The official rules to Kasparov-Junior were finally released today. They are almost identical to the preliminary ones I looked at and made small contributions to many months ago. Overall they are vastly superior to the favoritism built in to the Bahrain rules in Kramnik's favor. He got an exact copy of the program, access to the match machine, adjournments, and the Fritz team wasn't allowed to do more than add a dozen or so moves to the opening book between games.
The GK-DJ rules are a fair fight. You bring your computer, I bring my brain, let's get it on. The Junior team can make any changes they like between games. The only non-scientific bit is this: "10.a.: Should a position be reached which is in the machine's endgame databases and if the result from that position with correct play is a draw, then the game ends immediately and the machine operator must promptly advise the human player and the arbiter that the game has been drawn."
This is a strange thing to include if you are going to allow the use of tablebases at all. There are many endgame positions that a tablebase can draw that humans have a very hard time with and in which only one move will draw. We might all remember the Kasparov vs The World game that required reams of analysis of a queen + pawn versus queen endgame. It's unlikely this situation will occur, but it's a bit odd. (Radjabov forced Karpov to defend 50 moves of R+B vs R in Wijk aan Zee just the other day. GMs can lose these things.)
