Prematch commentary.
ROBERT J. T. MORRIS:  Chess is war.  You might have seen that on
  the front page of a major newspaper yesterday.  Of course,
  chess is just a game, okay, but it does represent a contest,
  and throughout history, in every contest, whether it's being
  man, in survival against his environment, or man against
  disease, or man against man, there's been a critical turning
  point when man has started to use tools.  You can call them
  tools, weapons, perhaps, or technology, to change that game, or
  that contest.
And in fact we're at a critical juncture in human history today,
  as man begins to use tools in contests of intelligence.
And not only that, today, with Deep Blue, playing as it is, I
  think we've reached a coming of age for this technology, the
  technology in contests of intelligence.  So I would like to
  introduce to you Monty Newborn, who is the chairman of the ACM
  chess committee.  Monty is overseeing the match.
MONTY NEWBORN:  I'd like to welcome everybody here to an
  afternoon of one of the most exciting chess games that we've
  anticipated in the history of the game.
The match right now stands at two points apiece, and there's two
  games to go, and it should be very exciting.  We're about five
  minutes from the start of the game, and before I introduce the
  commentators, I just want to point out that the ACM, of which I
  am the chairman of the ACM Computer Chess Committee is
  providing commentary for this match and officiating.  We have
  three great commentators who will bring your afternoon to life
  and answer your questions, and we'll have a few guests on
  stage.  And I suspect you'll have a wonderful afternoon.
In my capacity as chairman of the ACM Computer Chess Committee,
  I'd like to thank IBM for having us as the officiating
  organization, and it's been a real pleasure.
My three commentators will come out on stage one at a time.  The
  first one is Mike Valvo.  Mike is an International Master.
  He's also one of the best blindfold chess players in the United
  States, can play ten to 15 people with his eyes closed, kill
  them all, including myself, and he's also been involved as a
  commentator and an arbiter at countless computer chess events
  over the years.  Mike?
MIKE VALVO:  Thank you, Monty.  By the way, chess is life, not
  war.
MONTY NEWBORN:  The second commentator is Maurice Ashley.
  Maurice is an International Master and New Yorker, a coach for
  one of the outstanding high school chess teams in the United
  States, and has tremendous experience in bringing chess matches
  to life.  I think Maurice will make this sound like more of a
  football game by -- than a chess game by the time it's
  finished.  Congratulations.
Next, last, but not least, is Yasser Seirawan, who is three-times
  U.S. champion.
This is a team of outstanding talent and I'm sure you're going to
  have a very enjoyable afternoon, and I wish the opponents
  upstairs the best of luck and I'm sure you'll all have a
  wonderful time.  Maurice, I'll pass the mike to you.
MIKE VALVO:  We don't need a mike.
MAURICE ASHLEY:  Thank you.
YASSER SEIRAWAN:  Monty Newborn.
MAURICE ASHLEY:  Welcome to game five of the IBM Kasparov vs.
  Deep Blue rematch.  Are you as excited as I am?  The score is
  tied 2-2.  Many people didn't believe that this would happen.
  Yasser Seirawan, I know you were one of them that felt like
  Kasparov was going to destroy Deep Blue in the rematch, but it
  seems as if that has not happened.  This has captured the
  emergencytion -- imagination of the entire world, the New York
  Times, USA today, everybody is involved in this one, everyone
  wants a piece of the action, and I've never seen so much
  excitement in a chess match.  Yaz, when has there ever been
  such excitement for a match?
YASSER SEIRAWAN:  Well, it's a very interesting match.  The
  circumstance, once again.  We saw it in Philadelphia 2-2.  But
  in the Philadelphia match, there was a sense that Garry was
  just waiting, waiting to pounce.  And he won the last two
  games.  So, in a sense, it was anticlimactic, but there was a
  lot of tension in Philadelphia nonetheless.
This match feels a lot different to me.  It's the same
  circumstance, but the confidence isn't there that I had for
  Garry as in Philadelphia.  And it's really getting
  nerve-wracking.  Today, in my opinion, is the pivotal day.
  Garry won with white; he's lost with black; he drew with white;
  he drew with black.  And it's very clear his match strategy is
  to press for the white pieces.  He presses in an unorthodox
  kind of way, but I'm sure he doesn't want a draw today.  I'm
  sure he feels he needs to win today to clench the match, at
  least a tie.  And he doesn't want to draw today and go into the
  last round tomorrow with all the marbles at stake.
MAURICE ASHLEY:  And Kasparov has taken his seat.
