I woke up this morning with an idea about how to present joseki in a more useful manner. What if josekis were sorted in terms of their results, so that for example when looking at a low approach by white to a black 3-4 stone, we could see a list of josekis sorted for example so:

Josekis in which white gets a strong position on the approaching side
Josekis in which white gets a strong position on the other side
Josekis in which white gets a strong position facing the center
Josekis in which white gets solid territory in the corner
Josekis in which black gets a strong position on the approaching side
Josekis in which black gets a strong position on the other side
Josekis in which black gets a strong position facing the center
Josekis in which black gets solid territory in the corner

The idea is that this would enable users to better learn and remember josekis in the context of the desired result. What do you think?

_________________
Patience, grasshopper.

I have been thinking along similar lines. I see a problem though. Suppose in your example the 1 point low pincer is played. If I next check in Yoshio Ishida's Dictionary of Basic Joseki I see many ways white and black can steer the joseki according to their priorities. All of your outcomes are possible. Maybe all you can say is that by playing a pincer you get a bonus if you control adjacent corner. That doesn't mean that the pincer garantees influence or profit in that area.

This sounds very interesting. Of course there may be moves which can result in more than one of these results depending on the chosen variation. This seems sort of like a conceptual framework to help with solving problems from the whole board thinking in joseki series.

Right, the pincer itself won't guarantee any particular outcome - there are hundreds if not thousands of variations starting with a low approach to a 3-4 stone - but it might be useful when learning josekis to be able to concentrate on which variations steer towards a particular result, and which don't. Also, when learning a joseki, it could be helpful to have some prior information as to the outcome, so as to better understand the moves in the context of what they are trying to achieve.

_________________
Patience, grasshopper.

I don't think presenting joseki in this fashion necessarily makes it easier to remember them. You don't learn or remember joseki by the outcome, but by the meaning behind each move. Furthermore, these categories aren't even that clear. "Strong" can mean anywhere from a thick wall to a low position on the side with little defects to merely a group that has outside access to the center. Solid territory in the corner can depend on the value of losing/obtaining sente as there are plenty of joseki where either player can tenuki at various points in the sequence. These categories also only look at the benefits for a particular side, but makes no mention of the cost you must pay to obtain the desired result, which I think is just as important in understanding.

Hmmm. I don't like this idea.

When you're studying some move, the general idea should be clear from the variations anyway. Otherwise, the move is just above your level. But whether or not the actual result reflects your expectations of course depends on your opponent, and that is completely up to you to read out. A book cannot guarantee that a certain move will lead to a certain result, because the result depends on your opponent's moves as well. So I don't think it makes sense to categorise moves according to which result you will get.

For example, the purposes of both of these moves are kind of clear, but the actual result is unexpected!

Click Here To Show Diagram Code

[go]$$Bc Purpose: live in the corner easily
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . . O . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . X , O . . . . , . . . . .
$$ | . . . . 1 . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ +------------------------------[/go]

Click Here To Show Diagram Code

[go]$$Bc Result: big fight!
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . . O . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . X 3 O . . . . , . . . . .
$$ | . . 4 2 1 . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ +------------------------------[/go]

Click Here To Show Diagram Code

[go]$$Wc Purpose: make a base in the corner
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . X , O . . . . , . . . . .
$$ | . . 1 . . . X . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ +------------------------------[/go]

Click Here To Show Diagram Code

[go]$$Bc Result: lose the corner in exchange for a wall!
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ | . . O . . O . . . . . . . . .
$$ | . . X O O X X . . , . . . . .
$$ | . O O X X O X . . . . . . . .
$$ | . . X . X O X . . . . . . . .
$$ | . . . . . . . . . . . . . . .
$$ +------------------------------[/go]

_________________
My name is Gijs, from Utrecht, NL.

When in doubt, play the most aggressive move

Nonetheless, I like the idea. Even for joseki that I (sort of ) understand, I have trouble remembering the result. No, that sounds wrong. What I mean is, if I think "okay, I want the outside, facing that way, on this board" I don't really have lots of finished patterns pop to mind which I can then analyze and choose between. Instead I think "okay, that's a sensible move because it threatens that, and he'll respond there, and...", and at the end I can decide whether I like the result or not.

This is the reverse of how you want to go. If you're killing a group you figure out where the nakade will be, and work backwards. If you want to live, you find your two eyes first, and then figure out how to make them. Precisely for the reasons that gaius identifies, working your way forward through a joseki doesn't necessarily provide any insight into what it will look like at the end.

Actually, your second example is one of the reasons that the idea occurred to me. This joseki was one that I've periodically learned and forgotten over the last few years, but recently I decided to look at it with the outcome at least of this most basic variation in mind, and knowing what side the white wall ends up facing makes the decision to go down that road a good deal clearer than my previous state of mind of: "I remember the moves of this joseki, so I'm gonna play it!" (I'm sure that the above statement reinforces those who think 6ks have no business learning joseki, but after a while we get tired of L&D).

I'm sure you're right that you can't know the actual outcome of what starts as a joseki in a game because it depends also on what your opponent decides to play, but don't stronger players "choose" a joseki in a certain situation based on it's outcome?

_________________
Patience, grasshopper.

Hey All,

this my first post/reply since over a year, I guess.

In general I find the idea of classifying Joseki a good idea.

And I suggest to consider:
Much of above is quite relative to me personally and in the game in general.
There are many possible game positions of Go - even within 50 moves.
The qualifier 'strong position' for example, what if it is nullified by exactly the same, but colour-inversed joseki on the mirrored side?

Furthermore - for myself - I am much more interested in marking my Joseki books in the following way:

• i) who ends in 'sente'?
• ii) are ladders involved - and where do the lead to?
• (will 5-3, 4-4, 4-5 diametrically opposing corner make a difference?)
• iii) special aji (ko, type of ko?, who starts?, first threat by who?, ko-threat material, endgame etc.)

Especially point ii) is of absolute value.
That gote/sente endings are relative (to rest of board) I learned in the 80's when replaying Korean games
(interrupting 'Joseki'-sequence to start another in a different corner).

_________________
Greetings,
Tommie

3dan EGF (AGA no 13477) || Tommie on KGS: 'June'|| DGS: 'Zhi Laohu' 纸老虎 = 'paper tiger' || Senseis : http://senseis.xmp.net/?tderz ||
ENFP (MBTI) - 'Find your own style within the Fundamentals of Go! '

Welcome back Tommie!

My view of joseki is that the most useful thing they provide is the reference points for evaluating a local position or a variation. I mean, in early stages of the game, after reading a sequence, it can be very difficult to tell which player is better off, if you don't have any reference to compare with.

Apart from that, my impression is that josekis are widely used for avoiding early tactical blunders, i.e. safe sequences learned by heart. But I don't like this kind of use of joseki.

Since I like to have joseki results as reference points for local position evaluation, I think the proposed joseki classification makes some sense.

_________________
If you say no, Elwood and I will come here for breakfast, lunch, and dinner every day of the week.

How is this idea similar and different to what Kogo's does?

Eg. for the one space approach to a 3-4, it lists;



(Where letters are different responses)

_________________
Revisiting Go - Study Journal
My Programming Blog - About the evolution of my go bot.

Wow, that Gary Odom sure is quick! Hardly have I thought up a new feature and he's already implemented it, including Tommie's suggestion of mentioning ladders and sente!

Joking aside, yes, that is pretty much along the lines of what I was thinking, though I was imagining some sort of sorting mechanism. I guess I'll have to try using the available resource a bit to see if "my" idea was indeed useful for the likes of me.

_________________
Patience, grasshopper.