Ideas for Japanese-style rules

RobertJasiek · Post by **RobertJasiek** » Sat May 29, 2010 12:36 pm

The number of legal positions is ca. 1.2 * 10^170, i.e., orders of magnitude larger than the number of atoms in the universe. Those poses serious storage problems for a trivial approach of setting one exceptional rule per position. While it is in principle mathematically trivially possible, our physical environment is the problem for having Japanese style rules without an evaluation phase but with special extra rules. Or have you meant also withOUT precedents?

If compromises are being made and Japanese rules are simplified, the we can use an approach like the Simplified Japanese Rules or even more radical departures like the Simplified WMSG 2008 Rules. Of course, it becomes less and less similar to Japanese rules then.

HermanHiddema · Post by **HermanHiddema** » Sat May 29, 2010 12:52 pm

RobertJasiek wrote:The number of legal positions is ca. 1.2 * 10^170, i.e., orders of magnitude larger than the number of atoms in the universe. Those poses serious storage problems for a trivial approach of setting one exceptional rule per position. While it is in principle mathematically trivially possible, our physical environment is the problem for having Japanese style rules without an evaluation phase but with special extra rules. Or have you meant also withOUT precedents?

I think the Japanese approach has, in general, been decidedly local w.r.t. life & death issues, so precedents on global positions are not really Japanese style. (They are, one might say, unprecedented, if you'll forgive me the pun

)

If compromises are being made and Japanese rules are simplified, the we can use an approach like the Simplified Japanese Rules or even more radical departures like the Simplified WMSG 2008 Rules. Of course, it becomes less and less similar to Japanese rules then.

There are, of course, all sorts of ways to simulate a Japanese style territory score with combinations of area style rules, pass stones, buttons, etc. But I wouldn't consider that Japanese style rules.

Cassandra · Post by **Cassandra** » Sat May 29, 2010 1:13 pm

RobertJasiek wrote:First do some homework: Read and understand all my webpages on Japanese rules! Then correct your signature and replace it by a reference to finding written rules that explain professional Japanese rules, a problem existing from 1603 (or earlier) to 2004:)

Thanks for your suggestions, Robert, but I will not change my signature. I suppose the search for the solution of Igo Hatsuyoron's #120 more difficult than that for an consistent rule set. But perhaps I have found the solution already, not proven by professionals, who knows ? ;-)

What remembers me of some work I would have had to to with it ...

I thought long about the fact, why our discussions erupt from time to time.
Let me try finding an explanation.

When I look back at your rules texts (I haven't read them in toto, they are too complex for me), I feel that you are driven by something I would name "search for the origin of life".

To define "life" in the first instance, is the hard and long way in my opinion. You have come to a conclusion now, and I think a valid one. But your texts are written in a mathematical / logical / scientifical style, what perhaps is necessary to reach their aim after eons of intens research, so that the wording is not appropriate for the "normal" reader. In addition, starting with "life" makes it necessary to develop a clear definition for "eye". What explains why you are so keen with it.

It is not my intention to replace your definition of "life" with a new one of mine. Once I started with questions of "life", too, but I have given up, because I recognized that there is a simpler path (for me).

Evaluating the status of each chain is sufficient.

"Territory is what is inside a living group." and "When there is territory, you will have a living group." seems to be the same, but there is a tiny, but decisive difference.

If you are looking for "Will contain territory." and "Will become prisoner.", there is no need to have a definition of "life". The board point named "eye", when coming from "life", gets its value by being forbidden for the opponent. There is no trouble with "'lives' in Seki, but contains no territory."

If you think (and you are the only one who can judge it) that what I have written (OK, the wording could be better) could possible give the same results as what is your intention with your rules (may be after some tiny supplement and / or correction), you have the chance to get YOUR contents in a "language" probably more user will understand. There will be always the "scientifical" text in the background to prove that the "colloquial" one is correct and consistent.

What I meant with the correlation of "stone" and "chain" is the following and is related especially to evaluated chains, which can be captured.

All of the primary points of the chain must end in the same status, do you want to get "Will contain territory" or "Will become prisoner" for the chain. "Must become part of a '2-eyed' chain." (for "Will contain territory" for a capturable chain) had been my first intention, but I didn't dare, because I feared your comment on a wording like that. With this slight editing, some of my diagrams would change.

