What is "the direction of play?"

shapenaji · **#41**

Bantari wrote:

Hehe... The triangle is bigger, but the area is not necessarily so... And is it really bigger, or only visually so? Hat off to geometry, no? And - what does it mean and why should it be good that the shape itself is 'bigger'?

Base and Height are both 1 unit smaller.

However, the real reason I think the larger one is better is that you have 2 lines of "time" in which to confine an invasion rather than just 1.

Bantari · **#42**

shapenaji wrote:

Bantari wrote:

Hehe... The triangle is bigger, but the area is not necessarily so... And is it really bigger, or only visually so? Hat off to geometry, no? And - what does it mean and why should it be good that the shape itself is 'bigger'?

Base and Height are both 1 unit smaller.

However, the real reason I think the larger one is better is that you have 2 lines of "time" in which to confine an invasion rather than just 1.

Hey, this is actually a pretty good reason! Thanks.

To be a devil's advocate, how about this then:
More stretched triangle offers less space to confine invasion, but, since the area is less condensed, a successful invasion can potentially do less damage?

Tryss · **#43**

I guess, you could explain it like that: Put a stone as an extension, draw lines, and see what area is bigger :

Click Here To Show Diagram Code: [go]$$c $$ --------------------------------------- $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . O . . . . . . . . . . . . . . | $$ | . . O , . . . . . O . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . X . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . X . . . . . , . . . | $$ | . . . X . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ --------------------------------------- $${LN A5 C5} $${LN C5 K4} $${LN K4 K1} $${LN A16 K16} $${LN K16 K19}[/go]

40 intersections on the top
43 intersection on the bottom

Obviously, it's kinda crude reasonning, but it also explain why the komoku direction is in the 4 line direction, and not the 3rd line direction

daal · **#44**

Bantari wrote:

John Fairbairn wrote:

Quote:

My current understanding of the word "direction" in this context is: "The place where the stones currently on the board would prefer to develop."

But again this is changing Kajiwara's words, so it's no surprise if his message does not get through.

I lost track of all that, heh...
Are the words in question, the quote, directly from Kajiwara, or is it something John came up with, or the translator?

And, in either case, are we interested what Kajiwara means (or would mean) by that, or what was that John meant when he quoted what he quoted? Was it even a quote, or not?

I think all this needs to be cleared between daal and John first. Or at least - it would help *me* understand what is going on here.

It might all be repetitive, but I really lost track, and my head is buzzing from all kinds of other threads now...

Mine too

The impulse for this thread came from this quote:

John Fairbairn wrote:

But someone did make a comment recently about there being no books on go psychology, and I found three at once.

The most "fundamental" of these was a Go Super Book, Go Psychology for Fun by Mihori Sho. It has a chapter relevant to a couple of current threads here, on the psychology of mastering go theory. The most fundamental point was said to be deciding where to play and for this the essential mindset to ingrain was "First, direction of play; second, the actual point". Ignoring this is the commonest reason amateurs are admonished by professionals, apparently.

It sounded to me like getting this mindset might be a good idea, so I got Kajiwara's The Direction of Play off the shelf, in the vain hope that I might get a handle on this apparently common concept and be able to start thinkin' right.

shapenaji · **#45**

Bantari wrote:

Hey, this is actually a pretty good reason! Thanks.

np, always the way I thought about it

Quote:

To be a devil's advocate, how about this then:
More stretched triangle offers less space to confine invasion, but, since the area is less condensed, a successful invasion can potentially do less damage?

Well, it's true, if you have less to take, they can't take as much

TheBigH · **#46**

I've always thought of 'the direction of play' as encompassing more than just the physical location of the important points, or which area of the board the next fight will spill into. It includes higher-level evaluations of the situation like, is this turning into a big moyo based game, or a series of small skirmishes.

Bantari · **#47**

Anyways, for Ed and all others who wonder: can Bantari be ever satisfied with any answer? I say: yes, I can. And Shapenaji's post is an example of that.