(Applause.) The watch has come off immediately which generally
  means that he's ready to get down to business.  Kasparov with
  the white pieces, game five, similar conditions to last year,
  but as Yaz mentioned, game five last year, Kasparov had the
  black pieces and he was quite content to draw.  And, Mike, what
  was the feeling?  You were in the room as arbiter last year.
  What was the feeling in the room of game five last year and how
  is it different than this game?
MIKE VALVO:  Last year's game started off with what appeared to
  be a Petroff's Defense and Garry was a little bit surprised
  when they just moved very quickly and later on after the game
  he asked "Did you know what I was going to do?  How did you
  know what I was going to do?"  And it was very quickly a Scotch
  game, and Garry seemed very confident and in a commanding
  position he offered a draw knowing that he had white in the
  last game, and he didn't care what they did, and in a position
  where the computer was, to its own point of view, a little bit
  worse, they turned it down, continued to play, and Garry won
  the game.  It was an overwhelming situation.  They never
  recovered, Garry just crushed it like a peanut in the last
  game.
MAURICE ASHLEY:  This is a different story.  Again Kasparov with
  the white pieces, the pressure is on him to win this game.  He
  doesn't want to have to deal with the black pieces tomorrow.
  Having to win in order to win the match.  As you know, Kasparov
  wants desperately to win this match.
He is located now in the 35th floor of the Equitable Building.
  We should point out that we have an in-house audience but we
  also have many fans from all over the world watching live on
  WebTV, following live over the Internet.  So we will keep them
  abreast of what's happening visually.  They can't see, some of
  them cannot see it because they are only following on a
  chessboard, and we will be very descriptive at times of the
  events occurring in this program.
Kasparov now sitting taking a photo opportunity as there are
  many, many members of the press who are interested in taking a
  picture.
GK MOVE:  1 Nf3
MAURICE ASHLEY:  Finally he has played a move.  N if 3 is move
  one of game five.  A very, very important game.  And quickly
  Deep Blue --
DB MOVE:  1...d5
MAURICE ASHLEY:  Deep Blue has responded d7-d5.  This is
  continuing the same moves of the first game.
GK MOVE:  2 d3
DB MOVE:  2...Bg4
GK MOVE:  3 Bg2
DB MOVE:  3...Nd7.
GK MOVE:  4 h3
  (missing).
YASSER SEIRAWAN:  Mike, you're the computer expert here, please
  explain to us.  Throughout the entire matchbook, why we're
  laughing is throughout the entire four games of the match, the
  computer has jealously guarded its bishops.  It has not
  voluntarily given up its bishops.
DB MOVE:  4...Bxf 3
MIKE VALVO:  Well, you look at a position like this, and you say
  black's got a real potential to get a large control of the
  center, and that may be worth -- more worthwhile to the
  computer to get that than to keep the bishop vs. the knight but
  I don't see how they're going to do that right away because as
  soon as the bishop captures the bipartisan the pawn on d5 is
  going to be hanging.  And obviously Garry is going to do
  that -- I think.
YASSER SEIRAWAN:  I obviously agree with you.  I expect Garry to
  play Bg2xf3, attacking the d5 pawn.  Let's assume if you will
  Ng8-f6.  This keeps the possibility of e7-e5, grabbing a big
  share of the center.  And the position will very quickly
  transpose into themes of the King's Indian Defense, which Garry
  loves.  We may see a move, for example, like castles.  We are
  just trying to anticipate the further course of the game.
  e7-e5.  d2-d3, and a move like c7-c6, blocking the long
  diagonal h1-a8, Nb1-d2.  And we have a position which is
  similar, like I say, to a King's Indian, as well as a Pirc, and
  this is the kind of game that Garry would hope for where the
  opportunities of different strategic ideas all lie with white,
  with ideas like c2-c4, or e2-e4 or dropping back and keeping a
  defensive position.
MAURICE ASHLEY:  What do you think about Kasparov, having
  prepared this specific opening, he knew what happened in game
  one, but yet he has not moved.  Is he so surprised, as we are,
  by the fact that the computer has given up the bishop for the
  knight just after two minutes' thought?  Did he maybe not
  prepare for this particular variation thinking for sure the
  bishop was going to drop back?
GK MOVE:  5 Bxf3
MAURICE ASHLEY:  He has finally captured on f3 with the bishop.
MIKE VALVO:  Why did he take so long?  Why did he consider the
  only other move, exf3?
YASSER SEIRAWAN:  No, this is very important.  A chess
  Grandmaster, Garry Kasparov, for the last two days, has been
  comparing with his trainers and his computer specialists, and
  he sat there and he said to himself "I'm going to get this
  particular kind of opening, and this is the pattern that I'm
  going to get, this is how it's going to work, and then I'm
  going to beat the computer like a dog."
