30
2009
Interesting position for 1400 players
So I was playing against Shredder for iPhone, on the black side of my typical Alekhine Defense. The game was not excellent, but I did fairly well against the 1400-rated Shredder and got to this position after white played 48. c6:
            |
It looks like, despite Black’s Herculean effort to chase the white king all the way to the back rank, a queening is imminent; and strictly speaking it is. However, I found a nice move to save the situation: 48. ..Qa3!. Strange-looking (in fact, Shredder’s “coach” popped up saying that my move is a blunder and offered a takeback), but if white marches on with 49. c7, then ..Kf8 threatens mate next with ..Qd7, and once white’s king runs away, black has plenty of time to relieve the pressure.
An interesting problem, at least for players around my level.
8
2008
Several updates
There are a couple of updates and add-ons to make for previous posts which are too small to merit their own entry, so I’m putting them all together.
#google-dev
Firstly, this morning Google announced their new product, Google App Engine, wherein they provide hosting and a rich Python web application engine, and everyone else provides the apps. Although the 10,000 demo slots for the service were taken in a matter of minutes, the SDK and local webserver are freely available for experimentation, and I’m certain that more slots will open up very soon.
The reason this is relevant is because just a few days ago, I blogged about #google-dev, an IRC channel for Google development. I swear I had no idea this was coming — it was just my good luck. The way I see it, the channel’s utility just quadrupled, so once again, please come and idle, share your expertise, and ask questions.
Shusaku Documentary
For anyone who’s interested, there was quite a bit of demand for the actual video documentary, so I posted up a torrent last night; quite a few GD’ers got in on it, and at the time of writing there’s 11 seeds, but I’m not sure how long that will last. So if you want it, you should grab it as soon as possible.
EidoGo Extension
This is kind of old news by now, but Firefox 3.0b5 was released last week and so the Send to EidoGo extension has had to be updated. Also, I’ve fixed a bug (DGS game handling) and removed some extraneous code from the .xpi (an external .js file that I had migrated to inline code but forgotten to delete). The latest version is 1.1 and you can get it simply by reinstalling from the same link I had up originally, linked above.
Oh, and also, the blog has been updated to WordPress 2.5 and has an obviously different theme.
7
2008
Japanese documentary - 伝記の碁打ち本因坊秀策
So I recently obtained something that most Western Go fans have probably not seen yet — an NHK documentary about the life of Honinbō Shūsaku, the player which (as the show’s own subtitle puts it) “even beginners have heard of”.
Well, the whole video is in Japanese, but between my scanty knowledge of Japanese and a bunch of technical terms, my knowledge of Shusaku’s life, and the occasional help of my girlfriend, I’ve pieced together the basic storyline and most of the text. To be honest, they didn’t make it very hard.
Of course, it begins by looking at modern Go; they show the token game between an old grandpa and the cute 7-year old boy to emphasize how Go can break the generation barrier. Even Hikaru no Go is briefly mentioned and displayed, though (as far as I can tell), they neglect to mention that according to Hikaru no Go, Honinbo Shusaku was a colossal cheater who got all his moves from the thousand-year-old spirit of a dead master. Perhaps the most interesting segment of this part of the documentary is a match between two little girls, one of whom is receiving a brain scan as she makes her moves. A monitor highlights her brain activity. I couldn’t understand what their conclusions were, but I thought that was neat nevertheless.
The next section deals with Shusaku (”Kuwahara Torajirou”) during his childhood in Innoshima, Hiroshima. It follows his fascination with watching his parents play. It even includes a dramatization of an anecdote that I’ve read before; his angry (seemingly drunken) father is angry with him and punishes him by locking him in a cupboard which also happened to contain a Goban. After a few hours when his mother opens the cupboard door, Shusaku is quietly enthralled in studying his own positions on the board. If not for Invincible I probably wouldn’t have been able to follow this section; we see a fat Lord Asano playing a game with the very young boy, an approving smile on his face.
He is taken under the wing of a professional player and we see a cute montage of the young Shusaku performing various menial tasks for his mentor. At night they play, his mentor constantly wearing that awed mask of approval that all the adults in this documentary feel obligated to wear whenever they are in his presence.
