You don't say!Cassandra wrote:However, it is not at all unusual for people not to want to hear the truth.
And don't forget, 雨降って地固まる
GERMAN interpretation of J89's intended contents
-
CDavis7M
- Lives in sente
- Posts: 716
- Joined: Tue Jun 15, 2010 2:18 pm
- Rank: Shokyu
- GD Posts: 0
- Universal go server handle: CDavis7M
- Has thanked: 109 times
- Been thanked: 140 times
- Contact:
Re: GERMAN interpretation of J89's intended contents
And don't forget, 雨降って地固まる
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: GERMAN interpretation of J89's intended contents
Continued.RobertJasiek wrote:How / with which principles have you derived / created the German version?
Approach #6: Ambiguities brought about deliberately, which can only be removed again with a very high level of effort. Contd.
"or if capturing them"
(or whatever tranislation you prefer for "又は取られても")
We know from the past what is aimed for in the game of Go.
Cassandra wrote:VALUE can only be created within called-group-by-the-common-people formations that either already contain at least two permanently-for-the-opponent-forbidden-board-point construction elements, none of which is the sole connection between the called-group-by-the-common-people formation's stones along the lines of the board, which are compatible to called-true-eye-by-the-common-people construction elements, or that can be transferred into such a called-group-by-the-common-people formation with at least two permanently-for-the-opponent-forbidden-board-point construction elements, none of which is the sole connection between called-group-by-the-common-people formation's stones along the lines of the board, which are compatible to called-true-eye-by-the-common-people construction elements, even if the opponent plays first, and which enclosed board intersections are either unoccupied or occupied only by opposing can-be-made-to-disappear-permanently-even-if-the-opponent-plays-first called-group-by-the-common-people construction elements, which are compatible to called-dead-by-the-common-people called-group-by-the-common-people construction elements, which are compatible to called-two-eye-formations-by-some-rule-specialists construction elements, as well as to called-independently-alive-by-the-common-people called-group-by-the-common-people construction elements.
Board points permanently occupied by the stones of a player can be considered "owned" by this player / this player's stones. 
A region on the board enclosed by such stones can be considered "owned" by this player / the player's stones if it is either unoccupied or every fenced opposing stone could be permanently removed. 
A group of stones that includes at least two such "owned" regions (either of which includes a board point, which occupation is permanently prohibited for the opponent), none of which is the sole connection between "owned" board points along the lines of the board, even if the opponent is allowed to move first, can be called "two-prohibited-point-formation".An exemplary case.
Q.e.d #1.
Q.e.d. #2.
----------
Counter-example.
All the stones forming a group of stones that ALREADY IS such "two-prohibited-point-formation" are firmly connected to each other, exclusively either directly next to each other or via unoccupied points along the lines of the board.Exemplary case #1.
Exemplary case #2.
A group of stones that is not yet a "two-prohibited-point-formation" is successfully TRANSFORMED into a "two-prohibited-point-formation" if all the board points that were occupied or enclosed by the group of stones are still controlled by the resulting "two-prohibited-point-formation" after the transformation process.An exemplary case.
Q.e.d. #1.
Q.e.d. #2. A variation for Black.
----------
All you need is the distinction between "permanent" and "non-permanent".
To simplify the count at the end of the game, the "non-permanent" in the "permanent" can be taken off the board to make its way into the corresponding prisoner's lid.
No "or if capturing them" is required!
No "seki" is required!
----------
Overdetermination is an evil!
"Every fourth even number that is dividable by 8, 4, and 2, shall be regulated by ..."
This regulation refers to the set 8, 16, 24, 32, ...
However, that set is overdeternined!
Being dividable by 8 is an already existing property of every fourth even number.
Being dividable by 4 is a property of numbers, which are dividable by 8.
Being dividable by 2 is a property of numbers, which are dividable by 8, as well as by 4.
An even number is defined by being dividable by 2.If a regulation for this set of numbers starts with "If a number is dividable by 4, then ...", then 4, 12, 20, ... are also included in the regulations, but must later be filtered out again at great expense.
----------
J89 uses a different concept for determining "VALUE".
The property "being considered alive" is given to TWO DIFFERENT types of groups, one of which can enclose value (commonly known as "independently alive group") but the other cannot (commonly known as "seki").
The property "can be taken" applies to TWO DIFFENRENT types of stones commonly understood as "being alive", as well as to ANOTHER type of stones commonly understood as "being dead".
This results in
"being considered alive" AND NOT "can be taken" contains value. "being considered alive" AND NOT "can be taken" does not contain value. "being considered alive" AND "can be taken" contains value. "being considered alive" AND "can be taken" does not contain value.
NOT "being considered alive" does not contain value.It should be easy to see that a further property is absolutely necessary in order to be able to differentiate between
and as well as and .The property chosen is "possesses dame" (or whatever tranislation you prefer for "駄目を有する").