So he comes to the game full of vim and vigor --
MAURICE ASHLEY:  I should note --
DB MOVE:  5...c6
GK MOVE:  6 d3
YASSER SEIRAWAN:  And this is where we're anticipating e7-e5 by
  the computer.  But anyway, Garry comes to the game very
  resolute, he knows exactly what kind of opening he wants to get
  himself involved in, and then he comes and immediately he gets
  shaken by this capture of Bxf3.  So he has to mentally, as well
  as emotionally, readjust himself to the fact that his opening
  preparation for this particular game has been thrown off
  balance.  Right away, very early.  So he has to re --
MIKE VALVO:  "Re" something.
YASSER SEIRAWAN:  Regain control of the position, mentally and
  emotionally.
MAURICE ASHLEY:  Well, we see that the clocks show 1:55 each
  remaining for both sides.  They have, for those of you who
  don't know, two hours to make 40 moves, for each side has a
  clock, a two-faced clock, two hours delegated on each side for
  the moves.  We should also bring up to speed the people who are
  not able to see.  Our audience was not able to see what's
  happening in house.  We are three on stage.  We have Indus
  three video monitors.  To our left is a monitor that we keep a
  running tally of the position at all times.  And it shows the
  current position on the board.  Also it shows the moves right
  below it.  Directly behind us we have a monitor showing the
  computer program, Fritz 4, who helps us with analysis, and we
  find it's partially biased to Deep Blue, but will -- but is
  often turncoat when it sees the tile to -- time to switch
  sides.  And to our right is the monitor of the player area of
  the 35th floor again of the Equitable Building, where Kasparov
  is apparently pacing back and forth.  Seems he is content with
  the situation.  He often just paces.  He has a dressing room
  that he could go to whenever he wishes, and there's a
  television monitor in the room so that he can always know
  what's happening in the game and when it is his turn to move.
  He's facing an opponent it seems but it is in fact the Deep
  Blue operator.  I believe that is --
DB MOVE:  6...e6
MAURICE ASHLEY:  -- one of the Deep Blue operators, who has now
  move e7-e6, Jed of jumping into the center with a double pawn
  thrust, this move, Yaz, was unanticipated by us on stage.
YASSER SEIRAWAN:  Most definitely.  Many of the computer's
  algorithms describe to the computer that it's great to push
  pawns.  We have seen that several of times in the match that
  Deep Blue has thrust its pawn shield protecting its king up the
  board.  So computers love to push pawns.  We hadn't anticipated
  this restrained move e7-e6.  The only justification for
  capturing that knight on f3 was that it controlled the e5
  square, and therefore we had anticipated e7-e5.
I'd just like to explain one other thing.  It's highly unusual fo
  Garry Kasparov in a top tournament or match would be playing in
  an enclosed room.  It's far more appropriate that he should be
  on stage where we are and that we would be off in an enclosed
  room lecturing to an audience such as yourself.
However, because the audience would affect Garry Kasparov and not
  Deep Blue, they're playing in a closed, quiet, sound-proof area
  where there's no possibility of Garry being disturbed.
MAURICE ASHLEY:  In fact n world champion matches, prior world
  champion matches for example between Karpov and Kasparov in
  Moscow, there would be an audience of 3,000 in the hall and
  these guys would be on stage and you could not tell those
  Russian fans to keep quiet because they wanted to shout out
  moves.  (Audience laughter.)
YASSER SEIRAWAN:  He is absolutely right.  In fact, in several
  world championship matches, the Russian audience, who is
  really, really dare their culture loves and embraces chess, not
  quite like ourselves, although yourselves are making believers
  of me, the Russian audience understood when a mistake was being
  made, and when a mistake had been made there was a whole
  rustling crowd in the audience and it almost seemed to alert
  the players.  They go "Oh, thank you.  Thank you very much.
  How about that?  You blundered."  Because the audience
  understood what was going on.  Now let's go back to h3 where
  the bishop captured the knight.
MAURICE ASHLEY:  Okay.
MIKE VALVO:  From a computer's perspective if it played Bh5 the
  other move, if I was concerned -- it might be concerned about
  g4 Bg -- Bg67 -- it may have been concerned about the doubled
  pawn needlessly.  That may be the reason why it was concerned,
  because of the pawn structure.
YASSER SEIRAWAN:  An interesting point by Mike.  Obviously ux,
  Mike, you and I both know that Garry would not have pursued
  that variation, but it doesn't matter.  It's what the computer
  thinks, or worries about, and that is potentially one of the
  reasons why it played the way it did.
