A vague treatise on influence

Uberdude · Post by **Uberdude** » Mon Feb 20, 2012 11:25 am

This thread had degraded, as often seems to happen when Robert posts, into the usual to-and-fro about the merits or otherwise of Robert's methodoligical approach to Go and his books. We can talk about theory until the cows come home, but at the end of the day Go is about deciding where to play your stones. To that end, I would be interested in seeing how Robert, or indeed someone else who has read his books, would use his approach to decide where to play in the following late-opening/early-middlegame position from a game of mine (the topic of http://www.lifein19x19.com/forum/viewto ... f=4&t=5277).

Click Here To Show Diagram Code: [go]$$cm43 Black to play $$ --------------------------------------- $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . O O O . O O . . . . . . . X . X . . | $$ | . X O X X . . . . , . . . . . . X O . | $$ | . . X . . . . . . . . . . . . X O X . | $$ | . . . X . . X . X . . . . . . . O X . | $$ | . . . . . . . . . . . . . . . . O X . | $$ | . . . O . . . . . . . . . . . . O . . | $$ | . . . . O . . O . O . . . . . . . . . | $$ | . . X , . . . . . , . . . . . , O . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . X . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . O O . . | $$ | . . X . . . . . . . . . . . . O X . . | $$ | . . . , . . . . . , . . . . . , X . . | $$ | . . O . O . . . . . . . . X . X . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

There was a suggestion the last 4 moves weren't great, so an alternative problem would be this slightly earlier position.

Click Here To Show Diagram Code: [go]$$cm39 Black to play $$ --------------------------------------- $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . O O O . O O . . . . . . . X . X . . | $$ | . X O X X . . . . , . . . . . . X O . | $$ | . . X . . . . . . . . . . . . X O X . | $$ | . . . X . . . . . . . . . . . . O X . | $$ | . . . . . . . . . . . . . . . . O X . | $$ | . . . O . . . . . . . . . . . . O . . | $$ | . . . . O . . . . . . . . . . . . . . | $$ | . . X , . . . . . , . . . . . , O . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . X . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . O O . . | $$ | . . X . . . . . . . . . . . . O X . . | $$ | . . . , . . . . . , . . . . . , X . . | $$ | . . O . O . . . . . . . . X . X . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

These positions strike me as ones where influence and direction of play considerations are paramount. I find generating moves and evaluating my choices in such situations difficult. If Robert can come up with something to help this, it would be valuable indeed. I would also be interested how an intuitive player such as Magicwand would approach this position.

illluck · Post by **illluck** » Mon Feb 20, 2012 11:53 am

What if black just tries B12 in the second diagram? If black connects white is sort of floating.

Uberdude · Post by **Uberdude** » Mon Feb 20, 2012 11:57 am

illluck wrote:What if black just tries B12 in the second diagram? If black connects white is sort of floating.

Floating but light. See 3rd diagram of http://www.lifein19x19.com/forum/viewto ... 750#p87750 why I rejected b12.

illluck · Post by **illluck** » Mon Feb 20, 2012 12:00 pm

I don't understand your move 41 in that diagram - after black connects there's no rush to try to capture the two stones.

Uberdude · Post by **Uberdude** » Mon Feb 20, 2012 12:18 pm

illluck wrote:I don't understand your move 41 in that diagram - after black connects there's no rush to try to capture the two stones.

Indeed, where else do you propose to play? My feeling was that n5/6 was the key influence point of this board and thus whoever got sente to play there would have the advantage. Maybe Robert can confirm or deny this using his method? B12 is gote, whereas the game line strengthened the black group in sente.

illluck · Post by **illluck** » Mon Feb 20, 2012 12:21 pm

F4 looks really tempting to me for some reason - black getting shut inside corner is not as painful if the bottom is low and the left side grows.

RobertJasiek · Post by **RobertJasiek** » Mon Feb 20, 2012 12:43 pm

Uberdude wrote:
$$cm43 Black to play
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . O O O . O O . . . . . . . X . X . . |
$$ | . X O X X . . . . , . . . . . . X O . |
$$ | . . X . . . . . . . . . . . . X O X . |
$$ | . . . X . . X . X . . . . . . . O X . |
$$ | . . . . . . . . . . . . . . . . O X . |
$$ | . . . O . . . . . . . . . . . . O . . |
$$ | . . . . O . . O . O . . . . . . . . . |
$$ | . . X , . . . . . , . . . . . , O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . X . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . O O . . |
$$ | . . X . . . . . . . . . . . . O X . . |
$$ | . . . , . . . . . , . . . . . , X . . |
$$ | . . O . O . . . . . . . . X . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------
Click Here To Show Diagram Code
[go]$$cm43 Black to play $$ --------------------------------------- $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . O O O . O O . . . . . . . X . X . . | $$ | . X O X X . . . . , . . . . . . X O . | $$ | . . X . . . . . . . . . . . . X O X . | $$ | . . . X . . X . X . . . . . . . O X . | $$ | . . . . . . . . . . . . . . . . O X . | $$ | . . . O . . . . . . . . . . . . O . . | $$ | . . . . O . . O . O . . . . . . . . . | $$ | . . X , . . . . . , . . . . . , O . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . X . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . O O . . | $$ | . . X . . . . . . . . . . . . O X . . | $$ | . . . , . . . . . , . . . . . , X . . | $$ | . . O . O . . . . . . . . X . X . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

