"RobertJasiek","Strategic Consequences of New WMSG Rules Changes","2008-09-13 23:38:10","The following was written and formatted for rec.games.go. If somebody feels like it, feel free to provide nicer diagrams. I post it also here to give more WMSG participants a chance to develop good endgame strategy. \ \ ************************************************************************** \ \ The following discussion relies on the probably or possibly used rules \ version on 2008-09-14. For the sake of simplicity, no komi is used in \ the examples (unless stated otherwise) and shared intersections are \ not counted at all. The player to move is given due to the numbers of \ black and white stones on the board. Recall that under World Mind \ Sports Games Rules White making the first pass reduces Black's score \ by 1. \ \ ************************************************************************** \ \ 1) Forgotten ordinary endgame: \ \ . . . . . \ # # # # # \ # # . O O \ O O O O O \ . . . . . \ \ 1, 2 = pass \ Black makes the first pass. \ \ . . . . . \ # # # # # \ # # C O O \ O O O O O \ . . . . . \ \ C is a contestable point, which may not be taken during a resumption. \ \ Score = 12 - 12 - 0 = 0. \ \ It was a strategic mistake of White 2 not to play the two-sided dame, \ which is an ordinary endgame move; otherwise the score would have been \ -1. It is possible to play all ordinary endgame before the first \ succession of two passes. Also Black 1 might have taken the two-sided \ dame for the same score 0. \ \ ************************************************************************** \ \ 2) Ordinary basic endgame kos / study if passing is allowed when basic \ endgame kos are unfilled: \ \ . . . . \ # # # # \ . # O # \ # O . O \ O O O O \ . . . . \ \ Variation 2.1: \ \ . . . . \ # # # # \ 1 # O # \ # O 2 O \ O O O O \ . . . . \ \ 3, 4 = pass \ Black makes the first pass. \ There is no need to consider resumption. \ Score = 12 - 12 - 0 = 0. \ \ Variation 2.2: \ \ . . . . \ # # # # \ 1 # O # \ # O 3 O \ O O O O \ . . . . \ \ 2, 4, 5 = pass. \ White makes the first pass. \ White 2 is a strategic mistake. \ There is no need to consider resumption: Superko prohibits recapture. \ \ . . . . \ # # # # \ # # N # \ # O # O \ O O O O \ . . . . \ \ N is not a contestable point because each perfect play (starting from \ White pass, White divides his eyespace, or White plays in Black's \ territory) lets Black control N. \ \ Score = 14 - 10 - 1 = 3. \ \ Variation 2.3: \ \ . . . . \ # # # # \ . # O # \ # O . O \ O O O O \ . . . . \ \ 1, 2 = pass \ Black makes the first pass. \ There is no resumption. \ \ . . . . \ # # # # \ C # C # \ C O C O \ O O O O \ . . . . \ \ C are the contestable points because Black 3 could either connect the \ left ko or capture the right ko with White then capturing the left ko. \ \ B B B B \ B B B B \ S B S B \ S W S W \ W W W W \ B B B B \ \ B = scoring for Black, W = scoring for White, S = shared. \ Score = 10 - 10 - 0 = 0. \ \ It is particularly remarkable which the contestable and thus shared \ points are. In particular it is surprising that also filled \ intersections are shared. \ \ One may argue though that the rules contradict themselves because the \ contestable points rules see the ko stones' intersections as shared \ points while the all stones are alive after resumption rule and the \ area scoring definition suggest scoring points for the player \ occupying an intersection. \ \ Here we see a possible reason why mandatory ko filling rules are being \ discussed. \ \ Variations 2.4+: \ \ Black captures, White captures, etc.: not discussed. \ \ ************************************************************************** \ \ 3) Ordinary basic endgame kos / study if passing is prohibited when \ basic endgame kos are unfilled: \ \ . . . . \ # # # # \ . # O # \ # O . O \ O O O O \ . . . . \ \ Variation 3.1: \ \ . . . . \ # # # # \ 1 # O # \ # O 2 O \ O O O O \ . . . . \ \ 3, 4 = pass \ Black makes the first pass. \ There is no need to consider resumption. \ Score = 12 - 12 - 0 = 0. \ \ ************************************************************************** \ \ 4) Exceptional basic endgame kos / study if passing is allowed when \ basic endgame kos are unfilled: \ \ (Exceptional not because of the ko threat but because an early first \ pass is interesting.) \ \ . . . . . . \ # # # # # # \ . # O # . . \ # O . O # # \ O O O O O O \ . O . O . . \ \ Variation 4.1: \ \ . . . . . . \ # # # # # # \ 1 # O # . . \ # O 2 O # # \ O O O O O O \ . O . O . . \ \ 3, 4 = pass \ Black makes the first pass. \ There is no need to consider resumption. \ Score = 20 - 16 - 0 = 4. \ \ Black 1 is a strategic mistake because White chooses Variation 4.4 \ instead of 4.1, Variations 4.2 and 4.3 are better for Black. \ White 2 is a strategic mistake, Variation 4.4 is better for White. \ \ Variation 4.2: \ \ . . . . . . \ # # # # # # \ 3 # O # . . \ # O 2 O # # \ O O O O O O \ . O . O . . \ \ 1, 4, 5 = pass \ Black makes the first pass. \ There is no need to consider resumption. \ Score = 20 - 16 - 0 = 4. \ \ Variation 4.3: \ \ . . . . . . \ # # # # # # \ . # O # . . \ # O . O # # \ O O O O O O \ . O . O . . \ \ 1, 2 = pass \ Black makes the first pass. \ \ . . . . . . \ # # # # # # \ C # C # . . \ C O C O # # \ O O O O O O \ . O . O . . \ \ C are the contestable points. The ko threat region does not have \ contestable points because each perfect play would let Black defend \ there should White throw in simply or as a ko threat. \ \ If there should be a resumption, then - presumably (this is my current \ interpretation) - playing on C is prohibited because that would turn \ shared points into points of either player. \ \ . . . . . . \ # # # # # # \ S # S # . . \ S O S O # # \ O O O O O O \ . O . O . . \ \ S are the shared points. \ Score = 18 - 14 - 0 = 4. \ \ Variation 4.4: \ \ . . . . . . \ # # # # # # \ 1 # 6 # 4 5 \ # O 3 O # # \ O O O O O O \ . O . O . . \ \ 2, 7, 8 = pass \ White makes the first pass. \ \ . . . . . . \ # # # # # # \ # # O # . # \ # O . O # # \ O O O O O O \ . O . O . . \ \ position at the first succession of two passes \ There is no need to consider resumption: Superko prohibits recapture. \ \ . . . . . . \ # # # # # # \ # # O # . # \ # O N O # # \ O O O O O O \ . O . O . . \ \ N is not a contestable point because each perfect play lets White \ control N. \ \ Score = 20 - 16 - 1 = 3. \ \ Black 1 is a strategic mistake: Variations 4.2 and 4.3 are better for \ Black. \ \ ************************************************************************** \ \ 5) Exceptional basic endgame kos / study if passing is prohibited when \ basic endgame kos are unfilled: \ \ (Exceptional not because of the ko threat but because an early first \ pass is interesting.) \ \ . . . . . . \ # # # # # # \ . # O # . . \ # O . O # # \ O O O O O O \ . O . O . . \ \ Variation 5.1: \ \ . . . . . . \ # # # # # # \ 1 # O # . . \ # O 2 O # # \ O O O O O O \ . O . O . . \ \ 3, 4 = pass \ Black makes the first pass. \ There is no need to consider resumption. \ Score = 20 - 16 - 0 = 4. \ \ Variation 5.2: \ \ . . . . . . \ # # # # # # \ 2 # 3 # . . \ 4 O 1 O # # \ O O O O O O \ . O . O . . \ \ 5, 6 = pass \ Black makes the first pass. \ There is no need to consider resumption. \ Score = 20 - 16 - 0 = 4. \ \ If passing is prohibited when basic endgame kos are unfilled, then \ strategy is simpler than otherwise because the strategic option of \ delaying a ko connection to take the first pass when having the \ superior difference of ko threats does not exist. However, the \ exceptional rule about prohibiting passing when basic endgame kos are \ unfilled currently also makes the rules design inconsistent: There are \ other positions without basic endgame kos where passing before \ completing the endgame is better. \ \ ************************************************************************** \ \ 6) Passing before filling privilege dame / study if passing is allowed \ when basic endgame kos are unfilled: \ \ . # . # . \ O # # . . \ . O O # # \ O . O O # \ . O . O # \ \ Variation 6.1: \ \ 3 # . # . \ O # # . . \ 2 O O # # \ O . O O # \ . O . O # \ \ 1, 4, 5 = pass \ Black makes the first pass. \ There is no need to consider resumption. \ Score = 13 - 12 - 0 = 1. \ \ White 2 is a strategic mistake, Variation 6.4 is better for White. \ \ Variation 6.2: \ \ 2 # . # . \ 6 # # 4 5 \ 3 O O # # \ O . O O # \ . O . O # \ \ 1, 9, 10 = pass \ 7 @ 2, 8 @ 3 \ Black makes the first pass. \ There is no need to consider resumption. \ \ # # . # . \ O # # . # \ O O O # # \ O . O O # \ . O . O # \ \ position at the first succession of two passes \ Score = 13 - 12 - 0 = 1. \ \ White 2 is a strategic mistake, Variation 6.4 is better for White. \ \ Variation 6.3: \ \ 1 # . # . \ 6 # # 4 5 \ 3 O O # # \ O . O O # \ . O . O # \ \ 2, 7, 8 = pass \ White makes the first pass. \ \ # # . # . \ O # # . # \ . O O # # \ O . O O # \ . O . O # \ \ position at the first succession of two passes \ There is no need to consider resumption: Superko prohibits recapture. \ \ # # . # . \ O # # . # \ N O O # # \ O . O O # \ . O . O # \ \ N is not a contestable point because each perfect play lets White \ control N. \ \ Score = 13 - 12 - 1 = 0. \ \ Black 1 is one possible perfect play. \ \ Variation 6.4: \ \ . # . # . \ O # # . . \ . O O # # \ O . O O # \ . O . O # \ \ 1, 2 = pass \ Black makes the first pass. \ \ C # . # . \ ? # # . . \ ? O O # # \ O . O O # \ . O . O # \ \ From the 1-sided dame