I am fine with "unexpected" special effects on small boards, which are quite unlikely to be seen on a 19x19 board.

This is because the impact of the deciding fundamental difference between the "uncapturable" and the "two-eyed alive" worlds becomes the larger the smaller the board is.


I do not yet have a complete, ready-made framework for a set of "two-eyed alive" rules. That's why I'm still extremely flexible.
Some days earlier, I would have never dreamed of something like "J89 ko-bans must be lifted in order" or "cycles with (2 + 2n) moves are forbidden during status confirmation".

I think that a quick and dirty solution for your problem could be something like:
-- A "two-eyed alive" group that is neighboured to an opponent's "seki-alive" group does not contain any territory, if both groups share DAME. Opponent's "dead" stones inside must not be taken off the board as prisoners.

_________________
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)

Just allow territory in sekis: http://home.snafu.de/jasiek/sj.html

Dear Robert, this would not comply with Japanese tradition, would it?

However, I would not exclude this option per se, as "two-eyed alive" will result in special differences to the "uncapturable" world of e.g. J89 anyway. In positions, which are much more relevant than J89's example 4.

It leads us to the (here in this thread still to be discussed) question, what we (want to) understand by
"Having SOVEREIGNTY over an enclosed (and empty) part of the board."

The decision, whether we want to allow territory in seki, or not, is a direct consequence of that.
And it is independent of the ruleset used, because "identifying territory" (respective "... area") is only the second step, after all the work of determining the life-and-death status of all groups has been completed.


Now that you have opened this Pandora's box, we may want to remember that "Capturing stones" is NOT the aim of the game of Go.
Rather, the aim of the game of Go is to exercise sovereignty over a (n empty) part of the board.

"Uncapturable" is oriented towards "capturing stones", "two-eyed alive" is oriented towards "exercising sovereignty".

_________________
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)

Sorry Cassandra I think I did undersant your question (maybe to my bad english but I do my best)
Let me try again but I am not sure I am answering your question:

Click Here To Show Diagram Code

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

Surely J89 is not quite clear but J2003 is fine. All the corner is in seki and, as far as I am concerned I like this result. As a consequence it would have been better for black to not pass and to kill one of the white groups in normal play.

BTW the original position given by the author is the following:

Click Here To Show Diagram Code

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

Yes of course it is quite unlikely that such position appears in a real game.

In addition Cassandra you can look at the following very interesting article written by Robert M. Pauli : http://ljrg.eu5.net

Please don't worry.

Was just a tiny test for you

I knew that the transposition onto a 19x19-board was possible.
There are four other ones available...

_________________
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)

Probably it will become clearer with splitting the sequence in the third diagram of J89's "proof":

Click Here To Show Diagram Code

[go]$$B
$$ +----------------------
$$ | . X X O . X X O X . .
$$ | X X O O O O O O X . .
$$ | . O X X X X X X X . .
$$ | O O X , . . . . . , .
$$ | . O X . . . . . . . .
$$ | 6 O X . . . . . . . .
$$ | . O X . . . . . . . .
$$ | O O X . . . . . . . .
$$ | X X X . . . . . . . .
$$ | . . . , . . . . . , .
$$ | . . . . . . . . . . .[/go]

makes White's group at the left "alive".

Click Here To Show Diagram Code

[go]$$B
$$ +----------------------
$$ | . X X O 7 X X O X . .
$$ | X X O O O O O O X . .
$$ | . O X X X X X X X . .
$$ | O O X , . . . . . , .
$$ | . O X . . . . . . . .
$$ | O O X . . . . . . . .
$$ | . O X . . . . . . . .
$$ | O O X . . . . . . . .
$$ | X X X . . . . . . . .
$$ | . . . , . . . . . , .
$$ | . . . . . . . . . . .[/go]

is the FINAL move of J89's "proof".

Click Here To Show Diagram Code

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

Something has just been captured at the marked points at the top.
But the story does NOT continue.

Capturing White's stones at the top "enabled" NOTHING.
Therefore, White's stones at the top have been status confirmed "dead", not "alive".

Click Here To Show Diagram Code

[go]$$B
$$ +----------------------
$$ | . X X T X X X T X . .
$$ | X X T T T T T T X . .
$$ | . O X X X X X X X . .
$$ | O O X , . . . . . , .
$$ | . O X . . . . . . . .
$$ | Q O X . . . . . . . .
$$ | . O X . . . . . . . .
$$ | O O X . . . . . . . .
$$ | X X X . . . . . . . .
$$ | . . . , . . . . . , .
$$ | . . . . . . . . . . .[/go]

Quite apparently, (was ) has been mistakenly seen as the "new" move that cannot be captured.
However, this move was played much too early to be counted.