The list above becomes in a first step:
"being considered alive" AND NOT "can be taken" contains value. "being considered alive" AND NOT "can be taken" AND "possesses dame" does not contain value. "being considered alive" AND "can be taken" contains value. "being considered alive" AND "can be taken" AND "possesses dame" does not contain value. NOT "being considered alive" does not contain value.However, the property "possess dame" is also given to two
of different colours, which share unoccupied board points on their outside.Thus, the list above finally becomes:
a) "being considered alive" AND NOT "can be taken" AND NOT "possesses dame" contains value. b) "being considered alive" AND NOT "can be taken" AND "possesses dame" does not contain value. "being considered alive" AND NOT "can be taken" AND "possesses dame" does not contain value. a) "being considered alive" AND "can be taken" AND NOT "possesses dame" contains value. b) "being considered alive" AND "can be taken" AND "possesses dame" does not contain value. "being considered alive" AND "can be taken" AND "possesses dame" does not contain value. NOT "being considered alive" does not contain value.To get rid of this not very likely unintended side-effect (of
b) and b)), the players are implicitely enforced to occupy such shared unoccupied points outside their "independently alive" groups before the final counting starts.----------
To be continued ... (simply scroll down)
Last edited by Cassandra on Sat Oct 23, 2021 10:17 pm, edited 3 times in total.
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: GERMAN interpretation of J89's intended contents
Continued.RobertJasiek wrote:How / with which principles have you derived / created the German version?
Approach #7: Verification of life & death and clarity of wording.
We know from the past what the Japanese understanding is of the connection between "life and death" and "territory".
Cassandra wrote:
a) "independently alive", no capture, no "dame"White's group is "independently alive".
It is technically impossible to capture this group, thus it "cannot" be captured.
The correct wording for the status description at end of the verification of L&D is "it is not captured".
Two board points are White territory.
----------
White's group is "independently alive".
It is technically possible to capture this group, thus "it cannot be captured" is mistaken.
But it "will not" be captured.
The correct wording for the status description at end of the verification of L&D is "it is not captured".
Three board points are White territory.
b) "independently alive", no capture, "dame"No territory.
Occupy the "dame" 
. Goto a). "seki", no captureWhite's and Black groups are "alive", forming a "seki".
It is technically possible to capture Black's "alive" group, thus "it cannot be captured" is mistaken.
But it "will not" be captured.
The correct wording for the status description at end of the verification of L&D is "it is not captured".
The same applies for White's "alive" group.
Both groups share a "dame"
, so there is no territory.----------
White's and Black groups are "alive", forming a "seki".
It is technically possible to capture Black's "alive" group, thus "it cannot be captured" is mistaken.
But it "will not" be captured.
The correct wording for the status description at end of the verification of L&D is "it is not captured".
The same applies for White's "alive" group.
Both groups share two "dame"
, which implies that both sides surround something opposing "alive", so there is no territory.----------
To be continued ... (simply scroll down)
Last edited by Cassandra on Sat Oct 23, 2021 10:17 pm, edited 2 times in total.
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: GERMAN interpretation of J89's intended contents
Continued.RobertJasiek wrote:How / with which principles have you derived / created the German version?
Approach #7: Verification of life & death and clarity of wording. Contd.
We know from the past what the Japanese understanding is of the connection between "life and death" and "territory".
Cassandra wrote:
a) "independently alive", capture, no ko, no "dame"White's stones are "independently alive".
In the following, we will limit ourselves to the two White stones at the top.
It is not obligatory for Black to take these stones off the board during the verification of L&D.
At the end of the verification sequence, White still has stones on all the board points that were occupied by her stones in the very beginning.
Variation for Black.
At the end of the verification sequence, White still has stones on all the board points that were occupied by her stones in the very beginning.
Six board points at the top are White territory.
----------
White's stones are "independently alive".
In the following, we will limit ourselves to the three White stones in the upper left corner.
It is not obligatory for Black to take these stones off the board during the verification of L&D.
At the end of the verification sequence, White still has stones on all the board points that were occupied by her stones in the very beginning.
Variation for Black.
At the end of the verification sequence, White still has stones on all the board points that were occupied by her stones in the very beginning.
Six board points in the upper left are White territory.
b) "independently alive", capture, no ko, "dame"No territory.
Occupy the "dame" . Goto a).----------
To be continued ... (simply scroll down)
Last edited by Cassandra on Sat Oct 23, 2021 10:16 pm, edited 2 times in total.
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: GERMAN interpretation of J89's intended contents
Continued.RobertJasiek wrote:How / with which principles have you derived / created the German version?
Approach #7: Verification of life & death and clarity of wording. Contd.
We know from the past what the Japanese understanding is of the connection between "life and death" and "territory".
Cassandra wrote:
"seki", capture, no koPatients in intensive care: survival rate ZERO!
Exemplary case #1.
Black moves first in the corner.
Black captured five White stones and got one point of territory, so his total value is six points.