precedental(!) ruling, one has to imply that not \ taken privilege intersections are contestable points. Therefore C is a \ contestable point. \ \ Things are more difficult for the ? intersections. Whichever strategy \ Black tries, he cannot force White to abandon them. However, is each \ such strategy perfect play for both players? One can argue that \ perfect play is required for identifying contestable points. Proof: \ Because of the move-sequences of Variations 6.2 and 6.3 starting with \ move 2 and the trivial move-sequence starting with Black, each perfect \ play leads to the ? intersections controlled by White. Therefore they \ are not contestable points. Similarly, none of the points nearby the \ ko threat aji is a contestable point. Summary: \ \ C # . # . \ O # # . . \ . O O # # \ O . O O # \ . O . O # \ \ A resumption may not see a play on C because that would contest a \ point supposed to be shared. \ \ S # . # . \ O # # . . \ . O O # # \ O . O O # \ . O . O # \ \ S = shared point \ Score = 12 - 12 - 0 = 0. \ \ Black 1 is one possible perfect play. \ \ Conclusion: \ \ The effect of the implications of the 1-sided dame precedental ruling \ alter strategies and perfect play in this example. \ \ ************************************************************************** \ \ 7) Passing before filling privilege dame / study if passing is \ prohibited when basic endgame kos are unfilled: \ \ . # . # . \ O # # . . \ . O O # # \ O . O O # \ . O . O # \ \ Variation 7.1: \ \ 3 # . # . \ O # # . . \ 2 O O # # \ O . O O # \ . O . O # \ \ 1, 4, 5 = pass \ Black makes the first pass. \ There is no need to consider resumption. \ Score = 13 - 12 - 0 = 1. \ \ White 2 is a strategic mistake, Variation 7.4 is better for White. \ \ Variation 7.2: \ \ 2 # . # . \ 6 # # 4 5 \ 3 O O # # \ O . O O # \ . O . O # \ \ 1, 9, 10 = pass \ 7 @ 2, 8 @ 3 \ Black makes the first pass. \ There is no need to consider resumption. \ \ # # . # . \ O # # . # \ O O O # # \ O . O O # \ . O . O # \ \ position at the first succession of two passes \ Score = 13 - 12 - 0 = 1. \ \ White 2 is a strategic mistake, Variation 7.4 is better for White. \ \ Variation 7.3: \ \ 1 # . # . \ O # # . . \ 2 O O # # \ O . O O # \ . O . O # \ \ 3, 4 = pass \ Black makes the first pass. \ There is no need to consider resumption. \ \ Score = 13 - 12 - 0 = 1. \ \ Note how this differs from the other rules variant! Because White is \ forced by the rule not to pass, he fills the ko. He may not gain an \ extra point by making the first pass before filling the ko. \ \ Black 1 is the only perfect play. \ \ Variation 7.4: \ \ . # . # . \ O # # . . \ . O O # # \ O . O O # \ . O . O # \ \ 1, 2 = pass \ Black makes the first pass. \ \ C # . # . \ ? # # . . \ ? O O # # \ O . O O # \ . O . O # \ \ From the 1-sided dame precedental(!) ruling, one has to imply that not \ taken privilege intersections are contestable points. Therefore C is a \ contestable point. \ \ Things are more difficult for the ? intersections. Whichever strategy \ Black tries, he cannot force White to abandon them. However, is each \ such strategy perfect play for both players? One can argue that \ perfect play is required for identifying contestable points. Proof: \ Because of the move-sequences of Variations 7.2 and 7.3 starting with \ move 2 and the trivial move-sequence starting with Black, each perfect \ play leads to the ? intersections controlled by White. Therefore they \ are not contestable points. Similarly, none of the points nearby the \ ko threat aji is a contestable point. Summary: \ \ C # . # . \ O # # . . \ . O O # # \ O . O O # \ . O . O # \ \ A resumption may not see a play on C because that would contest a \ point supposed to be shared. \ \ S # . # . \ O # # . . \ . O O # # \ O . O O # \ . O . O # \ \ S = shared point \ Score = 12 - 12 - 0 = 0. \ \ Black 1 is a strategic mistake. \ \ Conclusion: \ \ The change of the ko filling rule changes Variation 7.3 from 6.3. This \ has severe effects on which moves are strategically correct. In this \ example, perfect play strategy under \"prohibited passing when there \ are unfilled kos\" is simpler than under \"allowed passing when there \ are unfilled kos\" but the analysis and especially the contestable \ rules interpretation why this is so is equally difficult. Application \ of the contestable points rules and the 1-sided dame precedent is far \ beyond ordinary players' ability. They may be lucky that with \ unawareness under \"prohibited passing when there are unfilled kos\" \ they will find correct strategy accidentally!!! However, in other \ exceptional shapes (which have not been discovered yet) they do not \ have a realistic chance to find correct strategy by means of \ supposedly correct rules and precedents interpretation. \ \ ************************************************************************** \ \ 8) 1-sided dame: \ \ +--------------+ \ | . O O # . . .| \ | # # O # . . .| \ | # # O # # # #| \ | . . O # O O O| \ | O O O # O O O| \ | # # # # O O .| \ | . # . # O . O| \ +--------------+ \ \ komi = 6.5 \ \ Variation 8.1: \ \ +--------------+ \ | 1 O O # . . .| \ | # # O # . . .| \ | # # O # # # #| \ | . . O # O O O| \ | O O O # O O O| \ | # # # # O O .| \ | . # . # O . O| \ +--------------+ \ \ 2, 3 = pass. \ White makes the first pass. \ There is no need to consider resumption. \ Score = (19+8) - (18+2) - 6.5 - 1 = 27 - 20 - 6.5 - 1 = -0.5. \ \ Variation 8.2: \ \ +--------------+ \ | . O O # . . .| \ | # # O # . . .| \ | # # O # # # #| \ | . . O # O O O| \ | O O O # O O O| \ | # # # # O O .| \ | . # . # O . O| \ +--------------+ \ \ 1, 2 = pass. \ Black makes the first pass. \ \ +--------------+ \ | C O O # . . .| \ | # # O # . . .| \ | # # O # # # #| \ | C C O # O O O| \ | O O O # O O O| \ | # # # # O O .| \ | . # . # O . O| \ +--------------+ \ \ C (or exactly one arbitrary point of all C points?) are contestable \ points, by precedental ruling. (Interpretation of the rules text of \ 2008-07-15 was ambiguous, so a precedental ruling has been necessary.) \ Therefore, during a resumption, play there is prohibited. \ \ +--------------+ \ | S O O # . . .| \ | # # O # . . .| \ | # # O # # # #| \ | S S O # O O O| \ | O O O # O O O| \ | # # # # O O .| \ | . # . # O . O| \ +--------------+ \ \ S = shared points \ Score = 26 - 20 - 6.5 - 0 = -0.5. \ \ Conclusion: \ \ Playing or not playing a 1-sided dame in this example are both \ strategically correct. In effect, the 1-sided dame here is unvaluable. \ \ ************************************************************************** \ \ 9) 1-sided dame (colour-inverse position): \ \ +--------------+ \ | . # # O . . .| \ | O O # O . . .| \ | O O # O O O O| \ | . . # O # # #| \ | # # # O # # #| \ | O O O O # # .| \ | . O . O # . #| \ +--------------+ \ \ komi = 6.5 \ \ Variation 9.1: \ \ +--------------+ \ | 2 # # O . . .| \ | O O # O . . .| \ | O O # O O O O| \ | . . # O # # #| \ | # # # O # # #| \ | O O O O # # .| \ | . O . O # . #| \ +--------------+ \ \ 1, 3, 4 = pass. \ Black makes the first pass. \ Score = (18+8) - (19+2) - 6.5 - 0 = 26 - 21 - 6.5 - 0 = -1.5. \ \ White 2 is strategically correct. \ \ Variation 9.2: \ \ +--------------+ \ | . # # O . . .| \ | O O # O . . .| \ | O O # O O O O| \ | . . # O # # #| \ | # # # O # # #| \ | O O O O # # .| \ | . O . O # . #| \ +--------------+ \ \ 1, 2 = pass \ Black makes the first pass. \ \ +--------------+ \ | C # # O . . .| \ | O O # O . . .| \ | O O # O O O O| \ | C C # O # # #| \ | # # # O # # #| \ | O O O O # # .| \ | . O . O # . #| \ +--------------+ \ \ C (or exactly one arbitrary point of all C points?) are contestable \ points, by precedental ruling. (Interpretation of the rules text of \ 2008-07-15 was ambiguous, so a precedental ruling has been necessary.) \ Therefore, during a resumption, play there is prohibited. \ \ +--------------+ \ | S # # O . . .| \ | O O # O . . .| \ | O O # O O O O| \ | S S # O # # #| \ | # # # O # # #| \ | O O O O # # .| \ | . O . O # . #| \ +--------------+ \ \ S = shared points \ Score = 26 - 20 - 6.5 - 0 = -0.5. \ \ White 2 is a strategic mistake. \ \ Conclusion: \ \ In this example, the player who has 1-sided dame available can play \ there before the first succession of two passes. The opponent gets the \ first pass because it is his move. Contrarily in example (8), the \ player to move cannot take advantage of having the right to move \ because by doing so (taking the first pass) he loses his right to take \ a 1-sided dame. Hence the position and its colour-inverse behave \ strategically differently. This is the effect of the contestable \ points rules in combination with the precedental ruling that 1-sided \ dame are contestable points. It is not the effect of the first pass's \ value because under the Simplified World Mind Sports Games Rules the \ player to move can take the first pass and the player with the 1-sided \ dame privilege may take them, regardless of whether we have position \ (8) or its colour-inverse (9). \ \ Not surprisingly, possible exceptional rules for 1-sided dame are \ being discussed to get out of the bad consequences of the official \ contestable points rules. \ \ ************************************************************************** \ \ Overall conclusion: \ \ Simplified World Mind Sports Games Rules or a certain interpretation \ of the World Mind Sports Games Rules of 2008-07-15 lead to (rather) \ simple rules application but unexpected strategic behaviour in \ exceptional positions where the first pass as tedomari is more urgent \ than endgame plays on the board with greater miai values. Although \ this is conceptually not different from middle game or earlier endgame \ tedomari, such strategic behaviour related to the first pass has been \ unknown under both Chinese Rules and Japanese/Korean Rules. Currently \ the Chinese and Japanese appear to be looking for a compromise that \ introduces exceptional, additional rules (or precedents, until further \ explicit, written exceptional, additional rules will be added) with \ the apparent aim to eliminate exceptional, previously unknown \ strategic behaviour in exceptional (but mostly rather common) \ positions. There is a problem though: Nobody knows whether or how many \ other shape classes with such exceptional, previously unknown \ strategic behaviour might exist. So there is the danger that an \ endless list of precedental rulings will have to be created and that \ the treatment is worse than what some perceive a desease. I recommend \ Simplified World Mind Sports Games Rules because understanding them \ and their strategic consequences is simpler than understanding a lot \ of difficult rules and their application. So far it seems that the \ Asian rules gurus (are going) fulfil their intended job to remove \ strategic difficulties by putting them into difficult rules instead; \ so the ordinary player might simply pray, not think, and expect \ everything to go well on the board. There is no guarantee though that \ what works for basic endgame kos will work in general. E.g., it is \ already difficult to distinguish basic endgame kos from dead kos or \ double ko sekis. Disputes will arise about whether having to fill such \ before making the first pass. Etc. This is a conceptually dangerous \ game, and rules history might see the troubles of endless precedents \ again as happened with the Japanese 1949 Rules. Why not do things \ right immediately by adopting the Simplified World Mind Sports Games \ Rules? We love go because of its strategic complexity - not because \ rules prohibit our thinking. Some strategy might be new - why not \ welcome it as an enrichment? Superko is also a strategic enrichment. \ Let us design our rules with courage rather than fear!" "Reed338","","2008-09-14 00:04:22","whoa, you've got some time on your hands" "John Fairbairn","","2008-09-14 02:35:03","[quote]whoa, you've got some time on your hands[/quote] \ \ But recall that to be an expert you have to know more and more about less and less." "Hicham","","2008-09-14 04:12:35","Hi Robert, could you try editing that to the diagram format used in this forum? That is way easier to read then this ascii looking go diagrams.\ \ (still reading the rest of the post, trying to understand it)" "kirkmc","","2008-09-14 05:09:33","Huh?" "Fedya","","2008-09-14 07:07:32","And yet, hundreds of thousands -- if not millions -- of people are able to play games every day and enjoy themselves despite the rules' being a complete cluster****." "RobertJasiek","","2008-09-14 07:14:19","I am not sure if I could. I guess I can figure out again how to create diagrams at all. But I would have problems with move numbers greater than 10, passes, character labels, same character labels on different black, white, empty intersections. The SL style diagrams are very bad for that purpose. It would be easier to create a PDF. \ \ Even worse, I have already spent 4 hours on the post and prefer to think about new contents (like searching for a general definition of \"basic endgame ko\") rather than fighting with online editors. \ \ You might a) copy & paste into an editor with fixed width font, b) do the editing here yourself, c) not read at all, or d) wait for my publication of similar contents elsewhere (maybe my webpage). \ \ But don't expect me to have arbitrarily much time." "RobertJasiek","","2008-09-14 07:27:52","Fedya, do those millions of people participate in the WMSG or have the intention to play their best endgame until the very end...? Strategic explanations of late endgames are not for those that don't care if they lose 0 to 20 points by not knowing the rules application well or that accuse players with perfect late endgame to have bad manners or to cheat instead of criticising the opponent's (or their own) weak late endgame." "TMark","","2008-09-14 07:52:28","Ithink that Robert is, or has become, the Large Hadron Collider of Western Go, searching for smaller and smaller particles of definition within the rules. Unfortunately, one word constantly misused by him is \"strategic\" which has no place within the ever diminishing minor nit-picking details of possible ko captures to the nth degree. \ \ Read them at your peril! Best wishes." "RobertJasiek","","2008-09-14 08:41:45","Replace \"strategic\" by \"tactical\" if you like, but the meaning in the context of rules game tree choices is the same. In case of doubt, either shall refer to the entire game tree. \"strategic\" is an exaggeration - \"tactical\" is an understatement." "ferl","by the way...","2008-09-14 11:08:29","Do you still play go ? \ And if so, do you still enjoy it ? \ liebe gruesse \ ferl" "RobertJasiek","","2008-09-14 11:59:52","Me? Do I belong to the 10 or only 100 most frequently playing Europeans...? :)" "sol.ch","","2008-09-14 13:27:16","Robert Jasiek's post (reformatted to show diagrams):\ \ The following was written and formatted for rec.games.go. If somebody feels like it, feel free to provide nicer diagrams. I post it also here to give more WMSG participants a chance to develop good endgame strategy.\ \ ************************************************** ************************\ \ The following discussion relies on the probably or possibly used rules\ version on 2008-09-14. For the sake of simplicity, no komi is used in\ the examples (unless stated otherwise) and shared intersections are\ not counted at all. The player to move is given due to the numbers of\ black and white stones on the board. Recall that under World Mind\ Sports Games Rules White making the first pass reduces Black's score\ by 1.\ \ ************************************************** ************************\ \ 1) Forgotten ordinary endgame:\ \ \ [go]\ $$ -------------\ $$ | . . . . . |\ $$ | X X X X X |\ $$ | X X . O O |\ $$ | O O O O O |\ $$ | . . . . . |\ $$ -------------\ [/go]\ \ 1, 2 = pass\ Black makes the first pass.\ \ [go]\ $$ -------------\ $$ | . . . . . |\ $$ | X X X X X |\ $$ | X X c O O |\ $$ | O O O O O |\ $$ | . . . . . |\ $$ -------------\ [/go]\ \ C is a contestable point, which may not be taken during a resumption.\ \ Score = 12 - 12 - 0 = 0.\ \ It was a strategic mistake of White 2 not to play the two-sided dame,\ which is an ordinary endgame move; otherwise the score would have been\ -1. It is possible to play all ordinary endgame before the first\ succession of two passes. Also Black 1 might have taken the two-sided\ dame for the same score 0.\ \ ************************************************** ************************\ \ 2) Ordinary basic endgame kos / study if passing is allowed when basic\ endgame kos are unfilled:\ \ [go]\ $$ ------------\ $$ | . . . . |\ $$ | X X X X |\ $$ | . X O X |\ $$ | X O . O |\ $$ | O O O O |\ $$ | . . . . |\ $$ ------------\ [/go]\ \ Variation 2.1:\ \ [go]\ $$ ------------\ $$ | . . . . |\ $$ | X X X X |\ $$ | 1 X O X |\ $$ | X O 2 O |\ $$ | O O O O |\ $$ | . . . . |\ $$ ------------\ [/go]\ \ 3, 4 = pass\ Black makes the first pass.\ There is no need to consider resumption.\ Score = 12 - 12 - 0 = 0.\ \ Variation 2.2:\ \ [go]\ $$ ------------\ $$ | . . . . |\ $$ | X X X X |\ $$ | 1 X O X |\ $$ | X O 3 O |\ $$ | O O O O |\ $$ | . . . . |\ $$ ------------\ [/go]\ \ 2, 4, 5 = pass.\ White makes the first pass.\ White 2 is a strategic mistake.\ There is no need to consider resumption: Superko prohibits recapture.\ \ [go]\ $$ ------------\ $$ | . . . . |\ $$ | X X X X |\ $$ | X X n X |\ $$ | X O X O |\ $$ | O O O O |\ $$ | . . . . |\ $$ ------------\ [/go]\ \ N is not a contestable point because each perfect play (starting from\ White pass, White divides his eyespace, or White plays in Black's\ territory) lets Black control N.\ \ Score = 14 - 10 - 1 = 3.\ \ Variation 2.3:\ \ [go]\ $$ ------------\ $$ | . . . . |\ $$ | X X X X |\ $$ | . X O X |\ $$ | X O . O |\ $$ | O O O O |\ $$ | . . . . |\ $$ ------------\ [/go]\ \ 1, 2 = pass\ Black makes the first pass.\ There is no resumption.\ \ [go]\ $$ ------------\ $$ | . . . . |\ $$ | X X X X |\ $$ | c X c X |\ $$ | c O c O |\ $$ | O O O O |\ $$ | . . . . |\ $$ ------------\ [/go]\ \ C are the contestable points because Black 3 could either connect the\ left ko or capture the right ko with White then capturing the left ko.\ \ [go]\ $$ -----------\ $$ | b b b b |\ $$ | b b b b |\ $$ | s b s b |\ $$ | s w s w |\ $$ | w w w w |\ $$ | w w w w |\ $$ -----------\ [/go]\ \ B = scoring for Black, W = scoring for White, S = shared.\ Score = 10 - 10 - 0 = 0.\ \ It is particularly remarkable which the contestable and thus shared\ points are. In particular it is surprising that also filled\ intersections are shared.\ \ One may argue though that the rules contradict themselves because the\ contestable points rules see the ko stones' intersections as shared\ points while the all stones are alive after resumption rule and the\ area scoring definition suggest scoring points for the player\ occupying an intersection.