Click Here To Show Diagram Code

[go]$$B Big points
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . 7 . . b . . |
$$ | . . . 2 . . . . . , . . . . . 1 . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . 6 . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , a . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . 4 . . . . . , . . . 5 . , 3 . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

Let's take this starting position, assuming the open corners are always the biggest points, i.e. the next major area. For the sake of the argument, I have played on a 3-4 point, creating an asymmetric position.

According to dwyrin's heuristic, the bottom area is now the biggest area, because and are separated by 12 open spaces, while all the other sides are 11 spaces wide.

Another heuristic says that asymmetric corner stones invite enclosures or approaches. Today, a high 2-space enclosure is more popular. takes a big point.

The bottom has now become the smallest area, the others still being 11 points wide. Conventional wisdom says that an enclosure invites an extension. Hence, a White splitting move at ''a'' is a traditional big point, perhaps the biggest point. However, modern play suggests that an approach from the inside is better, because corners are more important than sides, and bot play suggests that White even jumps into the corner with ''b''.

According to the strict definition of wide spaces, the 3-3 invasion is not "big", but we can extend the definition with such available invasion points. creates an "urgent" situation for Black, whose corner stone is threatened to be pincered, so that it can't make a base (borrowed, implicit definition of "urgent")

Now the question is if finds itself in an urgent position? It can still be sacrificed with a corner invasion, or extend along the side, unless Black kicks it and attacks next, creating a double purpose move. Is playing elsewhere worth all that?

Click Here To Show Diagram Code

[go]$$W Big points
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . X . . |
$$ | . . . . . . . . . . . . . X . . X O . |
$$ | . . . O . . . . . , . . . . . X O O . |
$$ | . . . . . . . . . . . . . . . X X O . |
$$ | . . . . . . . . . . . . . . . . O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . 1 . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . 2 . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . O . . . . . , . . . X . , X . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

Let's assume White chooses the modern bot corner pattern, finishing it with , an "urgent" point. Temperature has dropped in the area and Black can look for a big point. The widest area is now on the left. Although LZ slightly prefers to approach from the top or the bottom, when fed the approach from the wider side, she doesn't heavily object

Click Here To Show Diagram Code

[go]$$W Big points
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . X . . |
$$ | . . 6 . . 2 . . . . . . . X . . X O . |
$$ | . . . O . . . . . , . . . . . X O O . |
$$ | . . . . . . . . . . . . . . . X X O . |
$$ | . . . . . . . . . . . . . . . . O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . O . . |
$$ | . . 1 . . . . . . . . . . . . . a . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . 3 . . . . . . . . . . . . . . . . |
$$ | . O . . . . . . . . . . . . . . . . . |
$$ | . X X X . 4 . . . . . . . . . . . . . |
$$ | . X O O . . . . . . . . . . . . . . . |
$$ | . X X O . . . 5 . , . . . X . , X . . |
$$ | . X O . . O . . . . . . . . . . . . . |
$$ | . . O . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

A variation of the same corner pattern gives this result. Black is not stable yet, so, rather than playing a big point at the top, White applies pressure with the leisurely

With that extension in place, Black surely wants to approach from the top? Not really: LZ has her eye on the forcing move ''a'', which is not traditional. Her next choice is to invade the corner.

If we insist playing the approach at , LZ finds value in connecting her forcing move with her side extension, increasing the pressure on Black's lower left group. Both and are "urgent", next Black invades at because White has left the corner unattended and a double approach makes less sense due to White's stones on the side.

We can say this invasion is now a genuine "big point".

Click Here To Show Diagram Code

[go]$$W Big points
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . X . . |
$$ | . . X 1 . X . . . . . . . X . . X O . |
$$ | . . 2 O . . . . . 0 . . . . . X O O . |
$$ | . . . 3 . . . . . . . . . . . X X O . |
$$ | . 4 6 7 . . . . . . . . . . . . O . . |
$$ | . 8 5 9 . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . O . . |
$$ | . . O . . . . . . . . . . . . . b . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . O . . . . . . . . . . . . . . . . |
$$ | . O . . . . . . . . . . . . . . . . . |
$$ | . X X X . X . . . . . . . . . . . . . |
$$ | . X O O . . . . . . . . . . . . . . . |
$$ | . X X O . . . O . , . . . X . , X . . |
$$ | . X O . . O . . . . . a . . . . . . . |
$$ | . . O . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

After the corner pattern in the top left, Black plays the big point at the top and White does likewise at the bottom with ''a''.
The right side is the only remaining major area. LZ still wants to play ''b'' and it seems to make a big difference (10%) with any other move in that area. This suggests ''b'' is not so much a "big point" but one that raises the temperature, creating an "urgent" situation for White.