White captured two Black stones and got one point of territory, so her total value is three points.
Black's net advantage is three points.
White moves first in the corner.
Black captured six White stones and got one point of territory, so his total value is seven points.
White captured two Black stones but got no territory, so her total value is two points.
Black's net advantage is five points.
Uke: "It's quite confusing, isn't it? I'm afraid there will be disputes."
Tori: "I think the most elegant solution would be to declare the stones in the corner "dead" at the end of the game. After all, everyone should realise that these stones could be captured at any time."
Uke: "Ah, I see, the "dead" Black stone may not be removed from the board because it is not surrounded exclusively by "alive" White stones."
Tori: "Yes, you are right. And White's "dead" stones are not surrounded exclusively be "alive" Black stones, so they will not become prisoners as well."
Uke: "And what alternative do you have in mind?"
Tori: "Oh, I think we might as well declare the stones in the corner "alive"."
Uke: "Yes, that would work too. After all, the stones stay on the board."
Tori: "This alternative implies that there will be a "dame" at the 1-1-point. But which will not matter here, as the marked stones do not enclose anything anyway."
Uke: "Hmm, I think White will not really want to take action here. It would be best for her if both sides did nothing."
Tori: "Yes, so it will be Black who is required to act in a timely manner."
----------
Exemplary case #2.
White moves first in the corner.
White captured eight Black stones and got eight points of territory, so her total value is sixteen points.
Black captured two White stones and got two points of territory, so his total value is four points.
White's net advantage is twelve points.
Black moves first at the right.
White captured one Black stone and got one point of territory, so her total value is two points.
Black did not capture any White stones, but she got one point of territory, so his total value is one point.
White's net advantage is one point.
Uke: "Hmm, another confusing position. But can we really be sure that every player is aware of the above move sequences?"
Tori: "Let's simply declare the stones in the corner "alive" at the end of the game. The status of White's two stones at the right does not matter."
Uke: "This leads to a "dame" at the 2-2-point."
Tori: "Yes, you are right. Neither Black's nor White's groups surround anything of value. Either player will want to take action in due time."
Uke: "It seems to me that both have to hurry up a lot."
Tori: "Indeed. Sometimes during the game you can't see the wood for the trees."
----------
Examplary case #3.
White captured eight Black stones and got nine points of territory, so her total value is seventeen points.
Black captured six White stones and got two points of territory, so his total value is eight points.
White's net advantage is nine points.
Uke: "Alternatively, White could also have captured the Black stones on the left, couldn't she? How can we make her choose?"
Tori: "White's stones in the corner are "alive" anyway, so ..."
Uke: "Yes?"
Tori: "... let's declare Black's stones "alive" at the end of the game. The status of White's stones at the top and at the left do not matter."
Uke: "I see, this results in two "dame" at the edges of the board."
Tori: "Indeed, this is sufficient. And nobody surrounds anything of value."
----------
Exemplary case #4.
Black moves first in the corner.
Black captured eight White stones and got four points of territory, so his total value is twelve points.
White captured six Black stones and got three points of territory, so her total value is nine points.
Black's net advantage is three points.
White moves first in the corner.
Black captured eight White stones and got three points of territory, so his total value is eleven points.
White captured five Black stones and got three points of territory, so her total value is eight points.
Black's net advantage is three points.
Tori: "Uke, you will see that it does not matter here who plays first in the corner."
Uke: "Yes, but regardless of that, it is the same as in examplary case #1. We should give Black a hidden clue by declaring the stones in the corner "dead"."
Uke: "Or "alive"? Hmm, I have no idea what could be better."
Tori: "Don't worry about that. The only important thing is that Black resolves this position before the end of the game."
----------
It should have become evident that NONE of the positions discussed above will see the end of the game!
----------
To be continued ... (simply scroll down)
Last edited by Cassandra on Tue Oct 26, 2021 9:39 am, edited 3 times in total.
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: GERMAN interpretation of J89's intended contents
Continued.RobertJasiek wrote:How / with which principles have you derived / created the German version?
Approach #7: Verification of life & death and clarity of wording. Contd.
We know from the past what the Japanese understanding is of the connection between "life and death" and "territory".
Cassandra wrote:
"seki", capture, no ko (contd.)Projects under review by controlling: grounded because UNPROFITABLE!
Exemplary case #1.
Black moves first in the corner.
White captured nine Black stones and got six points of territory, so her total value is fiveteen points.
Black captured nine White stones and got five points of territory, so his total value is fourteen points.
White's net advantage is one point.
White moves first in the corner.
White captured eight Black stones and got six points of territory, so her total value is fourteen points.
Black captured nine White stones and got four points of territory, so his total value is thirteen points.
White's net advantage is one point.
Variation for Black.
White captured nine Black stones and got six points of territory, so her total local value is fiveteen points.
Black captured eight White stones and got one point of territory, so his total local value is nine points.
White's net local gain is five points.
Black will have gained globally by his ko-threat.