\ \ Here we see a possible reason why mandatory ko filling rules are being\ discussed.\ \ Variations 2.4+:\ \ Black captures, White captures, etc.: not discussed.\ \ ************************************************** ************************\ \ 3) Ordinary basic endgame kos / study if passing is prohibited when\ basic endgame kos are unfilled:\ \ [go]\ $$ ------------\ $$ | . . . . |\ $$ | X X X X |\ $$ | . X O X |\ $$ | X O . O |\ $$ | O O O O |\ $$ | . . . . |\ $$ ------------\ [/go]\ \ Variation 3.1:\ \ [go]\ $$ ------------\ $$ | . . . . |\ $$ | X X X X |\ $$ | 1 X O X |\ $$ | X O 2 O |\ $$ | O O O O |\ $$ | . . . . |\ $$ ------------\ [/go]\ \ 3, 4 = pass\ Black makes the first pass.\ There is no need to consider resumption.\ Score = 12 - 12 - 0 = 0.\ \ ************************************************** ************************\ \ 4) Exceptional basic endgame kos / study if passing is allowed when\ basic endgame kos are unfilled:\ \ (Exceptional not because of the ko threat but because an early first\ pass is interesting.)\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | . X O X . . |\ $$ | X O . O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ Variation 4.1:\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | 1 X O X . . |\ $$ | X O 2 O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ 3, 4 = pass\ Black makes the first pass.\ There is no need to consider resumption.\ Score = 20 - 16 - 0 = 4.\ \ Black 1 is a strategic mistake because White chooses Variation 4.4\ instead of 4.1, Variations 4.2 and 4.3 are better for Black.\ White 2 is a strategic mistake, Variation 4.4 is better for White.\ \ Variation 4.2:\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | 3 X O X . . |\ $$ | X O 2 O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ 1, 4, 5 = pass\ Black makes the first pass.\ There is no need to consider resumption.\ Score = 20 - 16 - 0 = 4.\ \ Variation 4.3:\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | . X O X . . |\ $$ | X O . O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ 1, 2 = pass\ Black makes the first pass.\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | c X c X . . |\ $$ | c O c O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ C are the contestable points. The ko threat region does not have\ contestable points because each perfect play would let Black defend\ there should White throw in simply or as a ko threat.\ \ If there should be a resumption, then - presumably (this is my current\ interpretation) - playing on C is prohibited because that would turn\ shared points into points of either player.\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | s X s X . . |\ $$ | s O s O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ S are the shared points.\ Score = 18 - 14 - 0 = 4.\ \ Variation 4.4:\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | 1 X 6 X 4 5 |\ $$ | X O 3 O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ 2, 7, 8 = pass\ White makes the first pass.\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | X X O X . X |\ $$ | X O . O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ position at the first succession of two passes\ There is no need to consider resumption: Superko prohibits recapture.\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | X X O X . X |\ $$ | X O n O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ N is not a contestable point because each perfect play lets White\ control N.\ \ Score = 20 - 16 - 1 = 3.\ \ Black 1 is a strategic mistake: Variations 4.2 and 4.3 are better for\ Black.\ \ ************************************************** ************************\ \ 5) Exceptional basic endgame kos / study if passing is prohibited when\ basic endgame kos are unfilled:\ \ (Exceptional not because of the ko threat but because an early first\ pass is interesting.)\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | . X O X . . |\ $$ | X O . O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ Variation 5.1:\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | 1 X O X . . |\ $$ | X O 2 O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ \ $$ ---------------\ [/go]\ \ 3, 4 = pass\ Black makes the first pass.\ There is no need to consider resumption.\ Score = 20 - 16 - 0 = 4.\ \ Variation 5.2:\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | 2 X 3 X . . |\ $$ | 4 O 1 O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ 5, 6 = pass\ Black makes the first pass.\ There is no need to consider resumption.\ Score = 20 - 16 - 0 = 4.\ \ If passing is prohibited when basic endgame kos are unfilled, then\ strategy is simpler than otherwise because the strategic option of\ delaying a ko connection to take the first pass when having the\ superior difference of ko threats does not exist. However, the\ exceptional rule about prohibiting passing when basic endgame kos are\ unfilled currently also makes the rules design inconsistent: There are\ other positions without basic endgame kos where passing before\ completing the endgame is better.\ \ ************************************************** ************************\ \ 6) Passing before filling privilege dame / study if passing is allowed\ when basic endgame kos are unfilled:\ \ [go]\ $$ -------------\ $$ | . X . X . |\ $$ | O X X . . |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ Variation 6.1:\ \ [go]\ $$ -------------\ $$ | 3 X . X . |\ $$ | O X X . . |\ $$ | 2 O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ 1, 4, 5 = pass\ Black makes the first pass.\ There is no need to consider resumption.\ Score = 13 - 12 - 0 = 1.\ \ White 2 is a strategic mistake, Variation 6.4 is better for White.\ \ Variation 6.2:\ \ [go]\ $$ -------------\ $$ | 2 X . X . |\ $$ | 6 X X 4 5 |\ $$ | 3 O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ 1, 9, 10 = pass\ 7 @ 2, 8 @ 3\ Black makes the first pass.\ There is no need to consider resumption.\ \ [go]\ $$ -------------\ $$ | X X . X . |\ $$ | O X X . X |\ $$ | O O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ position at the first succession of two passes\ Score = 13 - 12 - 0 = 1.\ \ White 2 is a strategic mistake, Variation 6.4 is better for White.\ \ Variation 6.3:\ \ [go]\ $$ -------------\ $$ | 1 X . X . |\ $$ | 6 X X 4 5 |\ $$ | 3 O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ 2, 7, 8 = pass\ White makes the first pass.\ \ [go]\ $$ -------------\ $$ | X X . X . |\ $$ | O X X . X |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ position at the first succession of two passes\ There is no need to consider resumption: Superko prohibits recapture.\ \ [go]\ $$ -------------\ $$ | X X . X . |\ $$ | O X X . X |\ $$ | n O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ N is not a contestable point because each perfect play lets White\ control N.\ \ Score = 13 - 12 - 1 = 0.\ \ Black 1 is one possible perfect play.\ \ Variation 6.4:\ \ [go]\ $$ -------------\ $$ | . X . X . |\ $$ | O X X . . |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ 1, 2 = pass\ Black makes the first pass.\ \ [go]\ $$ -------------\ $$ | c X . X . |\ $$ | a X X . . |\ $$ | a O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ From the 1-sided dame precedental(!) ruling, one has to imply that not\ taken privilege intersections are contestable points. Therefore C is a\ contestable point.\ \ Things are more difficult for the 'a' intersections. Whichever strategy\ Black tries, he cannot force White to abandon them. However, is each\ such strategy perfect play for both players? One can argue that\ perfect play is required for identifying contestable points. Proof:\ Because of the move-sequences of Variations 6.2 and 6.3 starting with\ move 2 and the trivial move-sequence starting with Black, each perfect\ play leads to the 'a' intersections controlled by White. Therefore they\ are not contestable points. Similarly, none of the points nearby the\ ko threat aji is a contestable point. Summary:\ \ [go]\ $$ -------------\ $$ | c X . X . |\ $$ | O X X . . |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ A resumption may not see a play on C because that would contest a\ point supposed to be shared.\ \ [go]\ $$ -------------\ $$ | s X . X . |\ $$ | O X X . . |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ S = shared point\ Score = 12 - 12 - 0 = 0.\ \ Black 1 is one possible perfect play.\ \ Conclusion:\ \ The effect of the implications of the 1-sided dame precedental ruling\ alter strategies and perfect play in this example.\ \ ************************************************** ************************\ \ 7) Passing before filling privilege dame / study if passing is\ prohibited when basic endgame kos are unfilled:\ \ [go]\ $$ -------------\ $$ | . X . X . |\ $$ | O X X . . |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ Variation 7.1:\ \ [go]\ $$ -------------\ $$ | 3 X . X . |\ $$ | O X X . . |\ $$ | 2 O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ 1, 4, 5 = pass\ Black makes the first pass.\ There is no need to consider resumption.\ Score = 13 - 12 - 0 = 1.\ \ White 2 is a strategic mistake, Variation 7.4 is better for White.\ \ Variation 7.2:\ \ [go]\ $$ -------------\ $$ | 2 X . X . |\ $$ | 6 X X 4 5 |\ $$ | 3 O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ --------------\ [/go]\ \ 1, 9, 10 = pass\ 7 @ 2, 8 @ 3\ Black makes the first pass.\ There is no need to consider resumption.\ \ [go]\ $$ -------------\ $$ | X X . X . |\ $$ | O X X . X |\ $$ | O O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ position at the first succession of two passes\ Score = 13 - 12 - 0 = 1.