MIKE VALVO:  I once played a match against a computer where we
  didn't let the computer have any book at all, and we just
  explore the opening lines and see what the computer would
  think.  And there would be situations where the computer would
  notice, in its analysis, better moves for me, even though I
  didn't make them, and then the next time we played the same
  line, it remembered the prior game it wouldn't go into this
  because it knew I had a better move.  And that's the
  consequence of something like this.  The computer knows that
  Garry could do something, even if he never intended to do it it
  wouldn't go into it because it knew it could be done.
MAURICE ASHLEY:  Good point.  Well, we do have a full afternoon
  for you.  There are a number of guests we often bring on
  stage.  Today we have that the IBM today, Matt who is the on
  the technical team, he won't give us too many secrets, but
  we'll try to get -- and later we'll also have woman's world
  champion Susan Polgar who was with us last time and she offered
  some tremendous insights into what was going on in the last
  game, and later we will have former U.S. -- actually two-time
  U.S. champion Grandmaster Patrick Wolff and he'll be on stage
  to offer his insights as to how the match is going -- how this
  game is going and how the match is going so far.
We should also mention that we have some youngsters in the
  audience who have been invited here by IBM.  I believe it is
  P.S. 116, P.S. 116.  Let's give them a hand.
(Applause.) May be a future Bobby Fischer in the crowd.  Who is
  the best player in P. S. 116?  Is there an agreement here?
  Okay, what's your name?  Youngster right here.  Give him the
  microphone.
AUDIENCE MEMBER:  Flynn.
MAURICE ASHLEY:  Flynn?
AUDIENCE MEMBER:  (Nods head.)
AUDIENCE MEMBER:  --
MAURICE ASHLEY:  Who do you like in this match, Flynn, Kasparov
  or Deep Blue?
AUDIENCE MEMBER:  Kasparov.
MAURICE ASHLEY:  And what do you think about this game so far?
  What's your opinion on this game right now?
YASSER SEIRAWAN:  Has white got a better position or --
AUDIENCE MEMBER:  It's even.
MAURICE ASHLEY:  Sorry?
AUDIENCE MEMBER:  It's even.  Said with authority!  It's even.
MIKE VALVO:  He's right.
MAURICE ASHLEY:  Good suggestion.  Okay, Flynn, we'll be back
  over there asking some more opinions and if you see some more
  good moves, let us know, okay?
YASSER SEIRAWAN:  Also, throughout the afternoon we'll be asking
  for questions from members of the audience.  There's a roving
  mike.  We do point out, please take the time to accept the mike
  and then ask questions of any of us on stage, and we'll be very
  happy to --
MAURICE ASHLEY:  What do you think about this position?  It seems
  a couple of moves so far -- e6 -- just as I speak --
GK MOVE:  7 e4
MAURICE ASHLEY:  A very aggressive move by Kasparov, saying he
  wants a piece of the space, a piece of the center, and he has
  even gotten up and left the stage to go his dressing room,
  shows his confidence in the situation.  What do you think?
YASSER SEIRAWAN:  I share his confidence, let me tell you.
  Actually I'm a little bit surprised by the early commitment of
  e2-e4.  Personally I, too, like the computer, have a prejudice
  for bishops.  Bishops work well in an open position after lots
  of pawns have come off the board, so in a sense, in the early
  part of the game, the knight, in a closed position is
  oftentimes more valuable.  But further you get into the endgame
  after more and more pawns come off the board, the bishops get
  better.  So from a long-term perspective, I like white, or
  Kasparov's, chances better, but in the short-term, what
  Kasparov has to do is take a stand in the center and figure out
  a way of opening up his diagonal.  And the move e2-e4 to my
  mind was a bit premature but not a bad move by any stretching
  of the imagination, but a bit premature.
MIKE VALVO:  There might be an advantage to this because Kasparov
  knows that the computer guesses his moves half of the time, 50
  percent of the time.  So he knows the computer is going to
  expect him to castle by playing a different move.  The computer
  has to start from scratch rather than having a running start of
  about three minutes.
MAURICE ASHLEY:  And the computer thinks very, very quickly.
  That I think is an understatement.  200 million positions per
  second quickly.  I had my students try to blink that first, and
  you should have seen the kids going...
YASSER SEIRAWAN:  A question, I think, in the audience, there is
  a young -- one of the young --
MIKE VALVO:  Is there a question over here?
YASSER SEIRAWAN:  Yeah, I was sure there was one.
AUDIENCE MEMBER: P I think he was just waving.
YASSER SEIRAWAN:  He got shy.
MIKE VALVO:  Everybody is hiding their hands all of the sudden.