Variation for White.
Black captured eleven White stones and got four point of territory, so his total local value is fiveteen points.
White captured three Black stones, but got no territory, so her total local value is three points.
Black's net local gain is thirteen points.
White will have gained globally by her ko-threat.
Tori: "Uke, you will see that it does not matter here who plays first in the corner."
Uke: "Yes, but regardless of that, we have examined this kind of position before. We should give White a hidden clue by declaring the stones in the corner either "dead" or "alive"."
Tori: "Are you sure that White really needs this hint? It seems a bit risky to capture the single Black stone in the corner, doesn't it?."
Uke: "Oh, hell, "ko" is my blind spot after all!"
Tori: "It is far more likely that Black will win the ko-fight than White. White will shy away from giving Black the opportunity, because she has more to lose."
Uke: "Ah, then this position will see the end of the game! Interesting."
----------
Exemplary case #2.
White moves first in the corner.
Black captured ten White stones and got five points of territory, so his total value is fiveteen points.
White captured four Black stones and got four points of territory, so her total value is eight points.
Black's net advantage is seven points.
Black moves first at the right.
White captured one Black stone and got one point of territory, so her total value is two points.
Black did not capture any White stones, but got one point of territory, so his total value is one point.
White's net advantage is one point.
Uke: "Tori, I know! We had this kind of position before. Let's declare all stones at the top "alive"."
Tori: "Do you really think that is necessary? Neither of the players will want to become active here. This position will see the end of the game."
Uke: "Yes, but there will be the "dame" at the 2-2-point again."
Tori: "What it is needed for?"
Uke: "To destroy any potential value. Otherwise White would profit from doing nothing with one point, and thus achieve the same result as if Black had played erroneously in the position."
Tori: "You have understood the deeper meaning, you are making progress!"
----------
Perhaps you have noticed that "if capturing ..." provides merely a pseudo-scientific explanation for the implicit compulsion to resolve ambiguous "seki" positions.
"dead"Everything remaining.
No territory.
----------
To be continued ... (simply scroll down)
Last edited by Cassandra on Sat Oct 23, 2021 10:16 pm, edited 5 times in total.
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: GERMAN interpretation of J89's intended contents
Continued.RobertJasiek wrote:How / with which principles have you derived / created the German version?
Approach #8: The connection between "life & death" and "territory".
We know from the past what is aimed for in the game of Go.
Cassandra wrote:VALUE can only be created within called-group-by-the-common-people formations that either already contain at least two permanently-for-the-opponent-forbidden-board-point construction elements, none of which is the sole connection between the called-group-by-the-common-people formation's stones along the lines of the board, which are compatible to called-true-eye-by-the-common-people construction elements, or that can be transferred into such a called-group-by-the-common-people formation with at least two permanently-for-the-opponent-forbidden-board-point construction elements, none of which is the sole connection between called-group-by-the-common-people formation's stones along the lines of the board, which are compatible to called-true-eye-by-the-common-people construction elements, even if the opponent plays first, and which enclosed board intersections are either unoccupied or occupied only by opposing can-be-made-to-disappear-permanently-even-if-the-opponent-plays-first called-group-by-the-common-people construction elements, which are compatible to called-dead-by-the-common-people called-group-by-the-common-people construction elements, which are compatible to called-two-eye-formations-by-some-rule-specialists construction elements, as well as to called-independently-alive-by-the-common-people called-group-by-the-common-people construction elements.
The trivial solution! Too simple to be true.
Any board point within a group that either already is,
or even if the opponent has played first can be transformed into,
a "two forbidden points formation",
is "territory".
++++++++++++++++++++++++++++++++++++++++++++++++++++
Even the L&D-Example-4-type of positions does not require special treatment, because ...
No stone can belong to more than ONLY ONE group!
----------
To be continued ... (simply scroll down)
Last edited by Cassandra on Sat Oct 23, 2021 10:16 pm, edited 2 times in total.
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: GERMAN interpretation of J89's intended contents
Continued.RobertJasiek wrote:How / with which principles have you derived / created the German version?
Approach #8: The connection between "life & death" and "territory". Contd.
We know from the past what is aimed for in the game of Go.
Cassandra wrote:VALUE can only be created within called-group-by-the-common-people formations that either already contain at least two permanently-for-the-opponent-forbidden-board-point construction elements, none of which is the sole connection between the called-group-by-the-common-people formation's stones along the lines of the board, which are compatible to called-true-eye-by-the-common-people construction elements, or that can be transferred into such a called-group-by-the-common-people formation with at least two permanently-for-the-opponent-forbidden-board-point construction elements, none of which is the sole connection between called-group-by-the-common-people formation's stones along the lines of the board, which are compatible to called-true-eye-by-the-common-people construction elements, even if the opponent plays first, and which enclosed board intersections are either unoccupied or occupied only by opposing can-be-made-to-disappear-permanently-even-if-the-opponent-plays-first called-group-by-the-common-people construction elements, which are compatible to called-dead-by-the-common-people called-group-by-the-common-people construction elements, which are compatible to called-two-eye-formations-by-some-rule-specialists construction elements, as well as to called-independently-alive-by-the-common-people called-group-by-the-common-people construction elements.