\ \ White 2 is a strategic mistake, Variation 7.4 is better for White.\ \ Variation 7.3:\ \ [go]\ $$ -------------\ $$ | 1 X . X . |\ $$ | O X X . . |\ $$ | 2 O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ 3, 4 = pass\ Black makes the first pass.\ There is no need to consider resumption.\ \ Score = 13 - 12 - 0 = 1.\ \ Note how this differs from the other rules variant! Because White is\ forced by the rule not to pass, he fills the ko. He may not gain an\ extra point by making the first pass before filling the ko.\ \ Black 1 is the only perfect play.\ \ Variation 7.4:\ \ [go]\ $$ -------------\ $$ | . X . X . |\ $$ | O X X . . |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ 1, 2 = pass\ Black makes the first pass.\ \ [go]\ $$ -------------\ $$ | c X . X . |\ $$ | a X X . . |\ $$ | a O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ From the 1-sided dame precedental(!) ruling, one has to imply that not\ taken privilege intersections are contestable points. Therefore C is a\ contestable point.\ \ Things are more difficult for the 'a' intersections. Whichever strategy\ Black tries, he cannot force White to abandon them. However, is each\ such strategy perfect play for both players? One can argue that\ perfect play is required for identifying contestable points. Proof:\ Because of the move-sequences of Variations 7.2 and 7.3 starting with\ move 2 and the trivial move-sequence starting with Black, each perfect\ play leads to the 'a' intersections controlled by White. Therefore they\ are not contestable points. Similarly, none of the points nearby the\ ko threat aji is a contestable point. Summary:\ \ [go]\ $$ -------------\ $$ | c X . X . |\ $$ | O X X . . |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ A resumption may not see a play on C because that would contest a\ point supposed to be shared.\ \ [go]\ $$ -------------\ $$ | s X . X . |\ $$ | O X X . . |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ S = shared point\ Score = 12 - 12 - 0 = 0.\ \ Black 1 is a strategic mistake.\ \ Conclusion:\ \ The change of the ko filling rule changes Variation 7.3 from 6.3. This\ has severe effects on which moves are strategically correct. In this\ example, perfect play strategy under \"prohibited passing when there\ are unfilled kos\" is simpler than under \"allowed passing when there\ are unfilled kos\" but the analysis and especially the contestable\ rules interpretation why this is so is equally difficult. Application\ of the contestable points rules and the 1-sided dame precedent is far\ beyond ordinary players' ability. They may be lucky that with\ unawareness under \"prohibited passing when there are unfilled kos\"\ they will find correct strategy accidentally!!! However, in other\ exceptional shapes (which have not been discovered yet) they do not\ have a realistic chance to find correct strategy by means of\ supposedly correct rules and precedents interpretation.\ \ ************************************************** ************************\ \ 8) 1-sided dame:\ \ [go]\ $$ -----------------\ $$ | . O O X . . . |\ $$ | X X O X . . . |\ $$ | X X O X X X X |\ $$ | . . O X O O O |\ $$ | O O O X O O O |\ $$ | X X X X O O . |\ $$ | . X . X O . O |\ $$ -----------------\ [/go]\ \ komi = 6.5\ \ Variation 8.1:\ \ [go]\ $$ -----------------\ $$ | 1 O O X . . . |\ $$ | X X O X . . . |\ $$ | X X O X X X X |\ $$ | . . O X O O O |\ $$ | O O O X O O O |\ $$ | X X X X O O . |\ $$ | . X . X O . O |\ $$ -----------------\ [/go]\ \ 2, 3 = pass.\ White makes the first pass.\ There is no need to consider resumption.\ Score = (19+8) - (18+2) - 6.5 - 1 = 27 - 20 - 6.5 - 1 = -0.5.\ \ Variation 8.2:\ \ [go]\ $$ -----------------\ $$ | . O O X . . . |\ $$ | X X O X . . . |\ $$ | X X O X X X X |\ $$ | . . O X O O O |\ $$ | O O O X O O O |\ $$ | X X X X O O . |\ $$ | . X . X O . O |\ $$ -----------------\ [/go]\ \ 1, 2 = pass.\ Black makes the first pass.\ \ [go]\ $$ -----------------\ $$ | c O O X . . . |\ $$ | X X O X . . . |\ $$ | X X O X X X X |\ $$ | c c O X O O O |\ $$ | O O O X O O O |\ $$ | X X X X O O . |\ $$ | . X . X O . O |\ $$ -----------------\ [/go]\ \ C (or exactly one arbitrary point of all C points?) are contestable\ points, by precedental ruling. (Interpretation of the rules text of\ 2008-07-15 was ambiguous, so a precedental ruling has been necessary.)\ Therefore, during a resumption, play there is prohibited.\ \ [go]\ $$ -----------------\ $$ | s O O X . . . |\ $$ | X X O X . . . |\ $$ | X X O X X X X |\ $$ | s s O X O O O |\ $$ | O O O X O O O |\ $$ | X X X X O O . |\ $$ | . X . X O . O |\ $$ -----------------\ [/go]\ \ S = shared points\ Score = 26 - 20 - 6.5 - 0 = -0.5.\ \ Conclusion:\ \ Playing or not playing a 1-sided dame in this example are both\ strategically correct. In effect, the 1-sided dame here is unvaluable.\ \ ************************************************** ************************\ \ 9) 1-sided dame (colour-inverse position):\ \ [go]\ $$ -----------------\ $$ | . X X O . . . |\ $$ | O O X O . . . |\ $$ | O O X O O O O |\ $$ | . . X O X X X |\ $$ | X X X O X X X |\ $$ | O O O O X X . |\ $$ | . O . O X . X |\ $$ -----------------\ [/go]\ \ komi = 6.5\ \ Variation 9.1:\ \ [go]\ $$ -----------------\ $$ | 2 X X O . . . |\ $$ | O O X O . . . |\ $$ | O O X O O O O |\ $$ | . . X O X X X |\ $$ | X X X O X X X |\ $$ | O O O O X X . |\ $$ | . O . O X . X |\ $$ -----------------\ [/go]\ \ 1, 3, 4 = pass.\ Black makes the first pass.\ Score = (18+8) - (19+2) - 6.5 - 0 = 26 - 21 - 6.5 - 0 = -1.5.\ \ White 2 is strategically correct.\ \ Variation 9.2:\ \ [go]\ $$ -----------------\ $$ | . X X O . . . |\ $$ | O O X O . . . |\ $$ | O O X O O O O |\ $$ | . . X O X X X |\ $$ | X X X O X X X |\ $$ | O O O O X X . |\ $$ | . O . O X . X |\ $$ -----------------\ [/go]\ \ 1, 2 = pass\ Black makes the first pass.\ \ [go]\ $$ -----------------\ $$ | c X X O . . . |\ $$ | O O X O . . . |\ $$ | O O X O O O O |\ $$ | c c X O X X X |\ $$ | X X X O X X X |\ $$ | O O O O X X . |\ $$ | . O . O X . X |\ $$ -----------------\ [/go]\ \ C (or exactly one arbitrary point of all C points?) are contestable\ points, by precedental ruling. (Interpretation of the rules text of\ 2008-07-15 was ambiguous, so a precedental ruling has been necessary.)\ Therefore, during a resumption, play there is prohibited.\ \ [go]\ $$ -----------------\ $$ | s X X O . . . |\ $$ | O O X O . . . |\ $$ | O O X O O O O |\ $$ | s s X O X X X |\ $$ | X X X O X X X |\ $$ | O O O O X X . |\ $$ | . O . O X . X |\ $$ -----------------\ [/go]\ \ S = shared points\ Score = 26 - 20 - 6.5 - 0 = -0.5.\ \ White 2 is a strategic mistake.\ \ Conclusion:\ \ In this example, the player who has 1-sided dame available can play\ there before the first succession of two passes. The opponent gets the\ first pass because it is his move. Contrarily in example (8), the\ player to move cannot take advantage of having the right to move\ because by doing so (taking the first pass) he loses his right to take\ a 1-sided dame. Hence the position and its colour-inverse behave\ strategically differently. This is the effect of the contestable\ points rules in combination with the precedental ruling that 1-sided\ dame are contestable points. It is not the effect of the first pass's\ value because under the Simplified World Mind Sports Games Rules the\ player to move can take the first pass and the player with the 1-sided\ dame privilege may take them, regardless of whether we have position\ (8) or its colour-inverse (9).\ \ Not surprisingly, possible exceptional rules for 1-sided dame are\ being discussed to get out of the bad consequences of the official\ contestable points rules.\ \ ************************************************** ************************\ \ Overall conclusion:\ \ Simplified World Mind Sports Games Rules or a certain interpretation\ of the World Mind Sports Games Rules of 2008-07-15 lead to (rather)\ simple rules application but unexpected strategic behaviour in\ exceptional positions where the first pass as tedomari is more urgent\ than endgame plays on the board with greater miai values. Although\ this is conceptually not different from middle game or earlier endgame\ tedomari, such strategic behaviour related to the first pass has been\ unknown under both Chinese Rules and Japanese/Korean Rules. Currently\ the Chinese and Japanese appear to be looking for a compromise that\ introduces exceptional, additional rules (or precedents, until further\ explicit, written exceptional, additional rules will be added) with\ the apparent aim to eliminate exceptional, previously unknown\ strategic behaviour in exceptional (but mostly rather common)\ positions. There is a problem though: Nobody knows whether or how many\ other shape classes with such exceptional, previously unknown\ strategic behaviour might exist. So there is the danger that an\ endless list of precedental rulings will have to be created and that\ the treatment is worse than what some perceive a desease. I recommend\ Simplified World Mind Sports Games Rules because understanding them\ and their strategic consequences is simpler than understanding a lot\ of difficult rules and their application. So far it seems that the\ Asian rules gurus (are going) fulfil their intended job to remove\ strategic difficulties by putting them into difficult rules instead;\ so the ordinary player might simply pray, not think, and expect\ everything to go well on the board. There is no guarantee though that\ what works for basic endgame kos will work in general. E.g., it is\ already difficult to distinguish basic endgame kos from dead kos or\ double ko sekis. Disputes will arise about whether having to fill such\ before making the first pass. Etc. This is a conceptually dangerous\ game, and rules history might see the troubles of endless precedents\ again as happened with the Japanese 1949 Rules. Why not do things\ right immediately by adopting the Simplified World Mind Sports Games\ Rules? We love go because of its strategic complexity - not because\ rules prohibit our thinking. Some strategy might be new - why not\ welcome it as an enrichment? Superko is also a strategic enrichment.