There was a recent dispute over in the Nongshim Cup where Kang Dongyun thought that he had captured a group of black stones, when in fact, the game ended with no result and they had to replay the game.

Here is the board position:

Click Here To Show Diagram Code

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

At a glance, it'd seem that the black stones are dead. They don't have two eyes, so they should eventually be captured, right?

But because black has an infinite supply of ko threats around the marked intersections, white cannot kill black.

I suppose that even pros have challenges now and then in determining the status of groups.

_________________
be immersed

There was actually a thread about this already. Though it's only 3 hours old it no longer shows up for me under "view new posts", I assume it is similar for others, I have seen it before that some posts kind of disappear. Anyway.


About the Nongshim Cup. Was it an actual dispute or something the player just hadn't realized?

Thinking about what is a dispute and the right of players to ask a referee to clarify rules makes me think of how then chess 960 world champion Wesley So made an illegal move in the the latest chess 960 world championship and when challenged about it consulted the referees about how to castle (it is by the way exactly the same in chess 960 as it is in regular chess), in that case he seems to have been so flabbergasted after hallucinating that the game variation was good that he forgot how to castle Maybe it is not foolish for a top chess player to have such an open and flexible mind that he can be confused about how to castle, in the heat of the moment, maybe this almost child like cognitive flexibility is necessary to keep improving and learning a game like chess to a level greater than almost anyone else. Maybe it's the same in Go.

Thanks - I didn't realize there was already a thread. I also don't know what the official definition of "dispute" is, I suppose. In this case, there was at least a misunderstanding.

_________________
be immersed

It is my understanding that the way that the rules for Chess 960 are (or were; don’t know if they were changed after that) implemented on chess.com, they allowed castling in the specific situation he was in. (The king was in check, but would not be moved when castling.) So it wasn’t so much a hallucination as allowing an online implementation to inform his understanding of the rules.

This particular board state was super interesting, especially since it has different outcomes in different rule sets. It is significant that the tournament is played with Korean rules.

You might want to hear his version of it in the interview at the end of this clip. Seeing the smiles on the faces of the arbiters is also priceless. Maybe there is some video of the incident in the Nongshim cup?

About the Go position, I don't think it is so much of a question of which rules are used. The rules that are actually used in professional tournaments in Asia, with the exception of Ing rules, allow for such kos to require a rematch. Of course other rules exist but if anything it looks like an example of a shared or a common understanding.

Yes, there WAS a dispute. Both players were unsure, and the situation came to the referee. Even the referee changed the call after realizing that the shape is listed in the rule book. Most pros at least knew that during a hypothetical play, a ko threat can be made only in the "relevant area", but they were not sure what is the relevant area in this case.

_________________
Jaeup Kim
Professor in Physics, Ulsan National Institute of Science and Technology, Korea
Author of the Book "Understanding the Rules of Baduk", available at https://home.unist.ac.kr/professor/jukim/bbs/board.php?bo_table=notice&wr_id=5

Just from looking at this position, I found reports of a dispute hard to accept. I am the least competent person to argue about rules, and don't even know the Japanese rules properly. But simple examination of this position shows that any argument over a "relevant" area must be irrelevant in this case. I don't really understand what is meant by "hypothetical play" but I'd be surprised if it doesn't just mean trying out a few moves after both players have agreed that proper play has ended, the idea being to clarify any points on which either player is unsure. On that basis, I can easily imagine that Kang (in byoyomi, tired after a tough 4-hour game and with his eyes firmly on a 10-million won winning-streak prize) may have felt, mistakenly, that he had killed the big Black group. When it was pointed out to him he hadn't, he may have reacted in disbelief. Given potential language problems and the fact that the game was on the internet, I'd find it easy to imagine that the referee, an Asian who would also be alive to problems of 'face', felt it safest to ponderously cite the rule book rather then tell a 9-dan pro he had hallucinated with a mis-read.