Yearning for complexity: "if capturing them ...".
"If capturing them ..." (or similar) overdetermines a sub-component of "even if the opponent has played first" that was previously sufficiently regulated. Unnecessarily, out of ONE use case
TWO different use cases are made,
which of course require a whole bundle of additional regulations.
In addition, the verification of "life and death" says goodbye to[/color] "groups" of stones,
and instead forces verification for each individual "chain / string" of stones.
The latter implies the need to consider dependencies and interactions (between several chains / strings)[/color] that had not played the slightest role before.
J89 made matters worse by assigning ONE attribute "alive" to TWO different types of chains / strings, "indepentdently alive" chains / strings
and chains / strings that are part of a "seki".
This also requires ADDITIONAL verification.[/color]
----------
To be continued ... (simply scroll down)
Last edited by Cassandra on Sat Oct 23, 2021 10:15 pm, edited 1 time in total.
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: GERMAN interpretation of J89's intended contents
Continued.RobertJasiek wrote:How / with which principles have you derived / created the German version?
Approach #8: The connection between "life & death" and "territory". Contd.
We know from the past what is aimed for in the game of Go.
Cassandra wrote:VALUE can only be created within called-group-by-the-common-people formations that either already contain at least two permanently-for-the-opponent-forbidden-board-point construction elements, none of which is the sole connection between the called-group-by-the-common-people formation's stones along the lines of the board, which are compatible to called-true-eye-by-the-common-people construction elements, or that can be transferred into such a called-group-by-the-common-people formation with at least two permanently-for-the-opponent-forbidden-board-point construction elements, none of which is the sole connection between called-group-by-the-common-people formation's stones along the lines of the board, which are compatible to called-true-eye-by-the-common-people construction elements, even if the opponent plays first, and which enclosed board intersections are either unoccupied or occupied only by opposing can-be-made-to-disappear-permanently-even-if-the-opponent-plays-first called-group-by-the-common-people construction elements, which are compatible to called-dead-by-the-common-people called-group-by-the-common-people construction elements, which are compatible to called-two-eye-formations-by-some-rule-specialists construction elements, as well as to called-independently-alive-by-the-common-people called-group-by-the-common-people construction elements.
Yearning for complexity: "if capturing them ...". (Contd.)
Board points that are occupied by an "alive" chain / string of stones under verification, ...
... which is captured, occupy such chain / string again the end of the verification.
This is because these board points are "owned" by such strings.[/color]
However, this REBIRTH of a chain / string is only the 99.9999% case.
An equivalence of "chain / string of length 1" with "forbidden point" must be ADDITIONALLY introduced to solve your famous J89 counterexample, ...
[/color]... which was NOT worth mentioning in the lost world of GROUPS.
[/color]
----------
A chain / string under verification, ...
... which is captured, but only PARTIALLY reborn, is a "seki" chain /string.
A chain / string under verification, ...
... which is captured, but NOT reborn, is a "dead" chain /string.
The formation in the upper left corner is commonly known as "seki".
The Black chain / string is "dead", but which does NOT matter here, as it is NOT EXCLUSIVELY fenced by "alive" chains / strings, and so will remain on the board after the verification of L&D has ended.
[/color]The Japanese understanding, on the other hand, is based on STONES that are played after the chain / string to be verified has been captured and that are still on the board at the end of the verification.
During verifikation, three White stones arised permanently on the board, so White's group under verification is "alive".
During verifikation, two (maybe three) Black stones arised permanently on the board, so Black's group under verification is "alive".[/color]
However, "being alive" is NOT a suitable property to differentiate between "independently alive" chains / strings and "seki" chains / strings.
An [/color]ADDITIONAL property "has 'dame'" must be introduced, but which is irrelevant here, because none of chains / strings in the upper left corner fences something.
----------
Everything would already be done if it were NOT for J89's L&D Example 4.
But this L&D Example exists, together with its Japanese understanding, so the matter will become even worse than it already was.
A catastrophe will just barely be averted, as we will see in subsequent posts.
Verification will be done for Black's chain / string at the top.
Black's string is captured, but NOT reborn, so it is "dead".
However, this result does NOT comply with [/color]Japanese understanding that sees this chain / string "alive".
However, we will NOT go that far here.
After Black's chain / string is taken by White with
, Black's move at 
completes a Black "two forbidden points formation", which did not yet exist in this area.In the presence of such an event, we will set the chain / string to be verified as a "seki" chain / string.
[/color]This insight ("birth of new independently alive group where none was before") may be helpful to end discussions in J89's world ...