At the point the documentary jumps to an animated 3D board to explain — the rules of Go. Personally I find this unnecessary; any Japanese viewer who’s watched this far into a documentary about a Go player who’s been dead for hundreds of years presumably also knows the rules of the game; and if they don’t, a ten-minute introduction isn’t going to much prepare them for understanding the life of Shusaku. Unfortunately this first tutorial segment marks a constant habit of the documentary to explain away complicated features like the ear-reddening move or the Shusaku fuseki using arrows that make it look as if Go stones sail across the board in parabolic arcs, and territory maps that are painfully inaccurate. Most of the time the documentary makes it look as if most professional moves are motivated by the need to stay out of atari for as long as possible. At one point the virtual board is demonstrating a 3-4 joseki, and of all the variations they could have shown, they decide to show why white cannot cut a bamboo joint.
Next, the documentary shows off one of Shusaku’s more famous games, played in 1839 against Itagaki Chuzo, when Shusaku was around 10 years old. Interestingly the documentary claims the game ended 72-70 for Black; however, every other source I’ve seen for this game (including the one linked above) simply says white resigned after Black 191. Maybe these producers had access to records we don’t, but I think it’s perhaps more likely that they thought a score would be more impressive, so they simulated a likely endgame and counted the score.
Next it takes a short visit to modern-day Innoshima, Shusaku’s birthplace. The Hikaru no Go fans will recognize a few of these destinations as being photo-identical to the scenes depicted in the anime when Hikaru is frantically searching for Sai. There are some nice shots of some of Shusaku’s calligraphy, his old board, and his shrine. There is an old lady there who seems to be leading the cameraman through the place, but my Japanese was not good enough to tell what relation she had to Shusaku or his shrine.
Next it goes through a few more of his games against various top players of his day. Then, of course…
What exposition on Honinbou Shusaku would be complete without a dramatic treatment of the famous ear-reddening move? The show the beginning of the game, including a step-by-step demonstration of the taisha joseki in the bottom-right which Shusaku messed up on. As far as I can tell, they never mention that Shusaku actually made a mistake in that corner, and indeed since their ‘analysis’ of the joseki involved mostly counting liberties to see who was further away from atari at any given moment, Shusaku seems to have actually performed admirably there since he captured more stones there than white 
When the moment finally arrives, however, they hold back no punches in explaining just how epic the move was; in true Hikaru no Go style, we see a reflection of the board glint in his eyes, his hand reach for the stone, cock it back up to his ear as if loading a shotgun barrel, and triumphantly slamming it down at 10-9 (this last part is replayed three times for effect). A glorious chorus of trumpets greets this brilliant stroke of genius, and Gennan Inseki’s ears become visibly red. My Japanese was just good enough to confirm that they kept to the anecdote of the doctor in the audience stating “I don’t know much about Go, but when Shusaku played that last move, the Master’s ears turned red, which told me that he was upset.”
It also mentions the castle games and plays through one of those matches (I did not recognize it). Finally, it shows the effect of the epidemic on the Honinbou House, including a touching scene of either Shusaku tending one of his students, or vice versa (it’s hard to tell with the poor lighting and the fact that they’re all bald). Finally, the two aesthetic shots shown above and some nice music ends it.
I think it’s an interesting video for Go players, particularly those for whom the culture and history of the game plays an important part in its appeal. I warn you that the actual technical Go content is plentiful but very light and in some cases just plain wrong. However I expect that a few people will be interested in getting a copy, especially once I post this on GD, so let me know through e-mail and we’ll figure out a way to transmit it.
All in all, the video runs for about 1 hour and 50 minutes, and should be enjoyable even for someone with no knowledge of Japanese.
4
2008
Kanata Chess Club
Lately on Wednesday evenings, I’ve been going to the Kanata Chess Club. I’m among the weaker players there, but that just means there’s more to learn. Yesterday I played three games against a much stronger player, and lost all three (though I had a winning position in one of them when my clock ran out). The fourth game was against someone else, and I won it. Here it is, below.