That is just my speculation, of course, but with that in mind, I decided to look on the internet for news about any "dispute". I found zero. I spent about 20 minutes and found it hard to see any official report that even mentioned the quadruple ko. There was clear mention of a re-match, but most talk was of Kang extending his winning streak, and there was quite a bit of rubbing the Japanese noses in their defeats. I saw no mention of rulesets except of the rule that said the re-match had to take place at once with the remaining times for each player. But no mention of any dispute. I had looked on the Hanguk Kiwon site and at the sites of several Korean newspapers - I can't remember which, but one was Sports Today and they would all have leapt on a "dispute" for journalistic reasons.

My search may have been lacking and my knowledge of rulesets is certainly lacking, but all of that seemed to confirm my initial impression of mountains and mole-hills (and mavens, just to keep the alliteration going ).

I found the stream https://youtu.be/jFcWz7ITud0?t=15032. The time in the link (4:10:32) shows the result of counting on the Go server, it seems to say everything is alive, then it goes into a screen with the positions blacked out. Maybe not much to be seen from this, it is a bit different when it is not in person and the Go server counts incorrectly.

The players didn't agree who won the game, and called the referees. In my book that's a dispute.

For the Korean rule, see this page, section 10, second to last example. White is not dead. Black can continue repeating (thus claiming no result) or accept a large seki.
https://www.kbaduk.or.kr/baduk/rule/

For comparison, see this page on Japanese rules example 16. Almost identical situation, but declares all white stones are dead.
https://www.cs.cmu.edu/~wjh/go/rules/Japanese.html

Given that Korean and Japanese rules give different results for this same situation, I'd say there is plenty of room for relevant arguments.

A few seconds after this timestamp the Korean announcer says "This is clearly a win for Kang Dongyun". The BadukTV announcers thought the stones are all dead. I didn't watch all of it and my Korean skills are very low. I did watch most of https://www.youtube.com/@leehyunwookTV Lee Hyunwook's videso (these are edits of his live broadcast so capture his live reactions). He thought it was a no-result, but his chat was telling him the BadukTV broadcast was saying Kang wins. He was surprised and kept saying according to him it should be no-result, asking his chat to confirm BadukTV was saying Kang wins?

I'm not really surprised pros had different opinions, Go creates some crazy situations sometimes.

ETA: I also watched the beginning of the BadukTV broadcast of the next day's game. They started off chatting about the previous day, and re-explaining the no-result. It was too complicated for me to understand the Korean exactly, but I heard them do a typical "we apologize and promise to do better in the future" type thing. I assume this was because in the live broadcast they had mistakenly assumed Kang won, and were surprised at his opponent calling for a ruling. And I didn't see it all unfold but they must have had to do some back peddling when the it was ruled to be a no-result!

I can see on the stream video that I linked that Kang Dongyun appears to decline or cancel a dialog which has a countdown (that is in the program they are using) and then clicks on a blue button and accepts something in a dialog. The computer screen has something in hangul written over it but after the dialog action there is something in Chinese characters. I am only interested in finding out what happened.

I don't think I can agree that if the players don't agree before consulting a referee it is therefore a dispute. Maybe this is a dispute now



When I refer to this translation it appears to say under article 10 that if the position is repeated it is a "draw" and this is similar to article 12 of Japanese 1989 rules. The position that you refer to is not same as in this game, the position in the game is one of those triple-ko like positions. Also 1989 is not really the rules of the game, in Japan or anywhere, for one thing the game position has the fun pass-ko cycle

https://www.cyberoro.com/news/news_view ... =1&cmt_n=0

이 바둑을 보던 한·중의 많은 기사가 백(강동윤)이 흑 대마를 잡은 것으로 착각했다. 사이버오로에서 해설하던 안성준 9단도 도중 “한국룰로는 백이 흑을 잡아 승리”라고 했지만 이후 규정집에 해당 형태가 있는 것을 확인한 뒤 바둑팬들에게 사과하며 “백이 흑을 잡은 것 같지만 자체 해결이 불가능한 모양이어서 무승부다.”라고 설명했다.