... why moves like
are NOT a means to make chains / strings "alive" somewhere in the shaded area of the board.[/color]----------
To be continued ... (simply scroll down)
Last edited by Cassandra on Sun Oct 24, 2021 4:37 am, edited 1 time in total.
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: GERMAN interpretation of J89's intended contents
Continued.RobertJasiek wrote:How / with which principles have you derived / created the German version?
Approach #9: The ultimate challenge: NO UGLY chôsei.
You may be surprised that the heading is not "NO ko-fights & NO UGLY chôsei"
Well, ko-fights are not problematic per se, as you will see soon! We will discuss J89's ineffectual attempt at a solution at length later.
The intended result of J89's L&D Example 4 makes this topic independent of how L&D is verified and how "territory" is defined.
One ko-shape.
Verification of L&D for Black's chain / string of length 1 at the top.
The single ko-shape in question is marked with
here.Black's chain / string is captured, but NO other Black stone has been established permanently on the board, so Black's chain / string is "dead".
The sequence above helps verifying L&D of Black's chain / string in the upper right corner, which is "dead" as well.
Fighting the ko is merely a waste of time.
Sooner or later, Black runs out of valid ko-threats.
Two ko-shapes coupled together: "double-ko seki".
The basic case of a "double-ko seki".
Dependent on the regulations used, the larger chains / strings are either "alive" or "seki", the chains / strings of length 1 are either "dead" or "seki."
So far so good as the verification of L&D for the upper left corner is concerned. However, there are ...
Implications for the rest of the board!
Verification of L&D for Black's marked chain / string somewhere else on the board, which is "hidden" behind a (third in total) ko-shape.
If White captures into the ko-shape at the bottom, Black captures into "his" ko-shape in the "double-ko seki". White also captures into "her" ko-shape in the "double-ko seki".
After having played a ko-threat successfully, Black re-captures into the ko-shape at the bottom. White passes. She cannot have any other options at her disposal, otherwise she would not (yet) have captured into the ko-shape at the bottom. Black also passes, as he has nothing to do.
If White captures into the ko-shape at the bottom for a second time, the mutual capturing into the ko-shapes of the "double-ko seki" is also repeated (this time the other way round).
After having played a ko-threat successfully, Black re-captures into the ko-shape at the bottom for the second time. We have reached the verification's starting position.
The cycle shown will be repeated forever, resulting in an "ugly chôsei" for Black's chain / string under verification, which will be never captured.
White has no possibility whatsoever to resist.
If she captures Black's chain / string under verification at the bottom, Black will capture her large chain / string in the "double-ko seki".
Black has established a "two forbidden points formation" in a region, where none was before.
Dependent on the regulations used, Black's chain / string at the bottom is either "alive" or "seki".
Additional regulations must be introduced to effectively prevent the "double-ko cycle" described above.
As we will see, this is easier said than done.
----------
To be continued ... (simply scroll down)
Last edited by Cassandra on Mon Oct 25, 2021 8:09 am, edited 1 time in total.
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: GERMAN interpretation of J89's intended contents
Continued.RobertJasiek wrote:How / with which principles have you derived / created the German version?
Approach #9: The ultimate challenge: NO UGLY chôsei. (Contd.)
We know from the past what is mandatory to prevent "UGLY chôsei":
There are several options for doing so.Cassandra wrote:Additional regulations must be introduced to effectively prevent the "double-ko cycle".
Prohibit double-ko capture as a KO-THREAT The simplest and most straightforward option that directly addresses, and eleminates, the cause of the main problem.
But it is obvious that Japanese culture does not consider it opportune to explicitly and publicly prohibit behaviour that is not considered opportune.
is prohibited for Black after White captured into the ko-shape at the bottom.Black's chain / string at the bottom is captured, so it is "dead".
----------
If you have a double-ko on the board and want to exclude the impulse to capture into one of its ko-shapes from the outset, you can use "HONTE's magic double-ko spell" to simplify the position:
Move the two individual stones each in the direction of their main groups.
The result is a "simple" seki without ko.
Sometimes there may be uncertainty about the extent to which two individual ko-shapes are linked together to form a double-ko.
In such a case, you might want to apply the "extended version" of HONTE's spell.
Add a stone to ONE of the SHARED liberties.
It does not matter which liberty and which stone colour you chose.
If the verification's result is the SAME with or without that added stone, the two ko-shapes form a double-ko.
Otherwise, they do not. And you have to undo HONTE's spell.
----------
However, L&D Example 23's disturbing chôsei remains unsettled.
repeats .White's chains / strings at the top will be never captured, therefore these are "alive" (or "seki", according to the regulations applied).
Per pass, Black captures three White stones, White captures two Black stones. The number of captured stones per pass is unequal, thus this type of cycle can be named "unparity cycle".
Prohibit double-ko capture as a ko-threat & the repeated START of an UNPARITY CYCLE Works successfully!
The point of
is prohibited for White. A move at this point would repeat .All White chains / strings are captured, so these are "dead".