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

would now end with

Click Here To Show Diagram Code: [go]$$ $$ |------------ $$ |W C . . O . $$ |. . . . O . $$ |O O O O O . $$ |M O M O O . $$ |O O O O O . $$ |. . . . . .[/go]

And your example 0000 would be

Click Here To Show Diagram Code: [go]$$ $$ |------------------- $$ |Q Q Q Q Q Q Q Q . | $$ |Q Y Y Y Y Y Y Y Q | $$ |Q Y . Y . Y Y Y Q | $$ |Q Y Y Y P Y Y Y Q | $$ |Q Y . P Y P . Y Q | $$ |Q Y Y Y . Y Y Y Q | $$ |Q Y Y Y . Y Y Y Q | $$ |Q Y Y Y Y Y Y Y Q | $$ |. Q Q Q Q Q Q Q Q | $$ |-------------------[/go]

Harleqin · Post by **Harleqin** » Sat May 29, 2010 4:49 pm

HermanHiddema wrote:Is it even theoretically possible to have Japanese style rules without an evaluation phase with special extra rules?

It depends on what you call "japanese style". If you want "locality", then no, but I am convinced that "locality" is a deeply flawed concept (and not "traditional" either!). However, if you just want "pure" territory scoring, I think that it should be possible. I actually made an attempt at that: After play stops, in the case of dispute, there is a single, global, hypothetical playout, in which the players determine control of the intersections. This control is then applied to the final position, removing all stones that are on an intersection controlled by the opposing colour. The implications have not been extensively explored yet, however.

Yes, it is a choice, but it is a wrong choice. I am sorry for the explicit wording.

Instead of saying "not acceptable" and "wrong choice", isn't is just better to write opinions like "I don't like", "I wouldn't want"? Or add "In my opinion" to your writing. These things are, after all, just a matter of taste and opinion.

Well, in my opinion, that opinion is wrong. What else can I write then? The alternatives do not have equivalent merit, therefore I just cannot lean back and say "it's just a matter of taste".

RobertJasiek · Post by **RobertJasiek** » Sat May 29, 2010 11:35 pm

Herman, that a score for the global position is assigned does not make the nature of the score global. Of course, it could be a score determined from local properties.

Harleqin, one global playout sequence is possible like in the Simplified Japanese Rules or like in New Amateur-Japanese Rules with one instead of two analysis sequences. Of course, there are very scarce differences in case of kos. Those scarce positions would differ from current Japanese tradition. The difference is relatively small though so that one can still speak of Japanese style rules. But one changed score every 1,000th to 50,000th (I can only guess the frequency) game is way beyond what Japanese traditionalists want to bear... They consider this more important than practical inapplicability of the rules as rules.

RobertJasiek · Post by **RobertJasiek** » Sun May 30, 2010 12:19 am

Cassandra, the search for a consistent ruleset is a matter of seconds. What took so long was the research for a written ruleset explaining professional Japanese rules tradition.

Our discussions become hot when a) you refuse to read something essential, b) you ignore already found facts, c) you write mistakes or things remaining unclear, d) you confuse opinion with facts etc.

If you are not able to understand my rule texts entirely, then you can try to understand at least parts. E.g., you might ignore everything on a lower definition level than "force" in J2003. However, spending less than 2 weeks on understanding, e.g., J2003 is a mistake anyway (unless you would understand it faster). You should not expect to understand it faster given that I needed 10 years to find and 11 months to write it.

If you do not understand something, then why do you not ask? Instead you do not even dare to read far enough to know what you might want to ask.

Start with the simplest Japanese style ruleset: the Simplified Japanese Rules. Are you able to understand them?

You want rules to be written like prose. You might have to be very patient until hopefully I write my 10+ volumes Go Rules Encyclopedia, which should explain everything also for the laziest reader. Writing such a commentary requires 2 to 6 weeks per Japanese style ruleset. Since you expect me to spend more time on writing on a lazy-user-friendly level than you want to spend on reading, you must wait and then pay for the books.

BTW, there is already an informal text version of J2003:
http://home.snafu.de/jasiek/j2003inf.html

Maybe you should accept the fact that Japanese style rules are more difficult than you want to invest energy in them?

