The "exactly the same code" requirement is kind of silly. People get better and worse over time, as do bots with changing code. The only real issue I see is if a bot enters with a serious bug that then gets fixed.
Sent from my iPhone On Aug 9, 2011, at 7:22 PM, David Doshay <[email protected]> wrote: > At the Portland Congress the situation reared its ugly (in minds of some > folks at the meeting) head. Alan, the President of the AGA is fine with > developing a procedure for ranking bots. At that time it was suggested that > AGA members with solid ratings would volunteer to play bots a fixed number of > times to set their rank. I tried to get people on this list involved in the > testing of bots for this purpose, but there were no volunteers. > > I believe that the previous head of the ratings system (who was pretty > strongly against rating programs) is no longer in that position. He raised > issues about "exactly the same code that was ranked" and was not receptive to > the idea that people make adjustments to the way they view the game too. > > Are there enough ranked AGA members on this list that I can raise the issue > within the AGA again and we can get some bots rated this time? Are authors > interested enough in getting AGA ratings to do this? > > I am willing to be the point person on this effort if there is enough desire > to make it happen. > > Cheers, > David > > > > On 9, Aug 2011, at 4:04 PM, terry mcintyre wrote: > >> Why not ask for volunteers with known AGA ratings to play games in KGS under >> tournament conditions with the top bots? You can use the KGS playing >> interface without trusting the KGS rating system. >> >> Terry McIntyre <[email protected]> >> >> Unix/Linux Systems Administration >> Taking time to do it right saves having to do it twice. > >> From: David Doshay <[email protected]> >> To: [email protected] >> Sent: Tuesday, August 9, 2011 6:57 PM >> Subject: Re: [Computer-go] KGS highest rank Bot >> >> ... >> At this time games played against bots do not have any effect upon your AGA >> ranking. The problem from the perspective of the guy who was in charge of >> the AGA ranking system is that there is no uniform way of giving a program >> its rating. He is completely against using KGS ratings because on the >> internet people can hide behind fake names and can game their ratings in >> various ways. >> >> Cheers, >> David > > _______________________________________________ > Computer-go mailing list > [email protected] > http://dvandva.org/cgi-bin/mailman/listinfo/computer-go _______________________________________________ Computer-go mailing list [email protected] http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