Prohibit the repeated START of a cycle Not applicable.
returns to the verification's starting position.A White move at
is prohibited. A move at this point would repeat .Black's chain / string remains NOT captured, so it is "alive".
Prohibit the repeated END of a cycle Works successfully in the case of double-ko!
A Black move at
is prohibited. A move at this point would repeat 
.Black's chain / string is captured, so it is "dead".
----------
However, L&D Example 4 does NOT achieve its intended result.
Verification of L&D for White's chain / string in the corner.
A White move at
is prohibited. A move at this point would repeat .White's chain / string is captured, so it is "dead".
Black's chains / strings in this position are also verified "dead", so it is a "seki".
J89, however, sees this White chain / string "alive", so "killing" Black's chains / strings.
The above result cannot be avoided in principle.
It agrees with the comparable result in the lost world of groups (White is unable to create a "two forbidden points formation" at the top), should thus be "correct".
However, the impact of this contradiction with J89 is extremely small.
It is strictly limited to games, where
(or White A), stopping Black from enforcing "no result", would lose the game by half a point.However, there is a way to force J89's intended result:
Limit the prohibition of repeating the end of a cycle to "ko-cycles", i.e. those that include a ko-shape or the creation of one.
BTW, cycles WITHOUT ko do not create any problems.
----------
What about L&D Example 23?
and 
started the verification sequence.The point of
is prohibited for White. A move at this point would repeat 
.All White chains / strings are captured, so these are "dead".
----------
What about L&D Example 18, J89's litmus test?
The point of
is prohibited for Black. A move at this point would repeat .Black's chain / string is captured, so it is "dead".
Prohibit the repeated end of a KO-cycle Works successfully!
L&D example 6 is no longer an issue.
"PASS for LIFTING end-of-ko-cycle-BAN" Works successfully!
A Black move at
is prohibited. A move at this point would repeat .Black's chain / string is captured, so it is "dead".
The result is the same as above.
However, as an IMPLICIT prohibition, this variant may be more fault-tolerant than the strict EXPLICIT one above in case potentially problematic formations have been overlooked.
"Pass for lifting SINGLE ko-ban" Not applicable.
This is J89's regulation.
repeats .Black's chain / string at the bottom will be never captured, therefore this is "alive".
White does not have any chance to resist.
If she captures Black's chain / string at the bottom, Black will create a "two forbidden points formation" THEREAFTER in the upper left, where none was before.
White must not recapture into the double-ko, as she did not play a "pass-4-ko" for "her" ko-shape there before.
Again, Black's chain / string at the bottom is verified "alive".
As if that wasn't enough, this forced double-ko cycle brings to life any chain on the board that has MORE THAN ONE liberty!
repeats .Black's chain / string at the bottom will be never captured, therefore this is "alive".
J89 chose the worst of all conceivable possibilities by prohibiting the immediate recapture into a SINGLE ko-shape, but which does not cause any negative effects at all, for the solution of a problem that arises in the INTERACTION OF A DOUBLE-KO with a single ko.
"Pass for lifting single ko-ban" & "pass for lifting end-of-ko-cycle-ban" Works successfully!
This is no wonder, as "pass for lifting single ko-ban" overdetermines what is already EXHAUSTIVELY regulated by "pass for lifting end-of-ko-cycle-ban".
White will start in the double-ko, in order to eliminate the source of all evil.
A Black move at
is prohibited. A move at this point would repeat .Black's "pass-4-cycle" is useless, as his large double-ko chain / string is taken off the board (remember the saying "There is NO double-ko in the varification of L&D").
Thereafter, Black's chain / string at the bottom is captured, so it is "dead".
----------
The verification of L&D for a double-ko will see all of its chains / strings "dead".
This formation remains a "double-ko seki".
"Pass for lifting ALL ko-bans" Works successfully in the case of double-ko!
This is the specific regulation in your J2003.
has to "pass-4-ALL-ko", if he wants to recapture into any ko-shape (either for saving his chain / string at the bottom or for idling around in the double-ko).However, White is faster, capturing his chain / string at the bottom, which therefore is "dead".
----------
However, L&D Example 23 will remain unsettled!!!
"Pass for lifting all ko-bans" & prohibit the repeated start of an unparity cycle Works successfully!
L&D Example 23 is NO longer an issue.
"Pass for lifting single ko-ban / RECAPTURE @ NO ko-ban LEFT" Works successfully in the case of double-ko!
White starts by capturing at the bottom.
After Black idled around in the double-ko, there are three ko-bans
on the board.White's "pass-4-ko" erases the ko-ban at the left edge.
White takes Black's chain / string at the bottom off the board. The ko-ban at the bottom vanished together with its ko-shape.
White is hindered to capture into the double-ko, as the ko-ban at the top is still active on the board (remember the saying "There is NO ko-fight in the verification of L&D.")
----------
Verification of L&D for White's chain / string of length 1.
