The book Joseki Vol. 3 Dictionary, which completes the Joseki series, is available now:

http://home.snafu.de/jasiek/Joseki.html

Sample pages:
http://home.snafu.de/jasiek/Joseki_3_Sample.pdf

Table of contents:
http://home.snafu.de/jasiek/Joseki_3_TOC.pdf

Cover:
http://home.snafu.de/jasiek/Joseki_3_Cover.png

Review by the author:
http://home.snafu.de/jasiek/Joseki_3_Review.html

**************************************************************************

Joseki / Volume 3: Dictionary
Review by the Author


General Specification

- Title: Joseki / Volume 3: Dictionary
- Author: Robert Jasiek
- Publisher: Robert Jasiek
- Edition: 2012
- Language: English
- Price: EUR 26.5 (book), EUR 13.25 (PDF)
- Contents: joseki
- ISBN: none
- Printing: good
- Layout: good
- Editing: good
- Pages: 256
- Size: 148mm x 210mm
- Diagrams per Page on Average: 5
- Methods of Teaching: decisions, classification, methods, principles, examples
- Read when EGF: 10k - 3d
- Subjective Rank Improvement: +
- Subjective Topic Coverage: +
- Subjective Aims' Achievement: ++


Motivation

Among the genres of go literature, joseki dictionaries have the worst reputation: their purely tactical selection of countless variations suggests rote learning. Everybody knows that this does not work. Joseki / Dictionary breaks with this tradition to emphasise understanding. Strategy and evaluation are more important than minor tactical variety. Therefore the book explains all the josekis' strategic choices and reasons in the global positional context and evaluates their territory and influence. These aspects are so fundamentally important that the book sets a new standard for joseki dictionaries.


Remarkable Top Achievements

Among the English language or probably world-wide go books, Joseki / Dictionary is the only joseki dictionary

- explaining strategic choices for all its josekis,
- classifying all relevant joseki types by their functions and values,
- stating a player's number of excess moves for all its josekis,
- evaluating territory and influence for all its josekis,
- distinguishing equal from one-sided results in general,
- providing all major reasons for decisions in the global positional context.


Characterisation

Dictionary

Chapter 4 dominates the by far greatest part of the book's space. It contains the dictionary of 3-3, 3-4, 3-5, 3-6, 4-4, 4-5, 4-6 and 5-5 josekis. Its top level is organised in the traditional reference style by coordinates of the initial moves. After the very few first moves, structure relies on a different means: the major strategic choices. There are decisions such as sente or gote, upper or left side, thin or thick shape, territory or influence, favourable or unfavourable ladder, support or no support in the adjacent corner etc. The usual row of three diagrams shows the prior position, one follow-up choice and sequence, and the alternative follow-up choice and sequence. Every follow-up sequence has the moves until the subsequent strategic choice or the sequence's end. This structure greatly eases both understanding and learning because strategic choices are clearly distinguished from minor tactical considerations.

The many inserted game examples enrich the text, show applications in the positional context, illustrate alternative variations or sometimes exhibit failures. Those diagrams show sufficient parts of the board or the whole board for understanding the reasons of choices. Where necessary, additional diagrams explain non-obvious tactics. Diagrams for ladders are put right after the strategic choices for such conditions. As for the strategic choices, the reader need not search but finds the relevant information exactly where he needs it. Nevertheless, the book comes with indexes for diagrams and keywords. Besides there are 33 mostly tactical problems.

Databases of professional games have been consulted to distinguish modern from outdated variations, identify currently popular moves, verify the joseki sequences and identify the practically occurring reasons for strategic choices in the global positional context. Thus the reader profits from several weeks of empirical research by the
author.

Go Theory

One of the great hurdles for joseki learning is the difficulty of recognising a joseki's functions. Chapter 2 classifies the types of josekis, characterises every type, studies typical strategic choices and provides the relation to relevant reasons for the strategic decisions. For example, there is the type 'mutual running fight' and the reason to 'make territory during the fight'. The chapter is as simple as valuable; its fundamentals are learnt easily.

Have you ever wondered whether a sequence creates a joseki or why traditional dictionaries do not tell you the why? Chapter 3 gives the answer by inventing the necessary theory! With a bit of practice, a player's current territory can be counted. Usually, assessing influence is even easier: count a player's number of influence stones, then compare Black's and White's numbers! The theory tolerates small calculation mistakes. Needless to say, one must also know whether both players have played the same number of stones in the corner. Josekis are classified according to their values of territory and influence. Without a proper balance, a result is 'favourable for a player'. The book offers solutions also for sequences in which a player has played one or two stones more than the opponent. The age of alchemy ends - now everybody can evaluate josekis!

