On Tue, 2008-09-16 at 17:01 +0200, Gian-Carlo Pascutto wrote: > > Disputes that beginners get into are another class of disputes that > > these rules cannot easily resolve without the beginner feeling as if > > they were being "handled." You pretty much have to rely on his good > > nature to eventually just accept the result without questioning it. At > > some point you say, "trust me, you really did lose here even though you > > cannot understand why." > > This is just really really bad. Not only will it annoy the hell out of the > beginner in question, the problems will not only occur with beginners but > every time someone starts to play the game in a matter that deviates a bit > from common practise. Like, let's say, a computer. > > Add in the factor that UCT bots like to play towards half point victories, > and you have a recipe for pain. > > > It's probably just as bad in chess. You have the arbitrary 50 move draw > > rule, castling, en-passant, and the insufficient material draw. For > > those who may not know, there is a class of positions that cannot be won > > no matter how stupid the opponent plays and these are draws and you have > > to know them. To more advanced players these are very simple to > > understand, but to the beginner they can be confusing. > > But the beginner could play out the game without complications. > > The rules can be perfectly implemented and are unambiguous. > > This is a huge difference to Japanese rules.
Yes, I completely agree on this point. You can a set of rules which are not subject to interpretation or ambiguity and this is a win. > > > So you actually have a situation where you can have a won game, play it > > perfectly, but are forced to accept a draw anyway. > > If you have an endgame that isn't winnable due to the 50 move rule, it > isn't a won endgame. It's as simple as that. In order to get unambiguous clear rules, we have had to accept a compromise on the purity of the rules. You have the ridiculous situation where you can announce a forced checkmate, but due to a technicality of the rules you are not allowed to claim a win. > You should play towards an > endgame that is winnable by the rules. I really don't even see the analogy > here at all. I agree, there is no comparison to the wart that is Japanese scoring. So I'm stretching the analogy a bit. The point I should be making about the 50 move rule is that it is unsatisfying in the ideal sense. Yes, it's a good practical concession but really serves as an (imperfect) substitute for the repetition rule. It's a wart. In chess, there is some logic that all games end (at least in principle) with with repetition, stalemate, or checkmate. The 50 move rule is a "practical substitute" for the repetition rule based on the assumption that these games would end anyway with a repetition. Same with insufficient mating material. Now the insufficient mating material is not so much a wart in my mind because it's an unambiguous practical shortcut. It is impossible to win such a game ANYWAY. Even if you opponent WANTS to lose, he cannot (short of actually resigning.) But the 50 move rule clearly fits under the category of strict interpretations of laws that go outside the intent of the law. For example you might save a life by running a red light. The law against running a red light was designed to SAVE lives, not COST lives. (There are probably much better practical examples of blindly obeying laws to the detriment of the spirit of the law.) The actual intent of the 50 move rule is to prevent unreasonably long games due to the fact that some stubborn player refused to admit he cannot checkmate his opponent. The intent is NOT to penalize someone for achieving a won game. Of course you are correct, the rule makes it simple by defining some classes of wins as draws. - Don >
signature.asc
Description: This is a digitally signed message part
_______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
