Life In 19x19

I hope this is the section to post this. It doesn't quite feel to belong to Beginners. I'll be using the 6th printing of the book, 2007.

Click Here To Show Diagram Code

[go]$$c Now... on dia 10, page 31
$$ ------------------
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . O . . . . |
$$ . O . O X X . . |
$$ . . O X O . c . |
$$ . . . X O . d . |
$$ . . . X a b . . |
$$ . . . . . . . . |[/go]

Kageyama (and my handheld Leela) recommends 'a'. Then 'c' answered by 'd'. As far as can see, this gives White the NE corner in exchange for his stones, influence and sente. However, Black at 'b' settles the matter (Leela doesn't want to play anywhere close; like, really doesn't wanna) and doesn't surrender, BUT losses sente. With the game prepared so that I could evaluate the position (stones of same color on opossite hoshi, a high 2-space pincer on NW White), the difference was about 10%, IIRC, in favor of closing at 'a'.

My question is... is that difference due only to sente (and maybe the lack of defects facing South) or am I missing a lot more?

Thank you. Take care.

Is there supposed to be a black stone at n17? Also if white plays c, black doesn't need to play the "slapping tesuji" at d but can just block at s16 if he doesn't want to let white get the corner benefit as white still dies if he gets to bamboo.

As to why thick turn instead of net, it is "firmer": it takes liberties away from white's stone and gives liberties to black. The net also leaves white a potentially useful peep at q12, or probably more useful the potential to exploit the p13 stones only having 4 liberties (with the turn they have 5), for example white attach at o13.

I think Kageyama just constructed this position to discuss the difference between playing a and b in terms of how we grasp the two White stones. I find nothing like it in real world positions in GoGoD. Hence we need to be careful about reading too much into the situation as a whole.

See viewtopic.php?p=256582#p256582

Ferran, I don't see any sente difference.

Robert Jasiek said:

Kageyama meant to explain what is explained explicitly in Joseki 1 Fundamentals, p.245:

"12.4.2 How to Play Nets

A net must take a firm grip on the captured string. It should leave the least aji. This means that the opponent has the fewest number of forcing moves related to the captured string.

This implied criticism of Kageyama - highlighted by being parked in another thread -is quite unfounded.

What Kageyama said (in Japanese) was "The answer is 'a'. If White tries to escape at 'c', Black stops him at 'c'. By now, I probably don't have to keep explaining the reason."

What James Davies wrote was ""The answer is 'a', If White tries to escape at 'c', Black stops him at 'c'."

So Kageyama was quite explicit. He had no intention of explaining again what he had already spent 11 pages explaining (entertainingly, too , be it noted). He didn't mess up.

I suspect the reason Davies (or Bozulich) omitted the last portion was to make the text fit the page - this was on the last line.

Is there supposed to be a black stone at n17?

No.

As to why thick turn instead of net, it is "firmer"

If this conjunction of thick and "firmer" is meant to imply these words overlap here, well, maybe they do to some degree, but that's not really what Kageyama had in mind. He used the negative of the verb yurumu = to be loose. He was saying choosing the less loose = firmer option is better.

John Fairbairn wrote:This implied criticism of Kageyama

Indeed. I needed years to figure out what he should have said explicitly so that then I could know for arbitrary nets how to identify nets as firmer.

- highlighted by being parked in another thread -

This is a consequence of the forum rules.

is quite unfounded.

Quite contrarily. Kageyama did not only show an example but his intention was to explain in general to choose the firmer net. Therefore, also his example should have explained it so that application could become general instead of being restricted to the example.

John, I suppose my focus when I used the adjective thick was on Black's shape towards the outside, whereas firmer focuses more on restricting White's liberties and scope for shenanigans inside.

Robert, I can't agree with your attempt to creative a rule for deciding which net is better that you could plug into djhbrown's attempt to create a rules based bot "This means that the opponent has the fewest number of forcing moves related to the captured string.". Simply counting the number of forcing moves is not sufficient, they also need to be weighted by how useful they are: move A which leaves the opponent 3 fairly useless forcing moves is better than move B which leaves them 2 useful ones. Imagine if that only 1 forcing move it leaves happens to be a ladder breaker for some really big ladder on the other side of the board. Of course judging what is useful is difficult and can change as the game progresses.

Uberdude, there is nothing wrong with further improving my principles by a global context of possibly playing elsewhere, assessing eventual points, assessing efficiency etc. My principles are designed for the assumption of not currently playing elsewhere. As such, my principles fall in the 90%+ category whereas Kageyama's less precise advice I found in the 60%+ category, often not knowing what "firmer" was sopposed to be.

I hope you do not misinterpret numbers of available forcing moves. We do not count the number of alternative forcing moves but we count the number of forcing moves a player can still get and are not bad exchanges (except for preliminary elimination of aji or ko threats if played prematurely).

The net A currently leaves 0 forcing moves that should be played. The net B leaves 1 forcing move that can become good at an appropriate time. 1 too many.