Click Here To Show Diagram Code

[go]$$W Big points
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . X . . |
$$ | . . X O . X . . . . . . . X . . X O . |
$$ | . . X O . . . . . X . . . . . X O O . |
$$ | . . . O . . . . . . . . . . . X X O . |
$$ | . X X O . . . . . . . . . . . 3 O . . |
$$ | . X O O . . . . . . . . . . . . . 6 . |
$$ | . . . . . . . . . . . . . . . 2 O . . |
$$ | . . O . . . . . . . . . . . . 4 1 . . |
$$ | . . . , . . . . . , . . . . . , 5 . . |
$$ | . . . . . . . . . . . . . . 7 . . . . |
$$ | . . O . . . . . . . . . . . . . . . . |
$$ | . O . . . . . . . . . . . . . . . . . |
$$ | . X X X . X . . . . . . . . . . . . . |
$$ | . X O O . . . . . . . . . . . . . . . |
$$ | . X X O . . . O . , . . . X . , X . . |
$$ | . X O . . O . . . . . O . . . . . . . |
$$ | . . O . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

The difference is too big to ignore, so let's take LZ's suggestion.

With some goodwill, is the last big point, although we should perhaps say that the right side was the last major area and Black developed it with through while putting pressure on White.
%%%%

In the above analysis, we have switched between the heuristic of playing a big point and LZ's top choice. We chose a big point when LZ only marginally objected (<2%). When LZ's top choice was considerably better, we followed her, which was particularly true for the "urgent" points.

In this example it was Black who took the last big point. However it was not the fact that Black claimed the last major area but the way he did, with that pressuring sequence, which made him stay in the game (at 47%).

I suppose we can. Would Ishida agree? BTW, Elf likes the 3-3 invasion, while AlphaGoTeach likes your approach.

I feel certain that this use of urgent does not accord with typical go usage. I am not at all sure if it accords with the usage of combinatorial game theory. It does accord with the fuzzy English usage. In any case, I am not sure that I agree.

AlphaGoTeach prefers a or b for over the jump attachment.

I feel a possible "big point" creep. The problem, as I see it, is that big, applied to go plays, implies a large gain or a global sente. It covers big points, urgent points, and points that have not been described as either one. The more we expand the meaning of "big point" toward "big" plus "point" the less clear we make the distinction. OC, that may be a good thing, I dunno.

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

OK, let's not call it one then. I'm not for fuzzy definitions. Let's reserve "big point" for "point in large(st) open space"

I try to play the latest big point on the board at every turn. I do not care about the largest empty space surounding my move, if this move is the latest big point. Sometimes the latest big point is surrounded by the largest empty space indeed. And if I am looking where to play a similar local shape on the board, say for example a knights move enclosure, I play it at the largest empty space, especially if I have no other clue where to play it.

In other words I enjoy your posts and sequences about taking the last big point in the opening, but I do not believe in the last big point in the opening heuristic. When I search for examples for this heuristic with AI, I found none up until now. (And if I tried to construct one, even with suboptimal fuseki moves, the advantage the last big move in the largest empty space was matched by an alternative move not in the largest empty space.)

I couldn't properly follow any of the various posts above as it was too hard to work out which text went with which diagram. In the internet age, the tolerance we give before moving on beyond even the slightest hiccup gets shorter and shorter - probably down to microseconds by now in my case.

But, with that caveat, a couple of things strike me from the text portions.

1. Any attempt to retain the word 'big' is going to fail. It has too many wrong associations among western players (and is inaccurate to start with).

2. If we are looking for definitions in English, we need more intervention from native speakers. Lurkers: speak up!

To get the ball rolling, I am going to suggest VANTAGE POINT.