MAURICE ASHLEY:  Well, Kasparov has returned to the board.  And,
  Mike, the opening, again, we know in chess is so, so
  important.  It's where the whole groundwork is laid for what's
  going to happen, the subsequent events in the middle game and
  beyond.  I know that Deep Blue has some really good analysis --
  a really good analysis team, some really good helpers.  Joel
  Benjamin is one but the New York Times even reported that
  Grandmaster Miguel Illescas from Spain is helping him.
MIKE VALVO:  That seems to be the case.  There are a lot of other
  Grandmasters around, and I wonder if they're also part of the
  team, too.
YASSER SEIRAWAN:  Do you think Deep Blue has a lot of secret
  helpers in the Grandmaster class?
MIKE VALVO:  I think he's got some helpers in there.
MAURICE ASHLEY:  How much can they help Deep Blue n actuality?
  What can they do -- 7...Ne5
MAURICE ASHLEY:  Deep Blue has now played a highly unusual move
  before you get to your next point, Ne5.  And this, to my eyes,
  is a very, very surprising move, Yaz.
MIKE VALVO:  A very bad one, too.  (Audience laughter.)
YASSER SEIRAWAN:  You're the one who believes in the future here,
  Mike.
MAURICE ASHLEY:  What an unusual move, this kind of move.  This
  breaks all principles.  I tell my kids don't move the same
  piece twice in the opening.
GK MOVE:  8 Bg2
MAURICE ASHLEY:  Kasparov didn't need to think too long about
  that lemon and moved back Bg2.
YASSER SEIRAWAN:  I actually think Ne5 is a good move, and that's
  one of the reasons why I didn't like e2-e4, although I didn't
  mention that, of course --
MAURICE ASHLEY:  You're keeping secrets from us, Yaz.  You can't
  do that.
YASSER SEIRAWAN:  If you just look at the position two moves ago
  from Deep Blue's perspective, we know that the computer has a
  prejudice for bishops.  So we could assume that the computer
  might like to go after Garry's bishop with the move Nd5-E.  And
  you're right, Maurice, as a teacher, the principle is not to
  move a piece that is already developed twice in the opening
  because you'll waste time.  Garry, to preserve his bishop had
  to retreat it, and this is the position that Deep Blue was
  contemplating.  Deep Blue is now contemplating d5xe4, giving
  Garry the choice.  If Garry wants, he can capture the pawn on
  e4 with his bishop, but then he in turn has been moving his
  bishop back and forth, and then Deep Blue gets to say to
  itself, "Well, I want that bishop.  I'll now play Ng8-f6,
  attacking the bishop and setting up NxB, and if you, white,
  really want to preserve the bishop you're going to have to move
  it again, and so I gain time.
MIKE VALVO:  Was that a great tap dance, or what?  Wow.
MAURICE ASHLEY:  The bishop went to g2, to f3, back to e2.
YASSER SEIRAWAN:  He's spending time trying to keep his advantage
  of the two bishops.  In this resulting position, you might say
  black has a fairly easy road to equality.  The bishop would
  move to c5, Bf8-c5, and they're in development.
Now, let's go back to the position at hand, that is to say, what
  Deep Blue is thinking about.  Garry has another alternative.
  He could play after dxe4, we have a new variation.  Garry could
  trade queens.  Now, one of the things that we've seen in this
  match, and especially in game four, Garry had all of his
  attacking potential that he sought to exchange queens, even
  when he had this possibility of Qc7 going for the attack.  So
  Garry may embrace the ending d5xe4 d3xe4 Qxd1+ Kxd1 O O+ and --
  O O O+ and Ke2.  In the endgame the king on e2 will be a good
  piece.  This is what Deep Blue and Garry are probably looking
  at.
DB MOVE:  8...didxe4.
MAURICE ASHLEY:  What we had anticipated has occurred.  The
  exchange dxe4 has occurred, and Kasparov has his first big
  decision of the chess game, and the whole character of the game
  is about to change.  Kasparov is bouncing in his chair, and he
  knows it's time to decide.  If he plays Bg2 -- takes Bg2, he
  keeps queens on.  Maybe he'll exchange queens, thinking it's
  best to play Deep Blue that way, play an endgame where
  computers have been known, Mike, to not be so good?
MIKE VALVO:  Yeah, it's an area where computers are probably at
  their worst, the endgame.  They would be even worse off in the
  opening except that the openings are programmed into them from
  the start, so they usually don't wake up until about move 10 or
  15 unless they're playing Garry, then they wake up about move
  three or four.