Readers

Why would a player learn josekis from a purely tactical dictionary consisting of two or three volumes? Surely he does not want to maximise his necessary effort, doesn't he? The author firmly believes that a dictionary consisting of just one volume suffices. What matters is rather how careful the dictionary's josekis have been selected. Probably Joseki / Dictionary with its 400 josekis and 130 (mostly professional, a few top amateur) game examples can serve as the only joseki dictionary a player needs for his go career. A beginner finds the simple variations, a kyu player finds exactly the practically relevant variations and a dan player finds the most frequent of the modern, recently invented josekis. When a player understands the book's variations, he will be prepared to rediscover or invent further josekis by himself.

Although Volumes 1 and 2 provide all the background go theory, Volume 3 presumes comparatively little prior knowledge. The essential terms' definitions of the preceding volumes are repeated. For a truly profound understanding, it is recommended though that the reader has at least a rough idea of development directions, sacrifice, large scale cuts and aji. In other words, a player profits the most by understanding well both strategic concepts and josekis. Volume 3 can be read independently, but a kyu player wishing to reach dan level some time cannot circumvent strategic concepts forever.

Joseki / Dictionary can be used as a textbook or a reference. It serves the joseki learning of players from 10 kyu to 3 dan while they get a very thorough understanding. Evaluation adds another layer to the book; players below 5 kyu can postpone chapter 3 and the values in the diagram captions. Since the strategic choices, evaluation, classification and reasons explain the core missing in purely tactical dictionaries, they can be understood better while consulting Joseki / Dictionary. Players from 4 to 5 dan can complete their knowledge gaps related to modern josekis not existing in their older source. For yet stronger players, the evaluation method is new.

What the Book is Not

Volume 3 does not repeat the go theory of Volumes 1 and 2. Since Joseki / Dictionary offers the most important tactical variations but presumes the reader's ability to find the easiest alternative tactics by himself, probably the book is not suitable for those double digit kyu players still having basic tactical difficulties. Joseki / Dictionary does not compete in terms of numbers of tactical variations, as found in detailed purely tactical dictionaries such as the Nihon Kiin's Joseki Dictionary. Joseki / Dictionary has intentionally skipped most of the too complicated, rare branches.


Comparison

++ (very good), + (in between average and very good), o (average), - (in between very bad and average), -- (very bad), ~ (to)

When is volume 1 going to be back in stock?

_________________
Studying Go in China- http://www.Aaronko.net - My Go Blog

Don't close your eyes or the stones die.

In Go it's not how thick it is, it's what you do with it!

My Kgs Rank Graph

I second this question.

RJ, I commend you for using examples from women's professional games.

The current planning is: roughly during the following 4 weeks. There are several factors I need to consider, and I need to negotiate with a printer, so I cannot know the exact date yet.

This week I will be too busy processing orders.

Next week, I will have more time to consider the details of Vol. 1 reprinting.

Question from section 3.3.

Here you evaluate the following joseki:

Click Here To Show Diagram Code

[go]$$B 3.3
$$ ------------------
$$ | . . . . . . . . .
$$ | . X O X . . . . .
$$ | . X O X . X . . .
$$ | . X X O . . . . .
$$ | . O O O . . . . .
$$ | . . . . . . . . .
$$ | . . . . . . . . .
$$ | . . . . . . . . .
$$ | . . O . . . . . .
$$ | . . . , . . . . .
$$ | . . . . . . . . .[/go]

The territory for white is analyzed as follows:

Click Here To Show Diagram Code

[go]$$B 3.3 1-10
$$ ------------------
$$ | . . . . . . . . .
$$ | . X O X . . . . .
$$ | . X O X . X . . .
$$ | 5 X X O . . . . .
$$ | 3 O O O . . . . .
$$ | 4 . 6 . . . . . .
$$ | . . 8 7 . . . . .
$$ | . . . . . . . . .
$$ | . . O . . . . . .
$$ | . 2 0 , . . . . .
$$ | . 9 1 . . . . . .[/go]

Click Here To Show Diagram Code

[go]$$Bcm11 3.3 11-12
$$ ------------------
$$ | . . . . . . . . .
$$ | . X O X . . b a .
$$ | . X O X . X . . .
$$ | X X X O . . . . .
$$ | X O O O . . . . .
$$ | O . O . . . . . .
$$ | . . O X . . . . .
$$ | . . 2 1 . . . . .
$$ | . . O . . . . . .
$$ | . O O , . . . . .
$$ | . X X . . . . . .[/go]

So you get 7 points of territory for white this way. This seems standard, but I wonder about . It's not quite the way I was taught. (I would assume at instead, giving a different count.) I suppose it is at least consistent in that white 'a' black 'b' is assumed when counting black. Maybe it doesn't matter but I was wondering if this was explained in one of your books.