You say "starting with 'life' makes it necessary to develop a clear definition for 'eye'". This shows that you have not understood even the rough basic structure of my Japanese rulesets yet. Neither J2003 nor my life model of WAGC Rules define or would need to define "eye"!!! I had to write an additional strategic level pile of definitions, where I defined eye; this is not used for the rules at all.

The J1989/J2003 life is not defined via "eye" but via capturability. Also for the definition of territory, "eye" is not needed but "eye-point" is used, which is a DIFFERENT concept. - The WAGC life model is also not defined via "eye" but via "2-eye-formation".

Although you take great pains to pretend not using "life" in your rules, what you actually do is to define forms of life: 2-eyed and stable are life while removable is dead.

Apart from ko rules, as I have said, your current life definition is like J2003-alive without capturable-2, which is removable / dead under your rules.

"All of the primary points of the chain must end in the same status": There I agree (for practically useful rulesets).

I do not understand this text of yours: "do you want to get 'Will contain territory' or 'Will become prisoner' for the chain. 'Must become part of a '2-eyed' chain.' (for 'Will contain territory' for a capturable chain) had been my first intention [...]. With this slight editing, some of my diagrams would change." I also do not understand your application of this text on the following diagrams. Please explain both the text and its application to the diagrams carefully!

Cassandra · Post by **Cassandra** » Sun May 30, 2010 2:03 am

RobertJasiek wrote:If you are not able to understand my rule texts entirely, then you can try to understand at least parts. E.g., you might ignore everything on a lower definition level than "force" in J2003. However, spending less than 2 weeks on understanding, e.g., J2003 is a mistake anyway (unless you would understand it faster). You should not expect to understand it faster given that I needed 10 years to find and 11 months to write it.

I think, I will start with your treatment of 1989 Nihon Kiin rules example #16, which I have found in your "j1989c".

This for two reasons:
First, I suppose that it will be easier for me to grasp the deeper meaning of your termini when I try to understand their application on a concrete position.
Second, you come to the same conclusion as the Nihon Kiin rules (what differs from what I had posted). #16 is one example of what I meant with the complaint that there is no clear distinction between "Evaluate" and "Count". "Dies trough the collapse of the Seki" can also be found in #17 and #18. "Collapse of the Seki" means "taking stones off the board", what in my modell can only happen during "Count".

You want rules to be written like prose.
...
BTW, there is already an informal text version of J2003:
http://home.snafu.de/jasiek/j2003inf.html

No. I would find a user-friendly version of the "correct" rules especially suitable.

I mean "user-friendly", not "informal". ;-)

Maybe you should accept the fact that Japanese style rules are more difficult than you want to invest energy in them?

May be true. #120 takes more that enough time.

Although you take great pains to pretend not using "life" in your rules, what you actually do is to define forms of life: 2-eyed and stable are life while removable is dead.

"removable" inside "2-eyed" is dead.

I do not understand this text of yours: "do you want to get 'Will contain territory' or 'Will become prisoner' for the chain. 'Must become part of a '2-eyed' chain.' (for 'Will contain territory' for a capturable chain) had been my first intention [...]. With this slight editing, some of my diagrams would change." I also do not understand your application of this text on the following diagrams. Please explain both the text and its application to the diagrams carefully!

I understood your example 0000 (concerning all of a captured chain's primary points) that my privious "must get 2 'taboo-points'" (what is equivalent to "get connected to a '2-eyed' chain") must be complemented by "or be turned into one of those".
So you can stop the evaluation sequence when it is clear that all of the primary points lay inside a "2-eyed" group. And in my solution to your #0000 Tengen turns into "2-eyed".

RobertJasiek · Post by **RobertJasiek** » Sun May 30, 2010 2:53 am

Starting with #16 is about the worst choice you could make. It has a too rare shape to be relevant for practice. The diagrams in my commentary are missing online due to difficulties with my text and HTML editors. Why don't you start with the simplest examples?!

"Collapse of the Seki" is terrible in the official J1989 commentary because it is entirely unrelated to the J1989 rules. They do not have any term or concept "Collapse of the Seki"! They do not even define "seki"!