My translation:
Many professionals who watched mistakenly thought white (Kang Dongyoon) captured Black's group. During the game, An Seungjoon 9p also said "According to Korean rules, white captured black's group and won", but later after confirming the relevant rules apologized to Baduk fans explaining "it seems like white captured black, but it's not possible [to actually capture them] in this shape so it's no-result".

So a bunch of people who make a living playing and explaining Baduk did not understand who won a completed game. Only after looking at the rulebook did they get the right result. I don't want to argue the definition of the word dispute. But I want to dispute the impression that the result is obvious and only a simple oversight by Kang Dongyoon. They had plenty of time to consider what was going on when White stopped playing in the gigantic dragon area and instead went back to endgame moves. Most of them were still just saying it's obvious that black is dead.




Yes the Korean rules say it's a draw.

For Japanese 1989 I'm using this source:
https://www.cs.cmu.edu/~wjh/go/rules/Japanese.html
You mention the Japanese 1989 rules "Article 12. No result". But "Life-and-Death Example 16" says "The ten white stones in the left corner are dead. The eleven white stones to the right also die through collapse of the seki." So under Japanese rules Kang would have ended the game and claimed both dragons are dead and he wins. It would not be an Article 12 No result.

I realize the position is not identical but it's the closest one. There are pictures of the Korean judge pointing to that position. So clearly they think even though the rule example is a sort of triple-ko not a quadruple-ko, they used that position to judge it.
https://www.cyberoro.com/news/news_view ... =1&cmt_n=0
"규정집의 참고도 19가 실전과 관계 있는 모양이다." "Diagram 19 is related to the game shape"

Unless someone has a source on updated Japanese rules that say otherwise?

ETA: I also don't want to imply that I think something is "wrong" with the rules and the must be fixed. Or argue the "logically" they should use rule-lawyer sets with triple-ko that a computer can easily judge etc etc. I think these rare occasions add some spice, and it's kinda fun to hash it out online. Go is a great and elegant game, these rare situations add some flavor.

ETA2: I also don't want to imply the Koreans who got this position wrong should be embarrassed and/or how dare they not know the rules. I just wanted to highlight that the position was confusing enough that they didn't get it right without referring to the rules. Relying on your instinct is not good enough because there are logical arguments to both rulings (dead or no-result) using the logical framework of territory scoring and determining life and death "locally". The definition of local is the tricky part here.

Oh I just realized jaeup posted in here.

Jaeup is what I wrote, especially that Japanese rules give the opposite result, right?

Yes, Black doesn't have a chance to make a life under the Japanese rule. The Chinese rule involves a "human judgement" in this case, and I was surprised to hear that most Chinese players think that Black can't make a life under the Chinese rule, especially considering that the "standard combined moonshine life" is a draw.

I think some people really did not understand what happened on that day. The referee first called that "Kang won the game" and then the call was changed. If that does not constitute a "dispute", I don't know what can do it.

_________________
Jaeup Kim
Professor in Physics, Ulsan National Institute of Science and Technology, Korea
Author of the Book "Understanding the Rules of Baduk", available at https://home.unist.ac.kr/professor/jukim/bbs/board.php?bo_table=notice&wr_id=5

Marvellous to see someone who clearly takes an interest in the arcana of rules but still applies common-sense and can see the sense in taking the "journalistic" view.



Just to make sure this thread does not descend into a real dispute, my own view of the word can be expressed with a topical analogy: the World Cup. Handball: the on-field referee points to the penalty spot. The VAR referee suggest a look at the replay. The referee looks and changes his mind. Both teams, both sets of supporters, both countries are aggrieved. They all agree that the handball rule is a mess, but they agree to disagree and get on with the game. They do not go to war, they do not go to the CAA, and the incident doesn't even make it to the news bulletins outside the two countries concerned. To me that's just a mild disagreement. This particular go case doesn't even seem to have got that far in my book, at least as far as the media is concerned. In other words, when I saw "dispute" in the title, I was expecting something much more juicy.