The thinking is that:

a. It is an accurate enough translation of oba, and certainly more accurate.

b. Vantage is not a very common word in ordinary speech and so is not burdened with not too many associations. Those that it does have (e.g. military positions as in Tennozan) are actually relevant in go. In particular, it has no significant association with size.

c. Point in 'big point' is too often prone to be taken as a specific intersection. In 'vantage point', it seems to me as a native speaker, that point is not taken in the same way but as an area. I suspect that is precisely because of the military connotations (brow of a hill, the high ground, etc).

d. I don't think a native speaker would ever say vantage point of a 3-3 invasion (I inferred from the above that there is some desire to exclude such plays from the definition???)

e. In practice, an oba is not just about occupation of an area, but also tends to imply follow-up moves from it. A vantage point can be seen as a launching pad for attacks.

f. Much of the proposed definition so far seems concerned with packing everything into a box labelled 'big point' and so moves that we would not normally think of as oba are being given that status. Japanese texts that I see that cover these sorts of positions seem to use a much wider palette, and one phrase that keep coming up is 'keep away from thickness.'

There are some other possible alternatives to vantage point. One that occurs to me is a 'colonising move.' Such terminology was common in Japan in 1930s and there was a form of chess invented there where the pawns were colonists. For the faint of heart, colonists could be changed to pioneers, I suppose There may be ideas lurking in thinking of an oba as a hillfort, or an outpost. Again; lurkers of the world, unite!


****

On a separate topic, mention is made of dwyrin here. Who he? More to the point, it seems he is being given credit for a theory that, as far as I know, belongs to O Meien in a 2002 book. His very first page is entitled "The widest side is the biggest" and his second diagram shows four small shimaris so arranged that the width between them on the third line is respectively 9, 10, 10 and 11 lines. The word oba seems absent in his book, and he talks instead a lot about the initiative (nb not sente). In fact, corner plays are mostly absent - it is definitely a book about sides and so must overlap in some way with obas. But he is not at all limited to obas. For example, there is one interesting position where he compares a position which has a 3-space width on the second line and a 3-space width on the 4th line. Incidentally, he also explicitly says his advice is designed to make the game easier and more fun for the reader, and also to enable him to make more powerful moves. He does not claim the moves shown are the best. The book was in the NHK Go Series so that will indicate the audience level. Maybe this book should be revisited as the launching pad for a new to look at go theory of the sides through AI eyes. The key word would then be 'wide'. But I can't resist pointing out what I said in an earlier post: the meaning of o- in oba is 'wide' or 'important', not 'big'.

I seem to recall hearing that idea in the 1990s as coming from lessons or lectures by Yang Yilun. My unstated reaction was, it's an oversimplification, and besides, who doesn't take that width into account? Maybe O and Yang have elaborated on the idea, but I expect that many people have discovered it.

Edit: One reason for my skepticism of "The widest side is the biggest" is that when I was learning go I heard that one criticism pros had of amateur play was playing "where the sound is the loudest."

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

It's very easy: the text under the diagram goes with the diagram.

Dwyrin is not a household name but he's a prolific high amateur youtuber who has been quite successful teaching Go in simple terms. His basic series is rather enlightening.

You don't have to be a genius to see which space is the widest but it takes a good teacher to bring simple concepts to the masses.

Voor de rest vind ik het ook spijtig dat ik geen native speaker ben, maar ik doe mijn best.

I seem to have a hard time conveying the message. I don't claim people should play in the largest open space. I don't know that. Or rather, I know that is not true.

The problem is "playing the biggest point" must mean something else than "playing the best move". If they are synonyms, ok, but then it's not very interesting to explore the concept. So "big point" must mean something else than "good move". So what is it, if it's not "a move in the largest open space"?

If we agree that "big point" means "available point in large open space", then we can discuss whether "try to play big points" or "take the last big point" is a good heuristic. Maybe it's not, but the debate is clear. If "big point" means "good move", then there is no debate: yes, you should always try to play the best move.

If there is a vantage point, a move that will provide a strategic advantage for you, and you can not play at this vantage point immediatly. In this case play the more urgent moves and sequences in such a way that you retain the privilige to play at the vantage point.

Sometimes you and your opponent struggle for the same vantage point. Whoever gets to play at the vantage point, before only tactical moves remain on the board, secures an advantage.


Now we only have to distinguish strategic and tactical moves.


Why should I care about sente and gote? Perhaps as a means to get on the next vantage point.