\ Let us design our rules with courage rather than fear!" "mdobbins","","2008-09-14 22:25:58","Thanks, sol.ch for reformatting, that makes the post much easier to follow. \ \ \ For some of the other responses, I think some posters have been unfair to Robert. Robert may be interested in exploring details of rules that most tend to just gloss over, but that is no reason to be so negative about such an intellectual pursuit. It feels a little like the general populations' anti-intelectual attitude towards go players, only up one level or two. Robert's posts tend to be in appropriatly identified threads and a sre few in number that if you are not interested, it is easy to mark them as read and move on to something more interesting to you. \ \ I am not saying you should not have your opinion, and fully support your right to expresss your opinion. I just ask that you consider what value you are adding with your expressions. If you are just trying to belittle someone's hard work, or are trying to make them feel less, are you doing anything more than trying to bloat your own ego? How skillful is that? What value?" "RobertJasiek","","2008-09-14 23:04:54","Thx sol.ch!!! On a very quick check, I see one editing mistake already I made and here is the correction: \ \ [go] \ $$ ----------- \ $$ | b b b b | \ $$ | b b b b | \ $$ | s b s b | \ $$ | s w s w | \ $$ | w w w w | \ $$ | w w w w | \ $$ ----------- \ [/go]" "sol.ch","","2008-09-14 23:08:41","no problem; also, fix'd. i'll keep checking this thread." "Bill Spight","","2008-09-15 15:09:07","[QUOTE=TMark]Ithink that Robert is, or has become, the Large Hadron Collider of Western Go, searching for smaller and smaller particles of definition within the rules. Unfortunately, one word constantly misused by him is \"strategic\" which has no place within the ever diminishing minor nit-picking details of possible ko captures to the nth degree. \ \ Read them at your peril! Best wishes.[/QUOTE] \ \ One meaning of [I]strategy[/I] is a prescription for playing a game. True, in the positions Robert is talking about the games have few plays, but Robert discusses how to play them. So \"strategy\" and \"strategic\" are appropriate words here. :)" "xela","","2008-09-15 18:09:24","[QUOTE=Hicham]Hi Robert, could you try editing that to the diagram format used in this forum?[/QUOTE] \ \ I imagine that Robert has already spent many many hours on this article. I'm grateful for his contribution and don't expect him to put in extra time on formatting. \ \ Thanks to sol.ch for the nice diagrams!" "Bill Spight","","2008-09-15 19:08:09","[QUOTE=sol.ch]Robert Jasiek's post (reformatted to show diagrams):\ \ The following was written and formatted for rec.games.go. If somebody feels like it, feel free to provide nicer diagrams. I post it also here to give more WMSG participants a chance to develop good endgame strategy.\ \ **************************************************\ \ The following discussion relies on the probably or possibly used rules\ version on 2008-09-14. For the sake of simplicity, no komi is used in\ the examples (unless stated otherwise) and shared intersections are\ not counted at all. The player to move is given due to the numbers of\ black and white stones on the board. Recall that under World Mind\ Sports Games Rules White making the first pass reduces Black's score\ by 1.\ \ **************************************************\ \ 4) Exceptional basic endgame kos / study if passing is allowed when\ basic endgame kos are unfilled:\ \ (Exceptional not because of the ko threat but because an early first\ pass is interesting.)\ \ {snip}\ \ Variation 4.4:\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | 1 X O X 4 5 |\ $$ | X O 3 O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ 2 = pass, 6 takes ko, 7, 8 = pass\ White makes the first pass.\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | X X O X . X |\ $$ | X O . O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ position at the first succession of two passes\ There is no need to consider resumption: Superko prohibits recapture.\ \ [go]\ $$ ---------------\ $$ | . . . . . . |\ $$ | X X X X X X |\ $$ | X X O X . X |\ $$ | X O n O X X |\ $$ | O O O O O O |\ $$ | . O . O . . |\ $$ ---------------\ [/go]\ \ N is not a contestable point because each perfect play lets White\ control N.\ \ Score = 20 - 16 - 1 = 3.\ \ Black 1 is a strategic mistake:[/QUOTE]\ \ Thanks, sol! Seeing the diagrams really helps. :)\ \ The same sort of thing can happen under area scoring when there is a single unfilled dame and two simple kos that are miai, if the second player is komaster. The first pass gains 1/2 point under area scoring instead of 1 point, but the principle is the same.\ \ [QUOTE]\ **************************************************************\ \ 6) Passing before filling privilege dame / study if passing is allowed\ when basic endgame kos are unfilled:\ \ [go]\ $$ -------------\ $$ | . X . X . |\ $$ | O X X . . |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ Variation 6.1:\ \ [go]\ $$ -------------\ $$ | 3 X . X . |\ $$ | O X X . . |\ $$ | 2 O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ 1, 4, 5 = pass\ Black makes the first pass.\ There is no need to consider resumption.\ Score = 13 - 12 - 0 = 1.\ \ White 2 is a strategic mistake, Variation 6.4 is better for White.\ \ Variation 6.2:\ \ [go]\ $$ -------------\ $$ | 2 X . X . |\ $$ | W X X 4 5 |\ $$ | 3 O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ 1 = pass, 6 takes back, 9, 10 = pass\ 7 @ 2, 8 @ 3\ Black makes the first pass.\ There is no need to consider resumption.\ \ [go]\ $$ -------------\ $$ | X X . X . |\ $$ | O X X . X |\ $$ | O O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ position at the first succession of two passes\ Score = 13 - 12 - 0 = 1.\ \ White 2 is a strategic mistake, Variation 6.4 is better for White.\ \ Variation 6.3:\ \ [go]\ $$ -------------\ $$ | 1 X . X . |\ $$ | O X X 4 5 |\ $$ | 3 O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ 2 =pass, 6 takes ko, 7, 8 = pass\ White makes the first pass.\ \ [go]\ $$ -------------\ $$ | X X . X . |\ $$ | O X X . X |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ position at the first succession of two passes\ There is no need to consider resumption: Superko prohibits recapture.\ \ [go]\ $$ -------------\ $$ | X X . X . |\ $$ | O X X . X |\ $$ | n O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ N is not a contestable point because each perfect play lets White\ control N.\ \ Score = 13 - 12 - 1 = 0.\ \ Black 1 is one possible perfect play.\ \ Variation 6.4:\ \ [go]\ $$ -------------\ $$ | . X . X . |\ $$ | O X X . . |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ 1, 2 = pass\ Black makes the first pass.\ \ [go]\ $$ -------------\ $$ | c X . X . |\ $$ | a X X . . |\ $$ | a O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ From the 1-sided dame precedental(!) ruling, one has to imply that not\ taken privilege intersections are contestable points. Therefore C is a\ contestable point.\ \ Things are more difficult for the 'a' intersections. Whichever strategy\ Black tries, he cannot force White to abandon them. However, is each\ such strategy perfect play for both players? One can argue that\ perfect play is required for identifying contestable points. Proof:\ Because of the move-sequences of Variations 6.2 and 6.3 starting with\ move 2 and the trivial move-sequence starting with Black, each perfect\ play leads to the 'a' intersections controlled by White. Therefore they\ are not contestable points. Similarly, none of the points nearby the\ ko threat aji is a contestable point. Summary:\ \ [go]\ $$ -------------\ $$ | c X . X . |\ $$ | O X X . . |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ A resumption may not see a play on C because that would contest a\ point supposed to be shared.\ \ [go]\ $$ -------------\ $$ | s X . X . |\ $$ | O X X . . |\ $$ | . O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ S = shared point\ Score = 12 - 12 - 0 = 0.\ \ Black 1 is one possible perfect play.\ \ Conclusion:\ \ The effect of the implications of the 1-sided dame precedental ruling\ alter strategies and perfect play in this example. [/QUOTE]\ \ [go]\ $$ -------------\ $$ | a X . X . |\ $$ | O X X . . |\ $$ | b O O X X |\ $$ | O . O O X |\ $$ | . O . O X |\ $$ -------------\ [/go]\ \ Black should pass instead of playing at [I]a[/I]. If White now passes, I think that [I]a[/I] should belong to Black and [I]b[/I] should belong to White. They are miai. I do not know the rules well enough to say how they would actually be interpreted. However, note that if the first pass is not counted for the purpose of ending play, this kind of problem does not arise. After two passes Black can play at [I]a[/I] with no problem.