Tryss' diagram is also a good try, although slightly weaker - the original diagram Bill presented had the 'squishy' triangle's extension stone(s) one line further - which would make both areas exactly same: 43 points. Each subsequent poster conveniently moved the extension stone(s) one line closer, not sure why.

I rejected topazg's numbers and counting because it is uncertain what is the correlation between the size of the triangle with the actual points it makes, so its really not very meaningful - until such correlation is indicated.

To iterate what I asked for: some kind of common sense explanation, does not need to be full, does not need to be 100% correct, just something showing a spark of understanding, something beyond simplistic "just because!" I swear, I *am* reasonable.... mostly.

jts · **#48**

I would rather see diagrams comparing the enclosures like so:

Click Here To Show Diagram Code: [go]$$B $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , W . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . X . . . . . . . . . . . . . O . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . X . . . . . B . . . . . O . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$------------------------------------------ $$ {LN D3 C5} $$ {LN D3 D1} $$ {LN A5 C5} $$ {LN K1 K3} $$ {LN K3 D3} $$ {LN Q3 R5} $$ {LN Q3 Q1} $$ {LN T5 R5} $$ {LN R5 R10} $$ {LN R10 T10}[/go]

I think this shows more clearly what's wrong with Ed's "tray" - to a first approximation* the corners are the same, but the black box is longer than the white tray and contains three more points along the side. Well, can we play Bill's "tray" instead, moving white out a point?

Click Here To Show Diagram Code: [go]$$B $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . W . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . X . . . . . . . . . . . . . O . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . X . . . . . B . . . . . O . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$------------------------------------------ $$ {LN D3 C5} $$ {LN D3 D1} $$ {LN A5 C5} $$ {LN K1 K3} $$ {LN K3 D3} $$ {LN Q3 R5} $$ {LN Q3 Q1} $$ {LN T5 R5} $$ {LN R5 R11} $$ {LN R11 T11}[/go]

Now the boxy black position and the white tray position have equally long extensions, but the problem seems clear to me - white's extension is too long, and black can (depending on the board) invade or reduce with impunity. But if the two extensions are the same length, how can one be safer than the other? Well, C5 is well-placed if white starts a fight around G3, but Q3 doesn't contribute much to an attack on a black move around R8. Since white is a move behind, the R11 tray is a harder extension to defend than the K3 boxy position.

Or, at any rate, that's how I think the position should be explained. But I don't really see this as being about the direction of play. These are basic ideas about efficiency in the opening. They're not too different from explaining why we finish of 4th line moves with an extension to the 3rd line, but treat 3rd line positions as settled and take sente; or why we use a two space extension along the side to make a base, rather than a two space jump up into the center.

Once we know what sort of moves are likely to be efficient/big in the opening, then we have the building blocks for what Kajiwara describes as the "direction of play" in the book, but only in the same way that the nakade shapes (heh, two solecisms for JF in the same sentence!) are the building block for life and death techniques. Knowing that points A, B, and C are efficient is only the introduction to the topic... it seems to me that whatever direction of play is, it involves using that knowledge to develop a plan for the entire opening (or possibly, more than one plan).

*(In fact I think B's corner should end up better because B benefits more if W probes the B's corner than vice versa.)

daal · **#49**

Nobody took me up on this one. Last chance!

daal wrote:

Here is a diagram from the book I would like to offer up for discussion (beware, it's a trap):

Click Here To Show Diagram Code: [go]$$B White to play - how should black respond? $$ +---------------------------------------+ $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . X . . . | $$ | . . O . X . . . O . . . . . O X . . . | $$ | . . . , . . . . . , . . O . O , X . . | $$ | . . X X . . . . . . . . . . . X . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . O . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . X . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . O . X . . . . . . . . . . . . . . | $$ | . X X . . . . . . . . . . . . . O . . | $$ | . O X X X . . . . , . . . . . , . . . | $$ | . O O O . O . . . X . . . . . O . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ +---------------------------------------+[/go]

Wrong. Kajiwara writes: "In relation to the corner enclosure White's checking extension at :w1:

is ideal. But this betrays thinking about only the local instead of the overall situation. Even if White checks at :w1:

, Black, having already defended at :bt:

, has little to fear from a attack on his lone stone on the lower edge.
:b2:

on the right-side star point is an ideal invasion which utterly spoils White's dream of monopolizing the right side. Indeed, :wt:

is now under attack and the initiative has passed to Black. Thinking locally and forgetting the overall position is tantamount to being blind to the direction of play."