MAURICE ASHLEY:  Good point.  And still Kasparov is thinking
  about this position, despite the fact that we've been eager, I
  mean eager, like a man from a mission, the moment he sees the
  opportunity to exchange queens, has done so in the blink of an
  eye.  But here for some reason, he's really hesitant, and maybe
  he's not so confident after that position after the exchange of
  queens, Yaz.
YASSER SEIRAWAN:  Possibly, but at the same time I do expect
  Garry to go into this endgame for all of the reasons that Mike
  stated.  His prematch strategy clearly was to trade queens, go
  in the endgame and do the best he can in the endgame.  We had a
  very interesting endgame in game four.  How many of our
  audience was here for game four.
GK MOVE:  9 Bxe4
MAURICE ASHLEY:  And he has avoided this endgame!  He has decided
  to avoid the endgame, by playing Bxe4, this is the Garry we
  know, and have come to love, I might say.  Garry Kasparov is
  know worldwide for his attacking ability, for the fact that
  when he's got queens on, he's got two bishops, he usually just
  rips the opponent's position apart, and clearly right here he's
  saying, "All right, kid gloves off, it's time to rock 'n' roll,
  let's get busy.  And now Kasparov, Bxe4, this bishop having
  moved already, is it four times in the opening -- Bg2 was one,
  takes on 1960 3 was two, back to g2, and now e4, and seemingly
  to move a fifth after the natural move Nf6, that Kasparov has
  decided that the natural pent-up power of his two bishops is
  going to be more than enough value for this waste of time.
So, Yaz, an aggressive posture by Kasparov.  This is not some
  kind of passive move.
YASSER SEIRAWAN:  Well, I had expected, all of Garry's obvious
  prematch strategy, to trade queens, with the move dxe 4.
DB MOVE:  9...Ngf6
MAURICE ASHLEY:  The natural move, developing the knight and
  attacking the bishop on e4.
YASSER SEIRAWAN:  Garry is going to have to retreat the bishop on
  e4.  There's no --
GK MOVE:  10 Bg2
MAURICE ASHLEY:  He has dropped back.
YASSER SEIRAWAN:  We have now predicted the move Bf8-c5 by Deep
  Blue.  One of the ideas in the opening is to castle early,
  protect your king, but also to bring your pieces to active
  squares.  When we say active squares, the definition is if you
  bisect the chessboard into two halves, right across from the a4
  rank, then you say 32 squares belong to white, 32 squares
  belong to black, and to activate our pieces we need to control
  more of the opponent's territory than the opponent controls of
  ours.  So the bishop on f8 clearly needs to move but where to
  move it.  e7, d6, b4 with check.  Of which those choices --
  which of those choices?  Bb4+ is attractive and Bc5 are the two
  most attractive squares because they control white's
  territory.  We're anticipating one of these two moves.  And
  quite frankly I think Deep Blue has a fine position, although,
  long-term, we know the bishop on g2 has no counterpart and is
  very meaningful.  In the short-term, the computer has good
  development.
MAURICE ASHLEY:  Well, Mike, you and I didn't like the Ne7-e5
  move.  It seems to have turned out okay.
MIKE VALVO:  That was a knee jerk reaction by me, but now is
  seems Bb4 check is a good move, because c3 is not possible,
  because of Nxd3.
MAURICE ASHLEY:  Let's look at that.  Bb4+.  And the normal move
  here would be to gain a tempo with c2-c3.
AUDIENCE MEMBER:  Not every check is a good one but there is the
  possibility in that case of Nxd3+ and black would just be up a
  pawn.  For example, after the move Ke2 attacking the knight on
  d3 and now the bishop is also under attack.  Black could play a
  simple move like Nxc1 check, and then Kasparov could resign
  pretty soon.
YASSER SEIRAWAN:  He snapped off a pawn.  So by the way on that
  previous variation, if you notice the red toolbar there,
  Maurice, you do a better job than I explain what Fritz is up
  to.  We use Fritz as a tool here on stage, and the idea is that
  every position has only a select number of legal moves.  If you
  look down into the lower part of the screen in the center --
DB MOVE:  10...Bb4+
MAURICE ASHLEY:  Fritz has decide to do play the move Bb4+ to
  test Kasparov's response.  I think Fritz -- excuse me -- Deep
  Blue has played this move.  Deep Blue has played Bb4+.  And I
  doubt Deep Blue thinks Kasparov is going to play c3.
GK MOVE:  11 Nd2
MAURICE ASHLEY:  Kasparov has played the much more natural move
  bishop to --
AUDIENCE MEMBER:  No!  Knight.
MAURICE ASHLEY:  Knight b1-d2 has been played.
YASSER SEIRAWAN:  Blocking the check.  And so here we have it.
