Commentary for black move 4: And Garry at the moment, I just wanted to say about his choice of openings. Let's be very clear. What do we mean when we talk about the opening book? Michael would probably speak to this better than I. But the idea is there's a database of games, hundreds of thousands of games in the database, and the computer is repeating the moves played by Grandmasters of the past and the present. But, who are the people that actually create the moves, if not the world champion Garry Kasparov? And Garry Kasparov has introduced I dare say more opening /TPHO*FLTS than any world champion in the history of the game -- maybe that's a slight exaggeration, but not by much -- so Garry I think should be challenging the computer in the openings, because he's the one who creates the openings himself. GK MOVE: 4...d6.

MAURICE ASHLEY: After much thought from Deep Blue's side, it has responded a bit more reservedly than we thought and has instead played the move d7-d6, which now Kasparov has to wonder if this -- I'm sure he must have thought about this idea, but certainly he's not going to get that favored position, the position he wanted with d7-d5 which would have given some kind of reverse Sicilian which is a formation that Kasparov plays extraordinarily well. We saw it a lot in the match against Anand in 1995 but teds all the time. So I think his state of mind now must be he's got to stop and think a little bit. Apparently -- and he's doing so now, he's not just whipping out his first move.

YASSER SEIRAWAN: a3 /HRAO*BGS looks like almost a wasted move in this position.

YASSER SEIRAWAN: I think this is the kind of opening that Garry actual hoped for. The move d7-d6 is slightly passive. I would have liked /TO*FZ the open Sicilian after c4xd5. The purpose of the move a2-a3 is to play b2-b4 in the near future, getting a space advantage on the queen-side following by fianchettoing the bishop with Bc1-b2. At the same time because of white's pawn structure c4, d3, e2, this white bishop on f1 needs a good diagonal, and the best way of doing that is to play g2-g3, fianchettoing the bishop by Bg1940 castling king-side as he did in game one. This is known as building a house, Nf3, g3, Bg2. White sticks his king behind a nice solid defense and the bishop on g2 is a very good protector indeed. So we probably see something like a double fianchetto very much as to game one. A game, by the way, that Garry did win and out played the computer nicely, and I must say another controversy seems to bey ruptureing on game one. We've heard rumors that the Deep Blue team thinks that Deep Blue could have drawn that game.

YASSER SEIRAWAN: We have no information, though.

YASSER SEIRAWAN: Not at the moment.

MIKE VALVO: We have no idea what they're talking about. We're going to ask them, though, when they get on stage.

MAURICE ASHLEY: We have a full day for you. At four o'clock, one of the Deep Blue team members will be appearing on stage, and we won't let them duck any key questions, or we'll try to get them to duck, and then at five we'll have the I will U.S. Russ Grandmaster Roman Dzindzihashvili, and he'll tell us about his understanding of the goings on in the match, and two days ago, the draw, possible draw. All of that, we'll try to get as much information from these illustrious people as possible.

Real-time text commentary is made possible by LiveNote, Inc. and Vincent Varallo Associates