_________________
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)

Very interesting approach, especially the idea of "removing a disturbing cycle".

But too long to be analysed in total now
Especially did not reach the status confirmation section ...


Nevertheless, just a quick and dirty comment on "removing a disturbing cycle".
If I understood this concept, and its conditions, right in a hurry, this rule -- as it is -- would result in a different game.

It can be easily overlooked that cycles with a length of 2n moves (n = 1, 2, 3, ...) have completely different properties than cyles with a length of (2 + 2m) moves (m = 0, 1, 2, ...).

As far as I can see now, "removing a disturbing cycle" should be restricted to the latter ones (don't know whether "m = 0" -- i.e. basic ko -- would be a special case here that had to be excluded -- and so be left with the basic ko rule alone).

_________________
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)

Let me discuss the Simplified Japanese Rules.

They do not agree to tradition for all positions for especially these reasons:
1) Life status is determined globally instead of mostly locally.
2) Life status is determined by skill instead of perfect play.
3) Sekis have territory instead of no territory.
4) Superko is used instead of basic ko and no result ko rules.
5) There are no instead of some exceptional rules for rare exceptional shapes.

In my opinion, (1) and (2) are good because go is a game of global play and skill.

On servers, players accept territory in sekis the most easily. Therefore, there is no need to let Japanese professionals dictate the opposite as an exception. Amateur players using Japanese style rules in real world games do not have a strong preference for this tradition.

I have chosen superko for the sake of simple rules design but it is not necessary to impose superko on Japanese style rules. Other ko rules can be used. Basic ko and no result ko rules can be used but require additional clarification or ko rules for life status "confirmation". E.g.: "Each pass enables basic ko capture. Long cycles are interpreted as follows: [...]"

Ordinary shapes, such as bent-4-in-the-corner and other hidden kos, do not need exceptional rules. Mostly, exceptional rules are relevant only for triple kos and similar rarities. Exceptional rules for rare exceptional shapes are not needed because one may tolerate the impact of ordinary rules on them. Mostly (especially Japanese and Korean) professional players want to maintain what they call tradition in the form of specific outcomes of some specific rare exceptional shapes. If they had studied all possible rare exceptional shapes, they would be forced to reflect their related inconsistent tradition, except that they rather stick their nose into the sand rather than admit inconsistency.

This is true, but certain rumours about "bent-4-in-the-corner is (always) dead" (what is not true) seemed to have brought some confusion into this subject.


This is why I suggested tailoring such "exceptional" rules as specific as possible (and e.g. not trying to establish an additional SIMPLE-ko rule for solving a TRIPLE-ko problem, as J89 did).


If I remember correct, this is what I tried to explain to you for decades

((This kind of tolerance might also help when comparing different status confirmation results between different rulesets.))


This desire is not bad per se.


However, if specific RULES are tailored to achieve these wanted results of status confirmation, these must be applied logically closed and without contradiction. J89's specific rules do not meet this condition, as we all know for sure.

Therefore, if the wanted outcomes are too contradictory to be solved by a (ONE!) specific mechanism by the rules, it would be better to keep the rules free of these, and add a "table of exeptional cases" as a supplement to the legal text.

_________________
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)

I think you (and many Western rules hobbyists) are on the wrong track here.

It is true that J89 text has many problems and, if applied literally, many flaws. But this is the problem of the text, careless wordings, oversights, etc.

The LOGIC behind the rules, however, is a different thing. There are not many things that just "happen" to be like they are, based on tradition etc. Most things have good reasons, J89 authors just haven't find good formality for them. It is no coincidence that when the text and examples contradict the examples (and contemporary pro practice) take priority.

What you consider "exceptional cases" are not exceptional for their rules logic, only for the (not too successful) textual formalization.

"Simplified Territorial Rules" would have been a better title.

"Japanese" and "territory in seki" is a contradiction in terms. Let me explain why (in my personal opinion) ...


In the "Chinese tradition" (i.e. for "area rule", to be clear), the fortifications are valued to the same extent as their contents.
Therefore, it is only locigal that the inner part (if there is such) of a fortification, which finds itself being a compound of a seki, is also taken into account in order to determine the outcome of the game.
=> "Territory" is counted in a seki.