What is perhaps harder is the statement that white can expect 7 points more influence from his position. I'm not sure where you get this number except by reverse engineering it assuming that it's already a joseki. In section 3.4 the same joseki is discussed from the perspective of influence by counting thick outside stones. (Or, as you say, the "influence stones.") But the number obtained there is not in points but in some other metric, so it's looking like apples vs. oranges.

Maybe there are some useful heuristics here, but statements like this are a bit discourarging:

"It is not always clear exactly which stones have or do not have significant impact; then one needs to
apply some, for example, visual heuristic to make the distinction."

This is a bit like saying one has to be strong in order to be strong.

While I admire the attempt to try to explain things from first principles, I just don't know how far one can take that approach. I really get the feeling that learning even exchanges is more like chick sexing---learned subconciously from a ridiculous number of examples---than like mathematics. And I say this even though I am rather fond of mathematics.

But I can appreciate what you are saying in the sense of "things to pay attention to while studying a ridiculous number of examples."

My motivation for selecting game examples is always to choose the most fitting examples, which can best illustrate what I want to teach. As it happens, some of those games involve female players and a few top amateur games were more interesting than pro games would have been at some places in the book. Sometimes, a professional failure is more instructive than correct play.

During assessment of what I call 'current territory', one must apply its definition. Sente and peaceful defense are assumed in that definition. What you get does not look like perfect play but looks like local endgame reductions from all outside directions. It is similar to Cho Chikun's territory judgement in his book Positional Judgement, except that I add the sente condition.



In fact, current territory is so important and useful that it is explained in each of the three Joseki volumes! :)



The example joseki has the stone difference 0 (Black and White have played equally many stones). By the assumption that it is a joseki, (almost) local equality of Black and White follows and thus the remaining potential of a later white excess of an additional 7 points.



Not exactly reverse engineering. We know the local position. To it, the general evaluation theory in chapter 3 allows to justify the assumption that we have a joseki of the value type 'ordinary sharing of territory and influence' here. (You will see...)



There are three values (lemons, apples, oranges :) ) to be considered for every joseki: stone difference, territory count and influence stone difference.

A direct comparison of territory and influence is possible only if the stone difference is 0. This is so in the particular example you cite.

Now, apples and oranges, which measure different things indeed, are compared by forming their ratio: How many excess territory points does Black have per excess influence stone of White? To get the ratio, divide the territory count by the influence stone difference!

Like you can ask how many oranges you get for one apple, you can ask how many excess territory points Black has got per excess white influence stone.



It is an approximative heuristic. Weaker players will assess Black's territory as 12 - 16 (average guess 14), intermediate players will get 13 - 15 (average guess 14) and strong players 14. Weaker players will assess White's excess of outside influence stones as, uh, 3 (it is hard to imagine that one gets a wrong number in that particular example).

There are, however, josekis for which value assessment can be much tougher:

Click Here To Show Diagram Code

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

It would be dishonest to say that here determination of either player's current territory would be easy.



For the 400 josekis in the book, I could always apply it, except for one follow-up variation leading to a huge ko exchange during the late middle game. Since the book has a broad and representative selection of joseki results, I am confident that the method is applicable to almost all corner sequences. I could apply it even to all hot fighting situations demanding immediate local continuations.



Read the book and report your impression afterwards! All maths you need is to form differences, divide to get the ratio or add / subtract very simple numbers.

Have a look here:

viewtopic.php?f=15&t=6018

Concerning heuristics for determining the influence stones, sector lines on the outside help a lot! Only when influence is strongly create also by rear stones, you would include also them.

Robert,

will you make PDF of Vol 1 available too?

I think you said, some time ago, you can't release PDF until you sell all books. Them, now is the time for a PDF.

Thanks, D

I will make available Vol. 1 as PDF pretty soon, Vol. 2 as PDF will come later. (In book layout, with wider inner than outer margin, both are already available for EUR 13.25 each.)