When I saw 2. Nf3 I was sure it was a mistake; but according to Chessbase, there have been a few dozen cases in professional games, and the most common response is 2. ..Nf6, but this makes little sense to me; after 3. Nc3 Nf6 4. Bb1, this is almost like a transposition to the Ruy Lopez, which seemed to me to be too good for what I perceived as an opening “mistake.” So I pushed the pawn.
After 3. ..d5 I felt quite good about myself. The e4 pawn was keeping his king’s bishop’s and queen’s pawns down, and I could aim for a four-pawns attack. Chessbase finds two modern games that proceeded this way, and both continued with 4. Nc3, after which I probably would’ve been happy with 4. ..c5, but instead I got the very surprising 4. h3?, which, needless to say, has not been seen before.
I think 8. ..fxe4 was my first real mistake; it really should have been 8. ..dxe4 to prevent the rather disastrous-looking 9. Qxd5 which happened in the real game. After 8. ..dxe4 it might go 9. Bd2 Be6 10. gxf5 Bxf5 11. Qh5 Be6, as below:
         |
Black’s pawn structure is just as bad as in the real game after 9. Qxd5 but at least I don’t have the disastrous Qxb7 to worry about. Still, materially I am up a knight already, so there’s not too much to complain about.
This led to some difficult fighting on the left side (well, right side for me) which was finally diffused with 23. ..Nxc2+. This was a good way to end the skirmish for me, having kept my minor piece advantage, gotten the short castle, and Nxc2+ prevented white from getting any castle at all. He replied with the suspect 24. Ke2; better would have been 24. Kd2 threatening the knight, which would gain a tempo as I withdraw with 24. ..Nb4 25. Kc1 Rab8 27. g5 (which is now safe because 27. ..Qxg5 28. Rg1 pins the queen now that the King is safely on the other side):
  |
This is much nicer than what happened in the real game, especially after the added blunder of 25. Rb1 which led to the more-or-less forced sequence 27. Nc4 Rxb1 28. Rxb1 Bd5 29. Qc3 Bxc4+ 30. Qxc4 Na3 forking the queen and rook, upgrading my material advantage from a knight to a rook. Entering the endgame ahead by a rook made me pretty confident that my position was won, and 33. f3 confirmed it, leading to a series of checks that destroyed white’s pawn island on the right, his last hope, and I won.
The game was a lot of fun, and probably the best part is that my opponent offered me a ride home at the end of the stay, which was quite welcome. 
2
2008
#google-dev
I’ve been playing around with various Google APIs lately, and since Web programming is definitely not “my thing,” I’ve had lots of difficult-to-answer questions along the way. This is not to say that Google doesn’t document them well (they do), but rather that for some things, nothing beats a live guru to talk to.
My usual approach is to connect to FreeNode’s IRC server, and join the appropriate channel. Well enough, except that #google proudly states
Note: This is not a support channel
What on Earth, then, I asked myself, are you, and why do you do whatever it is you do on FreeNode? Well, a channel member tells me after 20 mintues of total silence, “idlers” and “DALnet sucks” respectively.
It turns out that FreeNode simply does not have a support channel for Google APIs anywhere on it. None. This is coming from the network that hosts a support channel for a Linux distribution aimed at car dashboards and #ubuntu-de-offtopic, a channel specifically for overflow discussion for German Ubuntu developers.
Anyway, I’ve decided to try and fix this glaring omission in the FreeNode canon. Since #google was already taken by the non-supporting-idlers, I’ve gone and registered #google-dev with Chanserv. The connection information, for those who’ve never been on FreeNode before, is:
Server: irc.freenode.net 6667
Channel: #google-dev
If you have even the slightest inkling of interest in Google technologies, please come and idle — the more random people in the channel the more credible it is. With a bit of luck (and advertising on the appropriate Google Groups), we may even get some actual Google employees and/or gurus to help out. As for me, I have a passing familiarity with the GData client libraries (particularly the Python version), but don’t expect much more than that out of me. When it comes to web programming, I’m much better at asking questions than answering them 