  On the lower screen there, we see that after the move -- boy,
  it's moving around -- right now, Deep Blue has 48 possible
  moves.  48 possible legal moves.  That's in that lower screen
  there.  It's going through each and every one of those
  possibilities, and it's going to select what it believes to be
  the best choice out of the 48 possible.
Right next to it is you see the depth chart.  What that means is
  that -- actually, before I make a fool of myself, Mike, why
  don't you tell the audience what the depth chart means.
MIKE VALVO:  Let's way till Frederic comes back.  I'm not really
  sure what that means.
YASSER SEIRAWAN:  Okay, I'll take a guess.
MIKE VALVO:  It's supposed to mean --
YASSER SEIRAWAN:  The number of ply.
MIKE VALVO:  The number of ply that it's going into but he said
  something which doesn't quite make sense to me here.  I don't
  know why there's two numbers.  Usually it would say 9, which
  would mean 9 ply.
YASSER SEIRAWAN:  10 of 10 ply.  What a ply refers to as a half
  move.  So if I make a move and my opponent responds, that's two
  ply.  So what the computer is doing is analyzing every possible
  ply for five moves, five complete moves, or ten ply.  Below
  that, you have a little toolbar, and the toolbar is a visual
  aid to say who's doing what to whom, who has the advantage in
  the view of Fritz.  Red is for advantage for white.  Green is
  for advantage to black.  And, finally, above that you have a
  numerical value --
MAURICE ASHLEY:  Actually green is white and red is black.
YASSER SEIRAWAN:  Just reverse everything I just said.  And above
  that you have a numerical value.  You have 38.  What that means
  is that in this particular position, Fritz thinks that I
  believe white --
MAURICE ASHLEY:  White.
YASSER SEIRAWAN:  -- is doing very well.
MAURICE ASHLEY:  Doing very nice.
YASSER SEIRAWAN:  Has 38 points of a pawn advantage.  Pawn equals
  a hundred points.  So it thinks 38 points.  We do have a
  question in the aweience.
DB MOVE:  11...h5.
MAURICE ASHLEY:  No!
YASSER SEIRAWAN:  h7-h5!  Well, yeah.  h7-h5?  We've got to take
  a look at Garry now!
YASSER SEIRAWAN:  Garry is a happy camper!  Who's been
  programming this machine?
MAURICE ASHLEY:  I mean we've seen some strange moves and we've
  seen some strange moves.  That's a strange move!
MIKE VALVO:  I think I better go upstairs and see what's going on
  here.
MAURICE ASHLEY:  Okay, well, we will give Mike his well-timed
  break and he will go to the press room and get a few from the
  49th floor, the Grandmasters, there are many, many Grandmasters
  all over the world on that floor, and Mike, a little
  reconnaissance mission for us.
YASSER SEIRAWAN:  Absolutely, from the computer scientist.  IM
  Michael Valvo.
MAURICE ASHLEY:  All of us are shocked with the move h7-h5.  Let
  me just catch my breath and get a question and then we'll get
  back to the action.
YASSER SEIRAWAN:  Yes, sir?
AUDIENCE MEMBER:  Just a comment.  You were talking about the
  comment 10/10.  The second number is the maximum ply including
  extensions.  So, in other words, it might search to 10 ply and
  then it sees a forcing check line where it can check the
  opponent, say eight more ply, that would go to 18.  In this
  case the deepest it's going is 32 ply.  It's a special -- in
  some special situations.
YASSER SEIRAWAN:  So just to be clear that I understand you, the
  10 -- the first number refers to all the possibilities up to
  ten plies.  Fritz is analyzing all possible moves five moves
  ahead.  But the 32 means that in one of its extensions is
  looking 16 moves deep?
AUDIENCE MEMBER:  Yes.
MAURICE ASHLEY:  Thank you very much for that explanation.  We do
  have a slight error on our screen to the left.  It says move 10
  was Bd4, when in fact it's Bb4+ Nd2 and h5.
And I tell you, Yaz, some of the moves have been really strange.
  To my eyes, I was wondering, for example, before this, after
  Bb4+ Nd2, it seemed to me that after white castles at some
  point, let's say after h5 white were to castle, it seems to me
  white would be able to get in this c3 push without worrying
  about the d3 pawn and then an automatic d4 push because that
  pawn has protection, and black would be in a little bit of
  trouble, being pushed back with these center pawns.  So this
  move Bb4 to my eyes is sort of monkey see check, monkey do
  check, by the way.