After the idling around in the double-ko, White's chain / string of length 1 at the left edge will remain on the board forever!
The verification of L&D for a double-ko's chains / strings of length 1 will have to determine these chains / strings being "seki-stones" (or "dead", dependent on the regulations used).
Otherwise, e.g. with J89's regulation, ...
... there would be territory in "seki".
----------
However, L&D Example 23 will remain unsettled!!!
"Pass for lifting single ko-ban / recapture @ no ko-ban left" & prohibit the repeated start of an unparity cycle Works successfully!
L&D Example 23 is NO longer an issue.
----------
To be continued ... (simply scroll down)
Last edited by Cassandra on Mon Oct 25, 2021 11:33 pm, edited 4 times in total.
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: GERMAN interpretation of J89's intended contents
Continued.RobertJasiek wrote:How / with which principles have you derived / created the German version?
Approach #10: Do not exaggerate!
I have tried to produce an inherently logical and error-free version of the rules that is consistent with the intended outcomes (as well as a presumed "Japanese understanding") of J89 without being too subversive.
My personal favourite would be the verification of L&D for GROUPS, in combination with "pass-4-END-of-KO-cycle".
The END.
----------
Any questions, Robert?
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: GERMAN interpretation of J89's intended contents
For those, who are interested in studying the entire stuff.
- Attachments
-
- GERMAN interpretation of J89_s intended contents_Part#1.pdf
- (1.86 MiB) Downloaded 694 times
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: GERMAN interpretation of J89's intended contents
Part #2.Cassandra wrote:For those, who are interested in studying the entire stuff.
- Attachments
-
- GERMAN interpretation of J89_s intended contents_Part#2.pdf
- (1.14 MiB) Downloaded 604 times
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
-
drmwc
- Lives in gote
- Posts: 452
- Joined: Sat Dec 01, 2012 2:18 pm
- Rank: 4 Dan European
- GD Posts: 0
- Has thanked: 74 times
- Been thanked: 100 times
Re: GERMAN translation of J89's intended contents
Off-topic for Go, but I recently bought a translation of Newton's Principia Mathematica. This bought home the different approaches that can exist within a translation (or interpretation).John Fairbairn wrote:Cassandra: You have clarified your intentions (though not J1989's still misunderstood intentions), and it seems now there is a simple fix. Just change your title: to "German interpretation of J1989's contents".
There is a distinction made by careful speakers of English, and most certainly by professional linguists, between translators and interpreters.
The people who sit in offices at the EU typing up what they think a treaty text says in one language into another language, for publication, are translators (and above them they have, or should have, revisers and editors). The people who sit in booths with earphones on and pass on what they hear in a conference hall in one language into a microphone in another language are simultaneous interpreters. People who sit around a table together with two groups of businessmen or the like, and tell each group what the other group has said, are called consecutive interpreters.
The distinction boils down usually to time. A translator has time to think, and there is also time for revisers and editors to change whatever he has put down on paper. Interpreters have no time to think. As a result they are more prone to mistakes. These have happened on the world stage, when the interpreters who stand behind presidents whispering their interpretations have mis-whispered. With sometimes dramatic effect.
The distinction is tellingly encapsulated in the story of a brilliant interpreter at the UN who came up with something like the following for her audience of non-Polish speakers: "The Polish delegate has just made an untranslatable joke; he would appreciate it very much if you all laughed." They all laughed. Had this lady been asked to translate the joke on paper, I think she was brilliant enough to have found a word play that could pass as a translation, but strictly that, too, would have been an interpretation.
Something is always lost in translation, and the borderline between translation and interpretation, which depends on how much is lost, is bound to be fuzzy. But your additional comments have made it plain to me that you are way over on the interpretation side. Which is fine, and potentially useful.
But using the correct title would even better alert readers to what follows and what to look for.
The translation was by Charles Leedham-Green, a UK Go player who is a professional mathematician. In the introduction, he compares his approach to a previous translation by Cohen and Whitman (1999). The earlier approach was by professional linguists, and remained as true to the original Latin as possible. However, the upshot is that their translation is not immediately comprehensible to a reader. A specific example Charles highlights is Corollary 1 to Lemma 1 in Book 2, Section 2. Cohen and Whitman's translation is:
Hence in continually proportional quantities, if one term is given, the moments of the remaining terms will be as those terms multiplied by the number of intervals between them and the given term. Let A, B, C, D, E and F be continually proportional, then if the term C is given, the moments of the remaining terms will be to one another as -2A, -B, D, 2E and 3F
This is completely obscure to me. Charles' translation is:
Thus, in a geometric series of functions, if one term is constant, the moments of the any other term will be proportional to the product of that term with the number of steps in the series from that term to the constant term. Thus let A, B, C, D, E and F form a geometric series of functions, where the term C is constant. Then moments of the remaining terms will in the ratio -2A : -B : D : 2E : 3F
This makes a lot more sense to me, and is in fact trivial to prove.