Author: Robert Hyatt
Date: 19:50:46 10/21/03
Go up one level in this thread
On October 21, 2003 at 19:08:34, Sune Fischer wrote: >On October 21, 2003 at 15:55:40, Robert Hyatt wrote: > >>That's all irrelevant to this discussion. I was pointing out that the human >>should _not_ be able to either offer or accept a draw on behalf of the program. >>The program should be able to handle this itself. "draw" should be a legal move >>the machine has to accept. It can either make a decision or just print "draw >>declined" but it ought to accept the offer text and do whatever the author >>thinks is best. The human should be a bystander. Not an active participant. > >In automated events I would agree with you, but you know how dumb programs can >be, e.g. not seeing an obvious draw until the 50 move limit approaches. That's OK. It is an incentive to fix the problem. :) I had this facility in Cray Blitz for years, way before there were such things as "chess servers". > >When playing manually I can imagine both humans would feel pretty silly going >through that ordeal. >Don't you think it's embarrassing enough just watching it happen automated? :-). Again. "incentive" :) That's why Crafty resigns, offers draws, etc. No, it doesn't offer draws in _all_ the drawn endings. But it gets it right enough that I am not embarassed daily. That was why I fixed it. Ditto for acccepting/declining draw offers. > >That's not to say we shouldn't make programs smarter in understanding obvious >draws, I just doubt they will be as good as us any time soon. Programs are likely going to miss "exception" draws for decades. But obvious draws (KR vs KR for example) or KQ vs KQ can certainly be handled. Handle enough and you at least avoid much embarassment. To avoid _all_ embarassment, you should find another hobby. :) > >-S.
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.