GK MOVE:  12 Qe2
DB MOVE:  12...Qc7
MAURICE ASHLEY:  Protecting the knight.  Now it just seems, maybe
  I'm wrong, now it just seems as if c2-c3 can be played and then
  after some bishop move, let's say back to e7, d2-d4 can be
  played -- d3-d4, excuse me.  And now this knight has to drop
  back.  I don't know which square is best.  Maybe this square is
  best, but it seems these pawns have moved up the board with
  tempo, control of the board has been assured by white by his
  central pawns.  And I'm not really sure what this move h5
  does.
YASSER SEIRAWAN:  Okay.  Let's talk about that for a moment
  because this is the whole point of what IBM's concept is.  You
  have this incredibly difficult task called chess.  With so many
  possibilities, how on earth can we even get a computer to even
  play a decent game of chess?
But how's this for a fascinating question?  What if we can make a
  computer that by looking at all of these different chess moves,
  comes from a completely unconventional and is able to teach
  Grandmasters something about their game?  Well, this would be
  shocking.  A game that is centuries old, millenniums old, and
  to teach people something about their game of chess through
  unconventional means.
What we have here is a perfect example of what IBM's programmers
  are trying to do.  We can question, as humans, the judgment of
  Deep Blue.  Because he's doing something that is completely
  unnatural to strong players, like Maurice and myself, we
  wouldn't consider these moves h7-h5.
Now, what Deep Blue says to itself is, "Hey, this move seems to,
  c3, the moves d3-d4 are coming, pushing me pieces away.  I've
  God to do something about my knight on e5.  Let me figure out
  what I can do."
And it says to itself, "Huh.  How about that.  White has a weak
  square f4.  Of course, white does not have a weak square f4.
  And Deep Blue says that's the square I'm going off.  This is
  what I'm going to do, folks.  I'm going to play h7-h5.  And I'm
  going to play h5-h4."  So let's give Garry Kasparov a move.
Let's give Garry Kasparov the move castles.  New variation.  Now
  we play the move h5-h4.  After Garry plays g3-g4, bypassing the
  h-pawn, then Deep Blue says, "Aren't I brilliant?  I now
  control the f4 square.  I'll play Ne5-g6 and when Garry does
  something silly like c2-c3, well, then my bishop will drop back
  to d6 and I'll have played perfectly, I've got control of the
  f4 square."
So this is the way Deep Blue is trying to solve its problems.
What I had suggested the move Bc5, I thought that that was better
  than the move Bb4+ precisely because of the problems that the
  computer now faces, which is c3, attacking the bishop, and d4,
  attacking the knight.  So this is what Garry is now thinking
  about.  Is the move h5-h4 bothersome?  And if it is, well,
  white can always play the move h3-h4.
But then Deep Blue solves its problem in another way.  It now
  says, "Oh, it's not the f4 square that's weak, it's the g4
  square that's weak."
MAURICE ASHLEY:  Well, what if Kasparov tries to handle all of
  that and plays the variation beginning c2-c3, and after the
  bishop drops back, plays d4, and after the knight drops back,
  now plays your move, h4, and that knight didn't find the h4
  square and didn't find the g4 square.  In fact, it's found no
  square.  What's the deal with that?
YASSER SEIRAWAN:  Then Garry would be better!  (Audience
  laughter.)  And the strategy wouldn't work.
MAURICE ASHLEY:  Well, that seems like a very simple variation
  for Kasparov to play.  Maybe there's a response that we haven't
  noted, and I'm sure there are other moves we have not looked
  at.  Deep Blue is looking at all the moves in a given position,
  but this has come as a surprise to Kasparov and he's going to
  work out exactly what he feels is the best way to deal with the
  situation, and I have to say, I'm very suspicious of the some
  of the moves Deep Blue has played.  But as we've seen in so
  many games, Deep Blue has played some wacky looking moves?
YASSER SEIRAWAN:  Absolutely.  And he's gotten away with it, too.
MAURICE ASHLEY:  And he's gotten away with it.  This thing mixes
  it up like anybody.  The game gets wild, game gets sharp, Deep
  Blue sacks is piece or a pawn, and then Kasparov gets
  confused --
YASSER SEIRAWAN:  And then he gets angry.
MAURICE ASHLEY:  -- and we've seen that time and time again.
  Deep Blue has played some suspicious -- as far as we're
  concerned, the human side, strategical decisions, for its
  tactical ability, time and time again --
YASSER SEIRAWAN:  Saving the game.
MAURICE ASHLEY:  -- turns out fine.
YASSER SEIRAWAN:  Absolutely.  Throughout the match, three or
  four games, I think the match score is right, 2-2.  But quite
  frankly Garry has had the edge in three out of the four games.
  And sometime the comput