I don't watch videos. They are not much fun when you are deaf. So I don't really grasp what is going on here, but is there not some element of possible cheating (or of undue influence in order to stoke up a dispute) going on via the chats?

If this is a draw in Korean Rules but not Japanese or Chinese rules, then I prefer the Korean rules treatment of the position.

But the fact that relevant area hasn't been defined in a ruleset professionals use amuses me, haha, it's not really a ruleset then but just guidelines I guess. I think the best rulesets are the ones that feel as basic and not like a ruleset, rather than snobbishly fill itself with endless complexities like an upper class dinner with fancy rules to show how much better it is than you, that's what unnecessarily complicated rulesets feel like to me, haha

Perhaps in football standards it's not a dispute unless it looks like a prequel to a battle scene from the godfather movies, while in baduk just mildly not reacting in immediate agreement with your opponent or the officials counts as a major dramatic discourse

This is not a "normal" quadruple ko (which would be draw / no result in any Asian ruleset) but a combined moonshine life. W cannot capture during normal game (that would lead to quadruple ko), but he can try to pass instead and claim that B is not alive. Japanese and Korean rules verify those claims under altered rules.

So the question is whether B can use threats in the double ko seki to prove life of outer stones - AFTER the game have stopped. This is a rare case where even Asian rulesets differ. Both Chinese and Japanese rules see B not alive (each with its own logic, but both prevent using the double ko for life). Korean rules are an exception where it remains usable - seen as inside "relevant" area.


As you wrote Chinese rules use human judgement and list examples to help it. They list triple ko as draw and moonshine life as dead. The superko principle - when applied - should allow capturing B, and they seem to apply it for moonshine life while not apply for real triple ko. Otoh I never heard Chinese rules distinguish between remote moonshine life and a combined one. Closest they list seems a quadruple ko seki. What did you mean here?

I know a combined moonshine life has been finished as a draw in a Chinese tournament. But after a careful check, maybe it is a decision of the players, not the referee. http://sports.sina.com.cn/go/2008-07-19 ... 3722.shtml

But, it is weird to conclude that the combined moonshine life is dead in the Chinese rule. The rule text lists four examples of the draw, and the last one is a combined moonshine life with two external kos. Is that a draw because there are more than one external ko? But having one, two, or say 10 external kos does not really change the nature of this situation. If a group with no eye and two half-point ko's is made in a separate position (remote from the double ko seki), I don't think that group would be called alive. I am sure separated and combined moonshine lives are treated differently, at least in some cases.

_________________
Jaeup Kim
Professor in Physics, Ulsan National Institute of Science and Technology, Korea
Author of the Book "Understanding the Rules of Baduk", available at https://home.unist.ac.kr/professor/jukim/bbs/board.php?bo_table=notice&wr_id=5

I'm not sure which version you refer to, but beware a huge typo in Jasiek's English copy of the 2002 version (outer W stones missing in dia 4 - cf. in Chinese). The 1988 version shows the correct original of the same example (dia 5 there). "Four kos among three groups" if you meant this?

I don't think that example can be considered moonshine life, since both sides can press repetition in main game already, and one of them will likely do so. The game won't see two passes or a scoring position, so the question of moonshine life won't arise (the 1988 version is more expressive, and above dia 3 moonshine life it clearly says "at the end of the game"). Its dia 5 seems same as two remote double ko sekis - forced quadruple ko repetition in main game.

Wow.. that's exactly the confusion I had. Now I am speechless.

I actually checked both the English and Chinese versions when writing my book. But the Chinese version I looked up also had the same shape without the outer White stones (pdf file attached), and I never thought that there could possibly be a mistake. I assume Chinese Qiyuan distributed the wrong version at some time, and it is not the translator's fault.

_________________
Jaeup Kim
Professor in Physics, Ulsan National Institute of Science and Technology, Korea
Author of the Book "Understanding the Rules of Baduk", available at https://home.unist.ac.kr/professor/jukim/bbs/board.php?bo_table=notice&wr_id=5