The major topic is the two center groups and which of them provides the better influence. You ask for application of my book's concepts but, if I describe that in details here, it would need to long. So let me concentrate on the most interesting aspects. If Black allows White to make the forcing exchange h13-h14 etc., then the white center group creates significant influence because of being 0-connected and adjacent to a wide open area. So Black wants to prevent that. j10-h13 does not solve the problem. h12 helps White to become stronger. This brings us to the question if Black can cut h11/h11 or whether they are already 0-connected or else indirectly connected. Black f11 or g11 look promising for the sake of cutting and motivating White to conquer dame or c10 while Black becomes strong around h11. But I would want to check more variations than is fun on webpage diagrams and then consider more carefully whether White's connection was indirect, i.e. he can achieve still at least a fair result despite a black center cut. If yes, then b12 could be better.

RobertJasiek · Post by **RobertJasiek** » Mon Feb 20, 2012 12:51 pm

Uberdude wrote:My feeling was that n5/6 was the key influence point of this board and thus whoever got sente to play there would have the advantage.

P4 - P3 - M5 is good for moyo onfluence. So if Black b12, then H13 - H14, then that sequence. But b12 is too easy going for White. Black n6 misses the more urgent center groups topic. So see my other message.

RobertJasiek · Post by **RobertJasiek** » Tue Feb 21, 2012 1:05 am

Uberdude wrote:This thread had degraded, as often seems to happen when Robert posts, into the usual to-and-fro about the merits or otherwise of Robert's methodoligical approach to Go and his books.

You could also say that the dozens of followers of this thread don't take out their hundreds of other books to offer citations for other explanations of influence. Among my hundreds of books, I do not know any useful other citation though. But what about All About Thickness or hopefully existing Asian books about the topic? Don't they have any description of what "influence" actually is? If so, which and are they even worth considering in comparison?

Bill Spight · Post by **Bill Spight** » Tue Feb 21, 2012 4:25 pm

RobertJasiek wrote:
Bill Spight wrote:I think that there are two different senses of influence that are in common use. [...] power or outside strength. [...] The second sense comes from computer go (as far as I know), and means the effect that a stone or group of stones has on empty points or stones.
I am not sure if I understand what you are saying. Now I guess you might mean (the influence-generating aspect of, e.g.) thickness when you say "power or outside strength" as influence type I? Do you mean "influence" when you speak of (the influence absorbing) influence type II?

If so, then also type I can have negative connection or life values (it is not thickness then but thinness).

What is your research about, if I may ask?

It sounds like you are talking about influence as the effect of stones on points or other stones. The other sense is the traditional one of outside strength, or as in the phrase, sphere of influence.

My research is about the effect of stones on points or other stones. I am focusing on situations where it may be calculated precisely.

For instance, consider this corridor, with all stones alive.

Click Here To Show Diagram Code: [go]$$ $$ X X X X O $$ X . . . O $$ X X X X O[/go]

There are 1.25 points of territory, on average, in the corridor. We may allocate the points based on the possibility of a point becoming territory, as (0.75, 0.5, 0) for the three points. The values are the combined values of the influence of the surrounding stones. We may also allocate points based on the possibility of a point becoming area. In this case the values are the same.

Click Here To Show Diagram Code: [go]$$ $$ X X X O $$ X . . O $$ X . . O $$ X X X O[/go]

In this case the territory values are ((0.25, 0), (0.25, 0)) for territory and ((0.5, -0.25), (0.5, -0.25)) for area. The total influence is the same.

RobertJasiek · Post by **RobertJasiek** » Tue Feb 21, 2012 11:30 pm

Bill Spight wrote: It sounds like you are talking about influence as the effect of stones on points or other stones. The other sense is the traditional one of outside strength, or as in the phrase, sphere of influence.

Identifying something as a sphere (or region) of influence is an application of "influence". The effect of stones on intersections (possibly carrying) other stones can mean that a particular intersection has better influence values for one of the players. A string of such intersections then can be called a "sphere" of influence. More precisely, consider for every intersection the c-connected, a-alive and t-territory values for Black and the d-connected, b-alive and u-territory values for White. If c>d && a>b && t<u, then we call the intersection "belongs to a sphere of influence of Black". If c<d && a<b && t>u, then we call the intersection "belongs to a sphere of influence of White". In the other cases, neither player has an advantage for all three aspects (connection, life, territory). Alternatively it would be possible to consider only one or two of the three aspects or to make additional requirements like non-negative connection and life values for an intersection to be part of a player's sphere.

