >>used at the Dutch Open is bionic41.exe. The version used at >>the winboard tournament is BIONIC.EXE. But please do not link >>to the download directory on public web pages. >> >>To be clear: we have no commercial purposes with Bionic whatsoever. >>Even if there would be holes in the copyright law, as suggested, >>we would never use Bob's code for such purposes. >> >>Happy testing! >> >>Albrecht Heeffer > > >Thank you for being open to this Albrecht. We will check it out >and if this version proves to be consistant with the logs and the >moves played at the tournament (and our other testing) then I >would consider the matter closed as far as I'm concerned. > >There is the other issue about building on other peoples sources >but I'm not interested in attaching blame to you for this. It's >an issue that needs to be discussed openly and without too much >criticism. You were very up front and direct about this one. > > >- Don I don't really see anything to "close" here... I only see two issues, both 'water under the bridge, but which should be addressed for future chess tournaments of any kind': 1. how significant were the changes to the evaluation? I really don't care. As I said, this is being used in a way different from what I intended, and I think that's pretty obvious. IE the Linux example I gave. I viewed this as a 'single project' that many would be interested in working on and helping with. That has worked out in the past. This is an exception, because many parts of crafty were written by others. The improved compact attacks stuff from Mark Bromley, Eugene's tablebase code, assembly code and optimization suggestions, others too numerous to mention including one coming out that lets the annotate command produce an html file with the usual analysis plus bitmapped diagrams as well. _THAT_ was the entire purpose of this. IE Mike Byrne, another example, fiddled with evaluating "absolute 7th rank" before Jakarta, and that ended up in the stock evaluation... as did many other things by many other people... 2. it is quite clear that bionic did use the parallel search 'verbatim' which was the thing I disliked the most, because it gave them a significant performance advantage with zero effort on their part. And a fair number of amateur programs bit the dust as a result. That I certainly didn't want to see, any more than I wanted to see "Gunda-1" play in Jakarta, because I didn't like having two shots to do well (both finished tied for 4th place I think which makes the point). So 2 is quite clearly true... It isn't a 'major crime' at all, I simply view it as taking something I did and using it against others that didn't / couldn't do the same thing just yet (parallel search). 1. is interesting, but less important IMHO. IE does it really matter how much it was modified? Can we even tell? Since the spirit of the 'freeware' approach has been broken (ie with programs released under the same sort of agreement as Crafty, the changes to the source must _also_ be made public) all of this seems moot to me anyway, since this was not a road I had anticipated riding down...
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.