Reading a commentary by Takao Shinji, I was struck first by an expression he used. That in turn made me think about the move more than usual, at which my surprise increased. My flabber was gasted even more when I related it to some of the things we have been talking about here recently. One of those things was Dieter's "slow connections".
First, the background. The game was between young Mito Shuhei 1-dan, who was playing Black against big beast Hikosaka Naoto 9-dan. This was in the new series in Go World in which young players in the national squad present a radar chart of their various attributes, and a squad coach (Takao here, but it varies month by month) reviews that chart by reference to a game.
Mito thought his strongest suit was positional judgement and his worst was the endgame. He thought his opening and whole-board vision were average. He may wish to revise that after seeing Takao's comments. Takao was not concerned with criticising individual moves as with addressing problems with mindset. In particular, he decided the main theme was Mito's predilection to make prophylactic moves (mamori) too early.
The first example was Black 1 below (39 in the game).

- Click Here To Show Diagram Code
[go]$$B
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . b O X . . |
$$ | . . . O . . . . . . . . . . O X . . . |
$$ | . . . , O . O . . , . . . . . , X . . |
$$ | . . X . O X a . . . . . . . . O X . . |
$$ | . . . . X X O . . . . . . . . O . . . |
$$ | . . . X . X . . . . . . . O O X X . . |
$$ | . 1 . . O . . . . . . . . O X . . . . |
$$ | . . O . . O . . . . . . . . X . O . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . O . . . |
$$ | . . X . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . O O . . |
$$ | . . . X . . . . . , . . . . . , X . . |
$$ | . . . . . . . . . . . . . X . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]
Takao remarked that it's not good to play down towards the edge so early in the game. That's what surprised me initially. Talking about playing
on or
near the edge would not have surprised me. It was the
down that caught my attention.
But once he had my attention, I thought "hang on a mo". It's long been standard teaching that you should think about getting a base early, and it's something amateurs neglect horribly. But, if anything, AI is now telling us that even pros should be trying to settle their groups even earlier than they do.
On top of that, this move is concerned with the base of two opposing groups. If I saw this position in a book of next-move problems, I would choose Black 1 without further thought.
Hmmm. So I did a bit of further thought, and this confirmed I was "right". I wouldn't want to play 'a' because that just helps White settle his territory at the top. I would want to hold off on 'b' (tasty though it looks) for similar reasons.
Still puzzled I decided to visit Lizzie. Blow me, she said the best move was a toss-up between 'a' and 'b', but with 'b' being favoured, and Black 1 did indeed lose a couple of percentage points. Takao's preference, incidentally, was for 'a'. His terminology was interesting there, too. His recommended sequence was as below:

- Click Here To Show Diagram Code
[go]$$B
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . O X . . |
$$ | . . . O . . . . . . . . . . O X . . . |
$$ | . . . , O . O 2 4 , . . . . . , X . . |
$$ | . . X . O X 1 3 . . . . . . . O X . . |
$$ | . . . . X X O . . 5 . . . . . O . . . |
$$ | . . . X . X . . . . . . . O O X X . . |
$$ | . . . . O . . . . . . . . O X . . . . |
$$ | . . O . . O . . . . . . . . X . O . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . O . . . |
$$ | . . X . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . O O . . |
$$ | . . . X . . . . . , . . . . . , X . . |
$$ | . . . . . . . . . . . . . X . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]
The way he described this was "teatsuku kamaeru". For thickness aficionados, note the teatsui rather than atsui, and kamaeru refers to making construction moves, or making frameworks if you will, but on a smaller, tighter scale than moyos. Hata o motsu is an example, and that could be applied here. See my Go Wisdom if you need more detail such as what the te- adds, but the main point to take away from here is that Takao said
this is the way to attack. Or more precisely, he said this was the way to
aim at an attack. Up instead of down? But it was at this point that I remembered Dieter's slow connections idea, and wondered if there's a similar distinction slow attacks and fast attacks (but in reverse: slow attacks are better?). Over to the reader on that.
Takao's next example was Black 1 below (Black 49 in the game). But note that the so-called endgame move in the top right had been played already by now. This is properly yose in Japanese, because it concerns boundary plays. The idea is not 'approach' as someone said elsewhere (and a reason I don't like 'approach ko' for yose-ko). The -s- is the important letter, denoting a causative or permissive sense. Simply 'approach' would be yoru, where -r- is the important letter. We see the same distinction in amari (e.g. amarigatachi) and amashi. So the true idea is causing (or allowing) an approach. But I find the best analogy in go is that of pulling the draw strings of a bag or purse to make the opening smaller. You can pull it part way while you are still shopping, to minimise the chance of the territory you just bought falling out, and when you've finished shopping - the endgame - you can pull it tight. I mention this not to make the point for boundary lays over endgame yet again (though I'm far from averse to doing that), but because I'd like to ask whether other people have my impression that bots like to make these early boundary plays even earlier than human pros, in the same way that bots seems to like to settle groups earlier. A possible example is the diagram above is the bot choosing 'b' whereas Takao went for 'a'.
(Incidentally, the word used for endgame in the radar charts is not yose but shuuban.)

- Click Here To Show Diagram Code
[go]$$B
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . X . . . |
$$ | . . . . . . . . . . . . . O X . X . . |
$$ | . . . O . . . . . . . . . . O X . . . |
$$ | . . . , O . O O . O . . . . . , X . . |
$$ | . . X . O X X . . . . . . . . O X . . |
$$ | . . . . X X O . . . . . . . . O . . . |
$$ | . . . X . X . . . . . . . O O X X . . |
$$ | . X . . O . . . . . . . . O X . . . . |
$$ | . . O . . O . . . . . . . . X . O . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . O . . . |
$$ | . . X . O . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . O . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . O O . . |
$$ | . 1 . X . . . . . , . . . . . , X . . |
$$ | . . . . . X . . . . . . . X . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]
Again the focus should rather be on "slow attack" (my words), and the way the game should pan out here according to Takao is as below:

- Click Here To Show Diagram Code
[go]$$B
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . X . . . |
$$ | . . . . . . . . . . . . . O X . X . . |
$$ | . . . O . . . . . . . . . . O X . . . |
$$ | . . . , O . O O . O . . . . . , X . . |
$$ | . . X . O X X . . . . . . . . O X . . |
$$ | . . . . X X O . . . . . . . . O . . . |
$$ | . . . X . X . . . . . . . O O X X . . |
$$ | . X . . O . . . . . . . . O X . . . . |
$$ | . 5 O . . O . . . . . . . . X . O . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . O . . . |
$$ | . . X . O . 7 . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . O 1 . . . . . . . . . . . . . . . |
$$ | . . 2 3 . . . . . . . . . . . O O . . |
$$ | . . . X . . . . . , . . . . . , X . . |
$$ | . 4 . . . X . . . . . . . X . X . . . |
$$ | . . 6 . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]
He said this was more "positive". Even I can see that, at least, Black has nothing exposed to attack but White does, and Black has 2.5 corners to White's 1.5.