I am not sure vantage points are of great importance in go. I currently think of go as a mainly tactical game.

FWIW, Knotwilg, I’m a native speaker, and you write better English than I do on L19 (I’m pretty careless with my presentation here, admittedly).

I think any argument about requiring native speakers to come up with good terminology is a little obnoxious, so I hope you didn’t take offense.

_________________
be immersed

Dieter:



It may be a browser issue, but the problem is that much of the text I see above a diagram is physically attached to the diagram below (and well detached from the diagram above), which obviously suggests each goes with the other. And the behaviour is not consistent.

To answer your implicit question, yes I can read Dutch/Flemish without recourse to Google. But you do seem very quick to take my comments personally. Why? I don't know you, do I? Don't you recall previous similar discussions about other definitions in English? It was the definitions that were in dispute, not the person. It's a general problem, in any language. And even native speakers can argue amongst themselves. Is it eggs and bacon or bacon and eggs (it's the latter!)? In any case it wasn't even specially you that prompted my remark.

Bill:

Of course O Meien elaborates on his width theory. At least he spends 221 book pages on it. But there was one thing that caught my eye on riffling through it which may be of direct interest here. He gave the following position (Black to play) with the comment: "In this position the focus is on the left side."



That sort of comment is very, very common in the hints for whole-board problems in pro books. Here it is easy to know that's what he wants us to think because it's the widest area and that's what the book is about. But if you didn't know what the book was about, I suggest that it is not at all obvious that the focus is on the left side. Yes, I'd like to play there, but I'd like to play at the top, on the right, and everywhere else as well. Likewise in all the other books.

So, in the common way of building up definitions brick by brick, should we not start, before getting onto big points, with some sort of principle that would allow us to ascertain first which is the prime area of focus.

But still not easy, methinks. LZ started off here with lots of moves on the left side (though none mentioned by O), but after quite a few playouts it decided a play on the lower side was the main focus! And O didn't discuss that move (the lower centre star point).

Width was the main suggested factor for O. Other pros have peddled counting the number of stones on each side or the number of weak groups. Yet other have used ideas such as what Wilcox called sector lines. Amateurs have also been fecund. We see similar shortcut ideas in chess. Many seem to work up to a point, but not enough to trust very much.

Most things seem to get easier the more you think about them. Go just gets harder the more you think about it!!!!!!

It seems to me that the vast majority of mistakes in old books have to do with weak groups. AI says professionals were incorrectly evaluating moves that cause an effect on weak groups - in most cases I've seen, professionals were overestimating the strength of a weak group that can run. Perhaps the most useful thing we can do with such errors is not necessarily rewrite the books, but construct a new principle to help explain when such a group is more or less important than big (vantage) points.
I've been thinking on this a lot to myself, but certainly not enough to present a principle and call it gold. It seems like a principle like the one here would be "inferior" to one with more consideration to weak groups - it could only apply to a situation where the other principle doesn't, without any weak groups remaining on the board. So why bother dealing with this one, when what I really need in my games is a way to decide when to stop attacking and when to start attacking?

This is one reason that I have so much hope for the youngsters who will become tomorrow's pros. They have much less baggage to deal with, and they will have grown up thinking about the bot's moves and seeing what the bots think of their moves. Not a tabula rasa, but a fresh slate.

As for us, as a famous philosopher, who shall remain nameless, said, we start from where we are. Not only can we not throw out everything and start over, doing so would be counterproductive. The baby with the bath water, as you say. Can you imagine telescoping 300 years of go history into one decade? That, I think, is the magnitude of the problem we face.



It's a different game, isn't it? But not all that different, as far as K-10 is concerned. One impression I have gotten from the Elf commentaries is that Elf likes what I shall call anti-shape plays, that is, plays that prevent the opponent from making good shape. K-10 is a good shape point for White. If Black plays there he has some nice threats for attack, defense, and making territory. I doubt if these escaped O Meien's attention, but he believed that the left side was more important, and talking about the bottom side would distract from the point he wanted to make. Still, it seems to me that the question of where to play falls into the familiar oba vs. kyuba debate. Isn't K-10 a local kyusho? So how do we regard it globally? I can well imagine Sakata playing there. One thing that we may possibly be able to take away from the O Meien - LZ dialectic is that we should give higher priority to anti-shape plays like K-10.

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.