\ \ [QUOTE]\ \ ************************************************** \ \ 8) 1-sided dame:\ \ [go]\ $$ -----------------\ $$ | . O O X . . . |\ $$ | X X O X . . . |\ $$ | X X O X X X X |\ $$ | . . O X O O O |\ $$ | O O O X O O O |\ $$ | X X X X O O . |\ $$ | . X . X O . O |\ $$ -----------------\ [/go]\ \ komi = 6.5\ \ Variation 8.1:\ \ [go]\ $$ -----------------\ $$ | 1 O O X . . . |\ $$ | X X O X . . . |\ $$ | X X O X X X X |\ $$ | . . O X O O O |\ $$ | O O O X O O O |\ $$ | X X X X O O . |\ $$ | . X . X O . O |\ $$ -----------------\ [/go]\ \ 2, 3 = pass.\ White makes the first pass.\ There is no need to consider resumption.\ Score = (19+8) - (18+2) - 6.5 - 1 = 27 - 20 - 6.5 - 1 = -0.5.\ \ Variation 8.2:\ \ [go]\ $$ -----------------\ $$ | . O O X . . . |\ $$ | X X O X . . . |\ $$ | X X O X X X X |\ $$ | . . O X O O O |\ $$ | O O O X O O O |\ $$ | X X X X O O . |\ $$ | . X . X O . O |\ $$ -----------------\ [/go]\ \ 1, 2 = pass.\ Black makes the first pass.\ \ [go]\ $$ -----------------\ $$ | c O O X . . . |\ $$ | X X O X . . . |\ $$ | X X O X X X X |\ $$ | c c O X O O O |\ $$ | O O O X O O O |\ $$ | X X X X O O . |\ $$ | . X . X O . O |\ $$ -----------------\ [/go]\ \ C (or exactly one arbitrary point of all C points?) are contestable\ points, by precedental ruling. (Interpretation of the rules text of\ 2008-07-15 was ambiguous, so a precedental ruling has been necessary.)\ Therefore, during a resumption, play there is prohibited.\ \ [go]\ $$ -----------------\ $$ | s O O X . . . |\ $$ | X X O X . . . |\ $$ | X X O X X X X |\ $$ | s s O X O O O |\ $$ | O O O X O O O |\ $$ | X X X X O O . |\ $$ | . X . X O . O |\ $$ -----------------\ [/go]\ \ S = shared points\ Score = 26 - 20 - 6.5 - 0 = -0.5.\ \ Conclusion:\ \ Playing or not playing a 1-sided dame in this example are both\ strategically correct. In effect, the 1-sided dame here is unvaluable. [/QUOTE]\ \ If the first pass does not count towards ending play, then after two passes Black can fill a one-sided dame and pick up one point.\ \ [QUOTE]\ **************************************************\ \ 9) 1-sided dame (colour-inverse position):\ \ [go]\ $$ -----------------\ $$ | . X X O . . . |\ $$ | O O X O . . . |\ $$ | O O X O O O O |\ $$ | . . X O X X X |\ $$ | X X X O X X X |\ $$ | O O O O X X . |\ $$ | . O . O X . X |\ $$ -----------------\ [/go]\ \ komi = 6.5\ \ Variation 9.1:\ \ [go]\ $$ -----------------\ $$ | 2 X X O . . . |\ $$ | O O X O . . . |\ $$ | O O X O O O O |\ $$ | . . X O X X X |\ $$ | X X X O X X X |\ $$ | O O O O X X . |\ $$ | . O . O X . X |\ $$ -----------------\ [/go]\ \ 1, 3, 4 = pass.\ Black makes the first pass.\ Score = (18+8) - (19+2) - 6.5 - 0 = 26 - 21 - 6.5 - 0 = -1.5.\ \ White 2 is strategically correct. [/QUOTE]\ \ Note that not counting the first pass towards ending play does not affect the result. White can gain a point for the one-sided dame. It seems fair to treat Black and White the same in this regard.\ \ If you think of Button Go, where there is a physical button that is worth 1/2 point to whoever takes it, it is clear that taking the button gains something, unlike a pass, which does not. Even if you do not use a button but make the first pass gain 1/2 point, it should also be clear that the first pass is a play in the game, not a proposal to end it. Therefore it should not be treated like later passes in regard to ending play." "RobertJasiek","","2008-09-16 00:46:29","Bill, could you explain the two kos and one dame a bit further to see the similarities or differences to WMSGR more clearly? \ \ Whether in the privilege dame position Black has a right to get the dame and White a right to get the connection may indeed be considered to be still an open problem of interpretation. \ \ Currently WMSGR are not like Ikeda Rules though, which let the first pass not be game ending. I wonder whether the WMSGR designers were aware of the consequences when dropping the aspect." "Harleqin","","2008-09-16 03:07:41","I think that the game should never end just because of a certain number of passes. Instead, two consecutive passes (the exact number is not even important) should lead to a game stop during which players may agree about what stones are to be removed and to end the game. If they don't, they resume play until during some later game stop they do agree. If between game stops no stone is played, but the players still don't agree, there is obviously some sort of arbitration needed. Any preset ruling in this case will likely enable strange results to happen.\ \ And what is the fuss about 'contestable points' about? When I read the WMSG rules for the first time, I had the impression that the intent was to create straightforward area rules with last move compensation, period. Can't this point (Section 5, last sentence) just be thrown out?" "RobertJasiek","","2008-09-16 06:09:34","We wish but reality differs." "Bill Spight","","2008-09-16 09:52:44","[QUOTE=RobertJasiek]Bill, could you explain the two kos and one dame a bit further to see the similarities or differences to WMSGR more clearly?[/QUOTE] \ \ Here is an example with area scoring without the \"first pass compensation\". \ \ [go] \ $$ ----------------- \ $$ | . . . . . . . . | \ $$ | X X X X . X X X | \ $$ | . X O X X X . . | \ $$ | X O . O . O X X | \ $$ | O O O O O O O O | \ $$ | . O . O . . . . | \ $$ ----------------- \ [/go] \ \ Area scoring, 5.5 komi, Black to play. \ \ [go]$$ Black error \ $$ ----------------- \ $$ | . . . . . . . . | \ $$ | X X X X . X X X | \ $$ | 1 X O X X X 4 5 | \ $$ | X O 3 O 2 O X X | \ $$ | O O O O O O O O | \ $$ | . O . O . . . . | \ $$ ----------------- \ [/go] \ \ W6 takes ko. \ \ Black has no ko threat. \ \ Result: B 26, W 22 + 5.5 komi. W wins by 1.5. \ \ [go]$$ Correct play \ $$ ----------------- \ $$ | . . . . . . . . | \ $$ | X X X X . X X X | \ $$ | 3 X O X X X . . | \ $$ | X O 2 O 1 O X X | \ $$ | O O O O O O O O | \ $$ | . O . O . . . . | \ $$ ----------------- \ [/go] \ \ Result: B 27, W 21 + 5.5 komi. B wins by 0.5. \ \ Since playing the dame gains one point, the score difference is 2 pts. between these lines of play. By comparison, when the first pass gains 1/2 point, the score difference is 1 pt. between the comparable lines of play. \ \ [go]$$W \ $$ ------------- \ $$ | . O . O . | \ $$ | X O O . . | \ $$ | . X X O O | \ $$ | X . X X O | \ $$ | . X . X . | \ $$ ------------- \ [/go] \ \ Altering one of your examples, here is another one from ordinary area scoring. White to play. \ \ [go]$$W White error \ $$ ------------- \ $$ | 1 O . O . | \ $$ | X O O 4 5 | \ $$ | 3 X X O O | \ $$ | X . X X O | \ $$ | . X . X 2 | \ $$ ------------- \ [/go] \ \ B 6 takes ko. \ \ Result: W 12, B 13. Black wins by 1 pt. \ \ [go]$$W Correct play \ $$ ------------- \ $$ | 3 O . O . | \ $$ | X O O . . | \ $$ | 2 X X O O | \ $$ | X . X X O | \ $$ | . X . X 1 | \ $$ ------------- \ [/go] \ \ Result: W 13, B 12. White wins by 1 pt. \ \ These kinds of positions have direct analogs in area scoring, and, I dare say, in territory scoring, as well (although more complex). They have to do with the idea of komonster. \ \ [QUOTE] \ Whether in the privilege dame position Black has a right to get the dame and White a right to get the connection may indeed be considered to be still an open problem of interpretation.[/QUOTE] \ \ I have downloaded the rules, and I think that unfilled dame are to be shared. \ \ [QUOTE]Currently WMSGR are not like Ikeda Rules though, which let the first pass not be game ending. I wonder whether the WMSGR designers were aware of the consequences when dropping the aspect.[/QUOTE] \ \ I gather from your posts that they are working on the kinds of problems you have shown. Please let them know that there is a simple answer that has been known for a long time and is theoretically correct. :) The first pass, so-called, is not really a pass. The rules say, \"Playing alternately is the right of both players, but either player may give up the right to make a play and this is known as a pass.\" The first \"pass\" does not give up the right to make a play. It is a play that gains 1/2 point." "RobertJasiek","","2008-09-16 10:25:56","Bill, thanks! \ \ Although you are going to repeat it, please - in WMSGR terminology - state explicitly the rules that have long been known as a solution and explain why they are better to solve those problems and the contestable points issues, (I guess you do not mean Simplified WMSGR but are referring to Ikeda A3 or some version of Spight Rules.) I might guess and figure it out myself but I don't have the time before the WMSG. (Hint of why I am so busy besides a 4 day tournament: I have just defined(!) 5 of 7(?) existing types of disturbing kos and am now busy with the attempt to classify types of fighting kos. One of the types will be basic endgame kos, but you'd be surprised to see how not so easy it is to distinguish them from middle game connection kos... So far my definitions are independent of scores, values, shapes, cycle behaviour, and - as modules - rules details! You will be surprised about an entirely new approach :) )" "Bill Spight","","2008-09-16 10:30:05","[QUOTE=RobertJasiek]Not surprisingly, possible exceptional rules for 1-sided dame are being discussed to get out of the bad consequences of the official \ contestable points rules.