The correct rules are the Japanese 2003 Rules, version 35a. The original Japanese 1989 Rules are incorrect in many respects, see, e.g., my commentary on them. So when you want to see a user-friendly version of the correct rules, read the commentary on Japanese 2003 Rules: http://home.snafu.de/jasiek/j2003com.html

Maybe I will post a user-friendly summary on how to read the J1989 Rules.

Please show me in every detail how your (modified?) rules turn Example 0000's tengen into something "2-eyed"!

Cassandra · Post by **Cassandra** » Sun May 30, 2010 4:10 am

RobertJasiek wrote:Starting with #16 is about the worst choice you could make. It has a too rare shape to be relevant for practice. The diagrams in my commentary are missing online due to difficulties with my text and HTML editors. Why don't you start with the simplest examples?!

It will be important for me to recognise where the difference in evaluation comes from. Maybe thereafter I will understand your terminology better than before.

The correct rules are the Japanese 2003 Rules, version 35a. The original Japanese 1989 Rules are incorrect in many respects, see, e.g., my commentary on them.

But I suppose that your evaluation of #16 (whole j1989c is as of 2007) will match the J2003 requirements ?

Cassandra · Post by **Cassandra** » Sun May 30, 2010 4:32 am

RobertJasiek wrote:Please show me in every detail how your (modified?) rules turn Example 0000's tengen into something "2-eyed"!

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

The evaluation sequences for the stone at Tengen are the same as for Black's group.

We evaluate one of Black's chains, so it's White's turn.

Click Here To Show Diagram Code: [go]$$W $$ |------------------- $$ |O O O O O O O O . | $$ |O X X X X X X X O | $$ |O X M X . X X X O | $$ |O X X X O X X X O | $$ |O X . O X O . X O | $$ |O X X X 2 X X X O | $$ |O X X X M X X X O | $$ |O X X X X X X X O | $$ |. O O O O O O O O | $$ |-------------------[/go]

If she does nothing, Black connects. Both the Tengen stone and the chain possess 2 "taboo-points" now.

Click Here To Show Diagram Code: [go]$$W $$ |------------------- $$ |O O O O O O O O . | $$ |O X X X X X X X O | $$ |O X . X . X X X O | $$ |O X X X O X X X O | $$ |O X . O X O . X O | $$ |O X X X 1 X X X O | $$ |O X X X . X X X O | $$ |O X X X X X X X O | $$ |. O O O O O O O O | $$ |-------------------[/go]

If she decides to move, she can only capture Black's stone at Tengen.

Click Here To Show Diagram Code: [go]$$W $$ |------------------- $$ |O O O O O O O O . | $$ |O X X X X X X X O | $$ |O X . X 6 X X X O | $$ |O X X X O X X X O | $$ |O X 4 O 3 O 8 X O | $$ |O X X X O X X X O | $$ |O X X X 2 X X X O | $$ |O X X X X X X X O | $$ |. O O O O O O O O | $$ |-------------------[/go]

Click Here To Show Diagram Code: [go]$$Wm9 $$ |------------------- $$ |O O O O O O O O . | $$ |O X X X X X X X O | $$ |O X . X X X X X O | $$ |O X X X 6 X X X O | $$ |O X X 4 1 8 X X O | $$ |O X X X 2 X X X O | $$ |O X X X X X X X O | $$ |O X X X X X X X O | $$ |. O O O O O O O O | $$ |-------------------[/go]

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

The sequence comes to an end

Click Here To Show Diagram Code: [go]$$W $$ |------------------- $$ |O O O O O O O O . | $$ |O B B B B B B B O | $$ |O B M B X B B B O | $$ |O B B B X B B B O | $$ |O B X X M X X B O | $$ |O B B B X B B B O | $$ |O B B B X B B B O | $$ |O B B B B B B B O | $$ |. O O O O O O O O | $$ |-------------------[/go]

Considering Black's chain, it has 2 "taboo-points" now.

Click Here To Show Diagram Code: [go]$$W $$ |------------------- $$ |O O O O O O O O . | $$ |O X X X X X X X O | $$ |O X . X X X X X O | $$ |O X X X X X X X O | $$ |O X X X M X X X O | $$ |O X X X X X X X O | $$ |O X X X X X X X O | $$ |O X X X X X X X O | $$ |. O O O O O O O O | $$ |-------------------[/go]

Considering Black's stone at Tengen, it has been captured (what results in subtype (III)) and been transformed into a "taboo-point".