I am focusing on situations where it may be calculated precisely.

Given very limited calculation time?) I think though that every parameter value in my model can be determined within hours for almost all very difficult cases. My model assesses whether an intersection can be territory at all while you ask for the precise CGT-territory value. Naturally that can be much more difficult to assess.

$$
$$ X X X X O
$$ X . . . O
$$ X X X X O
Click Here To Show Diagram Code
[go]$$ $$ X X X X O $$ X . . . O $$ X X X X O[/go]

The intersections are *-connected and *-alive for Black.

They are -2-connected, -1-connected, *-connected for White.
They are -2-alive, -1-alive, *-alive for White.

They are 0-territory, 1-territory, not t-territory for Black.

They are not u-territory for White.

Click Here To Show Diagram Code: [go]$$ $$ X X X X X O $$ X . . . . O $$ X X X X X O[/go]

They are 0-territory, 0-territory, 1-territory, not t-territory for Black.

The CGT-territory method is more precise on the tiny, endgame scale than the n-territory, which is designed for the large influence scale.

The values are the combined values of the influence of the surrounding stones.

This is a different kind of influence from what my model considers. Maybe a different term can be used?

$$
$$ X X X O
$$ X . . O
$$ X . . O
$$ X X X O
Click Here To Show Diagram Code
[go]$$ $$ X X X O $$ X . . O $$ X . . O $$ X X X O[/go]

Each inside intersection is Black's 1-territory. After a black play, it reverts to your first example.

RobertJasiek · Post by **RobertJasiek** » Thu Feb 23, 2012 1:51 am

Magicwand asked whether particular fundamental findings (such as rules-related or ko definition) of mine are useful for getting stronger and I have replied that they are a bit useful in that respect. This misses two major points though:

1) The major purpose of fundamental research is to lay the basis of later applied research, which then can have much impact on getting stronger or (in case of science) making everybody's life richer. Einstein's general relativity theory was fundamental research. Later applied research enabled the installation of satellites with which GPS has become possible. My Japanese 2003 Rules are, for the sake of go theory, mostly fundamental research. Later reuse of their "force" concept has made practically useful definitions like n-connected and, relying on them, definitions of influence and thickness possible.

2) Not all my research has only marginal relevance for becoming stronger. While my fundamental research itself has an only marginal relevance for becoming stronger, my applied research, e.g., related to influence and thickness has great relevance for players becoming stronger. Of course, this does not include you, Magicwand, because you have an aversion against everything different from intuition and reading. Everybody open for conceptual learning can profit a lot though. a) My characterisation of influence and thickness to depend on connection, life and territory is much clearer than what existed before as a rough explanation: "nearby friendly stones bring an area under that player's influence". b) My degrees of influence and thickness allow whichever precision is needed while before the most precise descriptions were like "more and nearer friendly stones mean greater influence".

For other topics or concepts described by me, similar observations can be made.

Magicwand · Post by **Magicwand** » Thu Feb 23, 2012 11:05 am

[Admin]
MW expressed some concerns about Jasiak's skills, in a manner which was in violation of the TOS.
-JB
[/admin]

daal · Post by **daal** » Thu Feb 23, 2012 11:25 am

Magicwand wrote:...something in violation of the TOS...-JB

I can only conclude that either:

a) You have not read the TOS

or

b) You want to read more posts by Robert.

RobertJasiek · Post by **RobertJasiek** » Thu Feb 23, 2012 11:44 am

Magicwand wrote:...something in violation of the TOS...-JB

Ignorance of contents does not make it crap but means that the ignorants do not enable themselves to judge from first hand.

Everybody is invited to discuss the contents of my books!

My books (those I have published so far) are not research (a documentation of the research process) but contain a lot of my research's results. Therefore your statement "what you call research" is wrong.

show me some proof where others think highly of your research

I do not know if a comment like "Robert is a very careful and methodical researcher."
http://www.lifein19x19.com/forum/viewto ... 756#p48756
qualifies for your perception of "highly".

In general though public(!) comments really appreciating my (or, for that matter, any other go researcher's) research are sparse. I do not know why. Maybe consumption is so much easier than acquiring knowledge with which to properly identify contents as being new and comparing it with other literature.

EDIT: There is systematic observation though: Many more comments are about the fine details of English grammar than about factual research mistakes, if any. E.g., my ko definition paper still awaits a hint of any first factual mistake.

...something else in violation of the TOS...-JB.

You do ignore evidence presented earlier.

there are many non professionals tried but all failed.

You do ignore evidence presented earlier. Let me repeat: I have learned more from some weaker players (e.g. Andre Engels, then about 1 kyu) than I have learned from most professionals of those each having spent more time to teach me. James Davies (said to be about my strength) made me a lot stronger when I was a kyu.

...more in violation of the TOS...-JB.

You have never shown as clearly before that you cannot judge.

Life In 19x19

A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence

Re: A vague treatise on influence