The italics are mine, indicating what I believe to be the essence of direction of play.

Click Here To Show Diagram Code: [go]$$W Wrong $$ +---------------------------------------+ $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . X . . . | $$ | . . O . X . . . O . . . . . O X . . . | $$ | . . . , . . . . . , . . O . O , X . . | $$ | . . X X . . . . . . . . . . . X . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . Q . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . X . . . . . . . . . . . 2 . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . O . Y . . . . . . . . . . . . . . | $$ | . X X . . . . . . . . . . . . . O . . | $$ | . O X X X . . . . , . . . . . , . . . | $$ | . O O O . O . . . X . 1 . . . O . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ +---------------------------------------+[/go]

Thus, the correct move for White is:
How should black respond?

Click Here To Show Diagram Code: [go]$$W Correct. And now, how should black respond? $$ +---------------------------------------+ $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . X . . . | $$ | . . O . X . . . O . . . . . O X . . . | $$ | . . . , . . . . . , . . O . O , X . . | $$ | . . X X . . . . . . . . . . . X . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . O . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . X . . . . . . . . . . . 1 . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . O . Y . . . . . . . . . . . . . . | $$ | . X X . . . . . . . . . . . . . O . . | $$ | . O X X X . . . . , . . . . . , . . . | $$ | . O O O . O . . . X . . . . . O . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ +---------------------------------------+[/go]

Wrong again. Kajiwara calls this move "highly dubious" because it does not take into account that white is expanding the right side. Afterwards White can cap at "a", further expanding the right.

Click Here To Show Diagram Code: [go]$$W Wrong $$ +---------------------------------------+ $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . X . . . | $$ | . . O . X . . . O . . . . . O X . . . | $$ | . . . , . . . . . , . . O . O , X . . | $$ | . . X X . . . . . . . . . . . X . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . O . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . X . . . . . . . . . . . 1 . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . O . X . . . . . . . . . . . . . . | $$ | . X X . . . . . . . . . . a . . O . . | $$ | . O X X X . . . . , . . . . . , . . . | $$ | . O O O . O . . . X . . . 2 . O . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ +---------------------------------------+[/go]

Correct. The move to impede White's expansion.

Click Here To Show Diagram Code: [go]$$W Correct. $$ +---------------------------------------+ $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . X . . . | $$ | . . O . X . . . O . . . . . O X . . . | $$ | . . . , . . . . . , . . O . O , X . . | $$ | . . X X . . . . . . . . . . . X . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . O . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . X . . . . . . . . . . . 1 . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . O . X . . . . . . . . . . . . . . | $$ | . X X . . . . . . . . . . . 2 . O . . | $$ | . O X X X . . . . , . . . . . , . . . | $$ | . O O O . O . . . X . . . . . O . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ +---------------------------------------+[/go]

There you have it in a nutshell.

snorri · **#50**

I find it interesting that the examples given are mostly about expansion of stones of the same color. But there is also the question of whether it is interesting for the opponent to play in the same areas. IMHO, whether or not it is interesting to the opponent takes into consideration not just preventing expansion, but also whether playing there would be playing near strength. A complete isolation of the concept from sente and gote may be wishful thinking.

Think for a bit about the difference between extensions from a low enclosure vs. a high enclosure. The primary and secondary directions of development are the same, but an extension in the secondary direction from a high enclosure has a slight defensive feel of urgency to it that is not so much present for the low enclosure. Extending in the secondary direction from a low enclosure has a sense of overconcentration unless it's pretty far.

What is "the direction of play?"

Who is online