Click Here To Show Diagram Code: [go]$$W $$ |------------------- $$ |O O O O O O O O . | $$ |O X X X X X X X O | $$ |O X M X X X X X O | $$ |O X X X X X X X O | $$ |O X X X C X X X O | $$ |O X X X X X X X O | $$ |O X X X X X X X O | $$ |O X X X X X X X O | $$ |. O O O O O O O O | $$ |-------------------[/go]

This "taboo-point" at Tengen has a "partner" in the top left, so it belongs to a "2-eyed" chain of Black.

RobertJasiek · Post by **RobertJasiek** » Sun May 30, 2010 6:15 am

Which kind of difference in evaluation, which you would like to understand, are you referring to? I might tell you what to read if you specify that.

Of course, #16 analysis under J2003 meets the J2003 requirements.

I think now I understand how you want your modified rules to apply to Example 0000.

Cassandra · Post by **Cassandra** » Sun May 30, 2010 6:30 am

RobertJasiek wrote:Which kind of difference in evaluation, which you would like to understand, are you referring to? I might tell you what to read if you specify that.

If I had read it correctly, your result of #16 is that all of White's stones are dead (as they are with the Nihon Kiin rules).

My evaluation ends in Seki for all.

There must be a reason for this difference and I would like to know it.

RobertJasiek · Post by **RobertJasiek** » Sun May 30, 2010 6:57 am

The #16 diagrams are in the attached PDF.

I am not sure though whether you search these, which use the J1989 hypothetical ko rule. If you want the J2003 application, read http://home.snafu.de/jasiek/j2003com.ht ... alPosition and decipher the lexically annotated hypothetical-sequences!

Cassandra · Post by **Cassandra** » Sun May 30, 2010 8:16 am

RobertJasiek wrote:The #16 diagrams are in the attached PDF.

I am not sure though whether you search these, which use the J1989 hypothetical ko rule. If you want the J2003 application, read http://home.snafu.de/jasiek/j2003com.ht ... alPosition and decipher the lexically annotated hypothetical-sequences!

Thanks.

I'm sorry, but I hadn't read this text yet, due to the missing diagrams. :oops:

So you also come to the conclusion that there is a fault in the Nihon Kiin rules' evaluation.

What about the "application of analytical Ko-pass rule" on example #16 further down in the text, which ends with all of White's stones being dead ?

Cassandra · Post by **Cassandra** » Sat Jun 12, 2010 11:34 am

After going through the examples of my classification.pdf in the "...."-tread (see xxxxx) I think that we can simplify the topic much more by using only one status to be found (or even not for its complement) at the end of the evaluation of a string (may be a single stone).

What follows, is my new suggestion for processing "Evaluate".

"Evaluate" will determine the status of each string individually.
Determining the status of each string begins with a move of the opponent (may be a Pass).

There are two statuses:

"2-eyed": All primary points of the string will become part of a 2-taboo-point-formation of the string's colour.
"not-2-eyed": At least one primary point of the string will not become part of a 2-taboo-point-formation of the string's colour.

Some definitions:

"2-taboo-point-formation" = set of strings (may be one), which include at least 2 taboo-points.
"taboo-point" = point of the board, where a stone of the opposing colour must not be placed according to the rules (i.e. a stone must not be placed so that it has no lifeline left), excluding any Ko rule.
Minimum length of a string is 1.

Basic Ko rule, also valid for evaluation is

A player must not place a stone that would immediately recapture the last opponent's stone placed in a Ko shape.

During "Count", "not 2-eyed" strings inside opposing "2-eyed" strings are taken off the board as prisoners. Thereafter, "territory" is each empty point inside "2-eyed" strings.

"... will become" can also be understood as "the opponent cannot force that ... do not become"

2-eyed strings that cannot be captured remain unaffected, as remain Uttegaeshi and Nakade.

Colloquially named "dead" strings will be turned into prisoners as before.

Strings in positions colloquially named "Seki" will all get the status not-2-eyed, therewith not being inside opposing 2-eyed-strings:

The evaluation sequence will be caught in a cycle.
The string will be captured, but not succeeded in toto on its primary points.
The string cannot be captured.

Life In 19x19

Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules

Re: Ideas for Japanese-style rules