Therefore, also his example should have explained it so that application could become general instead of being restricted to the example.

But the point is, he did. He was talking to humans, not writing a computer program. Just because you choose to apply a non-fuzzy filter to a fuzzy world doesn't mean the rest of us have to.

A net must take a firm grip on the captured string. It should leave the least aji. This means that the opponent has the fewest number of forcing moves related to the captured string.

There's an example of why anti-fuzziness fails. Your definition talks about aji and forcing moves. That means you get to see only part of the image. Writers like Kageyama, and readers who follow them, understand there's a lot more. Apart from different kinds of forcing moves, consideration has to be given (as in any geta case) to related concepts such as fukimi, aya and "te ga aru". And, of course to different concepts such as liberties and thickness, and the possibility of sacrificing. Here, for example, a White move at S17 is more like te ga aru than kikashi. And that could matter big time. I acknowledge that you said your own definition is not watertight, but I think a claim of 90+% for you and 60% for Kageyama is totally the wrong way round. You are measuring a small portion of a bigger picture (and almost certainly invoking the uncertainty principle ). Kageyama is measuring almost the whole picture, down to its fuzzy edges. "Fuzzy-wuzzy was a bear, Fuzzy-wuzzy wasn't fussy wuzzy" is a good adage for most go players to follow. It's one way our human brains can work best in complicated situations. Stay flexible, head up; not preoccupied, head down, with fussy definitions. And remember what Lao Zi taught: what is not given substance is often more important than what is. Otherwise, where would we be without doors and windows? Stuck forever in lockdown!

Since he remained fuzzy, I tried every degree of fuzzy filter to possibly better understand him and it did not work. I saw many players with the same difficulties of not understanding this the fuzzy way but quickly understanding the explicit count of numbers of forcing moves.

Concerning my percentages, they have been my consistent experience from 1991 to 2020. The 90%+ is a modest estimate; for 95%+, I would need ro actually count instances to justify the pecentage.

Again, nothing is wrong with a broader picture of including more concepts to also get the last 10% right. However, just listing concept names does not do it.

For the non-Japanese readers, what are fujimi, aya, te ga aru? Even with fuzziness, I cannot decipher Linear A on such a small sample of phrases;)

Since he remained fuzzy, I tried every degree of fuzzy filter to possibly better understand him and it did not work.

I repeat: there is always a bigger picture. I wholly accept that YOU didn't understand it (or, more accurately, didn't like it?). I wholly accept that some other like-minded people didn't understand it. But the bigger picture is that an awful lot of people (the vast majority in MY experience) did understand it. That's one reason the book sold over 100,000 copies, I imagine.

For the non-Japanese readers, what are fujimi, aya, te ga aru? Even with fuzziness, I cannot decipher Linear A on such a small sample of phrases;)

There's a typo there: it should be fukumi. But I've talked about these things (and aji) before and they are discussed at length in Go Wisdom, so I'm not going into it again here - I'm taking a leaf out of Kageyama's book, in fact! もはや、理由をくどくどとは申しますまい。

Reminds me of a problem from Sakata's book Tesuji and Anti-suji of Go:

Black to capture the right way:

Click Here To Show Diagram Code

[go]$$
$$ ------------------
$$ . . . . . . . . |
$$ . . . O O O O . |
$$ . . O X X X O . |
$$ . . O . . O X . |
$$ . . . . X . X . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

Sakata says this is the cleanest, and correct way.

Click Here To Show Diagram Code

[go]$$
$$ ------------------
$$ . . . . . . . . |
$$ . . . O O O O . |
$$ . . O X X X O . |
$$ . . O . . O X . |
$$ . . . . X 1 X . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

This could be correct in special circumstances, but it does leave aji.

Click Here To Show Diagram Code

[go]$$
$$ ------------------
$$ . . . . . . . . |
$$ . . . O O O O . |
$$ . . O X X X O . |
$$ . . O . . O X . |
$$ . . . 1 X . X . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

A bad move,

Click Here To Show Diagram Code

[go]$$
$$ ------------------
$$ . . . . . . . . |
$$ . . . O O O O . |
$$ . . O X X X O . |
$$ . . O . 1 O X . |
$$ . . . . X . X . |
$$ . . . . . . . . |
$$ . . . . . . . . |
$$ . . . . . . . . |[/go]

A reminder that sometimes Kageyama was wrong about nets:
viewtopic.php?p=247929#p247929
viewtopic.php?p=247941#p247941

Edit (with admin hat on): I have moved John's reply and subsequent discussion to a new thread: viewtopic.php?f=15&t=17510, to avoid cluttering this one with 2 parallel discussions.

John Fairbairn wrote:That's one reason the book sold over 100,000 copies

As I have often said, the book is a must-read but for a different reason: its great motivation to take the fundamentals seriously (and this might be called "the big picture"). It just does not go enough into the details of what are the fundamentals, especially not for dans. Kageyama should be praised for the former - not for the latter.