The "Japanese tradition" seems to assess the fortifications as a necessary evil. In a certain sense, they are mandatory, but otherwise take only up space, which could be otherwise utilised much better.
In addition, it seems to be grounded on a principle of "absolute independence".
Therefore -- after Black placed a stone into a White eye-point -- it makes an all-decicive difference, whether WHITE's group can still "breath" because there is another eye point inside WHITE's group, or whether BLACK was so very kind to not occupy all outside liberties of her group (for what reason ever).
=> Territory is NOT counted in a seki.

_________________
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)

In this case, they should not have neither choosen "examples of confirmation of life and death" as the headline of that supplement to the legal text, nor "The results in the following examples would be reached through confirmation of life and death" in that supplement's introduction.

As additional (implicite) rules for status confirmation, which are not listed in the legal text, we have identified up to now

"Enabled" means permament occupation of any board point, which is possible in no other case but after the group under status confirmation has been taken off the board.

"Pass-for-ko bans" must be played in the same order as the respective ko have been captured. Or (/ and) ...
The completition of a double-ko cycle is forbidden.

As I coincidentally found in the current original version of J89 at the Nihon Kiin site, ...

Click Here To Show Diagram Code

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

... this compound of "bent-four" and "double-ko with nested ko shapes for the sandwiched group" is really life-and-death example 18.
As with ALL compounds of "bent-four" and "double-ko", NO additional rule is needed to achieve the intended status (this is what Robert also mentioned in his posting above).

However, you would have to find implicite rule for confirming White's double-ko group below dead, what of cource can be expected to be the intended result.

Click Here To Show Diagram Code

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

This is a variation of example 16, having the nested type of double-ko, instead of the simple one combined with a solid eye-point each.

-------------------
EDITED

_________________
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)

No.

Japanese (or Korean) rules have an extra status assessment phase with hypothetical play. Simple territory scoring rules would have ordinary play for status assessment but whether pass fights occur is the problem. (Button go is yet another thing.)

This one is pretty obvious. Just a side effect / oversight of formalization, which - as is - would break way too many things.


Hypothetical play is necessary for any pure territory scoring rules, actual playout (with pass stones or similar) is partial switch to area scoring.

I also don't see why you call your rules "Japanese", but for a different reason: draw on repetition in triple kos and similar cycles is not ignorable rarity but essential in Asian go.

If so, rules (or / and illustrative examples or illustrative examples alone) that were specific for triple-ko and for any of the "similar" cycles should have been created.

I am sure, you will not use pears for baking an apple-pie.

_________________
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)

"Simplified" will be a reference to a comparison with the legal text of J89 and / or J2003.
Just because simpler than "simple" cannot be.

Therefore, "simple territory scoring" PLUS "territory in seki" PLUS "status assessment" would be equivalent to "extended simple territorial rules".

_________________
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)

I'm not sure what you mean here. You won't find examples for this among the LD example section (since long cycle games won't reach any end position where LD would be in question), but J89 still contains illustrative examples at commentary on "no result".

Also player awareness. Void games are so widely known, the other day my opponent tried to claim that for a sending-two-returning-one. So players learn about long cycle draws even before they recognize real cycles.

Probably you have already noticed that I edited my posting above.

The previous is now (prohibiting triple-ko cycles was an idea to make the ko-ban for simple ko superfluous), re-worded to "The completition of a double-ko cycle is forbidden."

Probably you want to reconsider your demand for such a prohibition.

Getting rid of a cycle means that the terminating move of that cycle has to be forbidden, NOT the initiating one.
I am sure that you will not like the resulting consequences.

Why the terminating move?

Choosing the INITIATING move for that purpose would have the following consequences (among others):
-- Application case "simple-ko": You must not capture into any ko shape.
-- Application case "double-ko": You must not capture into any double-ko shape.

J89 shows several sequences of capturing into a double-ko shape. Thus, getting rid of that cycle cannot be intended.

#####################################

Apparently, the impact of a ban for every single-ko onto a double-ko flew under the radar of the J89 authors. But the features of a double-ko were changed dramatically.

During "play", it is the player who captured the SECOND ko, who has the priviledge of recapturing first.
If he does, the priviledge for the next cycle switches to the other player.
This implies that a double-ko (seen in isolation) during play is a 50:50 issue.
During this cycle, both players played one move each elsewhere.

But during "status confirmation", it is the player who captured the FIRST ko, who has the priviledge of recapturing first.
This implies that a double-ko (seen in isolation) has become a 100:0 issue.
During this cycle, no move is played elsewhere, what makes an endless repetition possible.
If there is NO endless repetition, the double-ko has effectively disappeared.

It is evident now that Robert's one-pass-for-all-ko-bans in J2003 eleminated that difference between "play" and "status confirmation" for the application case "double-ko".

_________________
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)