Hey Russ, On Mon, Oct 28, 2013 at 08:08:41PM -0700, Russ Allbery wrote: > I think I may be missing something here, but what semantics do you believe > ending the current game in the middle of a match should have? Off-hand, > that doesn't seem like a sensible action to me, since wouldn't it then > leave the match status in an undefined state? > > Or, put another way, why would you use end game rather than resign?
The semantics is indeed a bit confused, at least in my mind; I don't exclude that a proper solution for this bug would be improved documentation of what "end game" is supposed to do --- I didn't find any in the doc. So, first of all, whereas resigning implies that you lose, "end game" does not. Using "end game" it's the AI that will play in your stead until the end of the current game. So you can also win. I believe the intended use case is that you use to quickly terminate a game, if you are at present not interested in playing, say, the bear off phase. But if you're ahead in that specific game, you do expect (statistically) to win that game. If that happens in, say, the first game of a match to be played at 11-points, you do expect to be able to play yourself the subsequent games. It is this that seems to be impossible, and it smells like a software bug (something like: triggering the boolean "match ended" instead of the boolean "game ended" once done). Hope this clarifies, Cheers. -- Stefano Zacchiroli . . . . . . . z...@upsilon.cc . . . . o . . . o . o Maître de conférences . . . . . http://upsilon.cc/zack . . . o . . . o o Former Debian Project Leader . . @zack on identi.ca . . o o o . . . o . « the first rule of tautology club is the first rule of tautology club »
signature.asc
Description: Digital signature