[/QUOTE] \ \ One sided dame gain nothing, they are like filling in territory. Therefore the proper time to play them is after the first \"pass\", which gains 1/2 point. Therefore the first pass should have no effect upon stopping play. As I have pointed out, by the definition of \"pass\" in the rules, it is not really a pass. \ \ [QUOTE]There is a problem though: Nobody knows whether or how many other shape classes with such exceptional, previously unknown strategic behaviour might exist.[/QUOTE] \ \ Well, if you treat the first \"pass\" as a play instead of a pass, it simplifies matters considerably. It is true that nobody knows what moves that are worth less than a dame exist. I have discovered two, and others have discovered two more. But as for the examples that you have shown, the problems arise from allowing the opponent of the first passer to stop play with a pass. \ \ [QUOTE]So there is the danger that an endless list of precedental rulings will have to be created and that the treatment is worse than what some perceive a desease.[/QUOTE] \ \ Indeed. Special rulings are unnecessary, and, as the Japanese experience has shown, undesirable." "Bill Spight","","2008-09-16 10:56:45","[QUOTE=RobertJasiek]Bill, thanks! \ \ Although you are going to repeat it, please - in WMSGR terminology - state explicitly the rules that have long been known as a solution[/QUOTE] \ \ Ikeda Area III, Rule 6 \ End of the game: After the first pass, the game ends when both players pass in succession. \ \ Also see [url]http://senseis.xmp.net/?ButtonGo[/url]. WMSG rules are a form of button go, as are Ikeda Area rules, only without a physical button. \ \ It is not that these rules have been known as a solution to the problems you have shown us, the problems do not even arise in button go, because button go simply adds a move worth 1/2 point to area scoring. Nothing essential is changed. Having a 1/2 point play typically makes the question of who gets the last dame irrelevant, and thus makes the result more like territory scoring. So button go is a good compromise between the two major forms of scoring today. \ \ \ [QUOTE]and explain why they are better to solve those problems and the contestable points issues,[/QUOTE] \ \ The contestable points issue will not go away. It may punish a player for not making gainful plays before passing. However, without a rule such as the Ikeda rule above, the current rules may also punish a player for making the theoretically correct play (the first pass),-- in the case of one sided dame, when it is the biggest play, and in other cases, when it is the correct play to get tedomari." "Harleqin","","2008-09-16 12:33:37","In theory, if one player firstpasses and the other passes immediately afterwards, there exists the mechanism of section 5, number 2. However, this means that everything would have to be played out, for example in the case of one-sided dame, and no removal by agreement would be possible.\ \ Currently, I have the impression that someone is intentionally screwing up the rule wording in order to increase frustration with this kind of rules among the players." "Bill Spight","","2008-09-16 13:17:05","[QUOTE=Harleqin]In theory, if one player firstpasses and the other passes immediately afterwards, there exists the mechanism of section 5, number 2. However, this means that everything would have to be played out, for example in the case of one-sided dame, and no removal by agreement would be possible.\ \ Currently, I have the impression that someone is intentionally screwing up the rule wording in order to increase frustration with this kind of rules among the players.[/QUOTE]\ \ I don't think so. I think that the \"contestable\" points questions comes from concerns of Japanese and Korean players. First, it has been customary to leave dame unfilled under territory rules (although that seems to have changed recently in Japanese pro play). Sharing unfilled dame at least sounds like an accommodation until you think about the interaction with the \"first pass\" rule. More important, I think, is the \"both lose\" provision in the Japanese rules, that punishes the presumptive winner for overlooking a gainful play. Sharing contestable points also metes out a punishment for failing to make a gainful play, although it is typically less severe than prescribing a loss.\ \ The problem stems from rule 5.3, which states:\ \ [QUOTE]Play may only be resumed to confirm the life or death of stones and the rights to the vacant points enclosed by the living stones. If one or both players find other contestable points on the board after two passes, those points shall be left as they stand and are evenly divided between the two players.[/QUOTE]\ \ Neither one sided dame nor sente dame are vacant points enclosed by living stones (of one color), and so presumably must be played before the first pass to count. But both types of dame are worth less than the first pass. Therein lies the problem." "RobertJasiek","","2008-09-16 14:35:41","Bill, things are getting clearer! I don't know though when the Japanese / Koreans can be convinced. They dislike virtual plays and also dislike plays of value after one and / or two passes (and have had rules for that). For some reason, it is very difficult to move them exactly where we have understood the best solution lies. Maybe excellent explanations by us can help to overcome their aforementioned feelings. The Chinese are willing to compromise to get the Japanese / Koreans in at all. We are observing this, know the theory a bit better, but will have a hard time to encourage immediate perfection. As Thomas H. puts it: After the WMSG 2008, there will be more time for improvements. \ \ My first impression is that the following are essentially the same: \ Ikeda A3 = Simplified WMSGR with resumption on request = WMSGR with Ikeda pass succession rule and without contestable points rule = Button Go without extra effects of ko bans. Or am I overlooking something? \ \ How do you conclude that the 1/2 pass is not really a pass in the WMSG text? \ \ It would be nice to see a list of the 4 plays smaller than a dame." "Bill Spight","","2008-09-16 16:18:59","[QUOTE=RobertJasiek]Bill, things are getting clearer! I don't know though when the Japanese / Koreans can be convinced. They dislike virtual plays and also dislike plays of value after one and / or two passes (and have had rules for that).[/QUOTE]\ \ I have given that last a little bit of thought. Like not calling the first \"pass\" a pass, but something else. How about \"knock\"? Or \"check\"? Our experiences with coupon go show that players easily adapt to taking a card or piece of paper. Doing so rather than saying \"Pass\" might help make the distinction clear.\ \ [QUOTE]For some reason, it is very difficult to move them exactly where we have understood the best solution lies. Maybe excellent explanations by us can help to overcome their aforementioned feelings.[/QUOTE]\ \ I think that the best explanation is that the first \"pass\" is not really a pass.\ \ [QUOTE]The Chinese are willing to compromise to get the Japanese / Koreans in at all. We are observing this, know the theory a bit better, but will have a hard time to encourage immediate perfection. As Thomas H. puts it: After the WMSG 2008, there will be more time for improvements.[/QUOTE]\ \ Indeed there will. :) As you know, I have predicted that the future of go lies in button go. I was pleasantly surprised that the WMSG are already a form of button go. That is progress. :)\ \ [QUOTE]How do you conclude that the 1/2 pass is not really a pass in the WMSG text?[/QUOTE]\ \ The rules define pass this way: \"Either player may give up the right to make a play and this is known as a pass.\" \ \ But taking an action to gain 1/2 point is not giving up the right to make a play. It is itself a play. Much better not to call it a pass at all. \ \ ----\ \ Edit: The rules also say, \"Two consecutive passes signal that both players believe that there are no more points to contest for in the game.\" But if the first pass is made for the purpose of gaining 1/2 point, then it does not give that message. The examples that you give underscore that fact. In territory scoring you do not say, \"Filling the last half point ko signals that there are no more points to contest.\" It is the fact that a pass gains nothing that allows the interpretation that the players do not believe that there are any more points to contest. (And even that is not necessarily so. ;)) Making a play that gains something does not convey any such message.\ \ ----\ \ The more I think about it, the more I like \"check\" in English, especially if you pick up a slip of paper. You can call the paper a check that is worth 1/2 point to whoever takes it.\ \ [QUOTE]It would be nice to see a list of the 4 plays smaller than a dame.[/QUOTE]\ \ The first one I know of was discovered by Bill Fraser. It is a 2/3 pt. sente by area scoring. Then I discovered a 1/3 pt. play by area scoring, and then a 1/2 pt. play. Later Bernd discovered another 1/2 pt. play." "RobertJasiek","","2008-09-16 22:27:54","The Asians do not like physical means to express a button and they also do not like another move type. So I think the easiest way to convince them is either \"a play not made on the board\" or \"a pass not contributing to the game end and made before regular passes\". \ \ Which are example positions for \"Bill Fraser. It is a 2/3 pt. sente by area scoring. Then I discovered a 1/3 pt. play by area scoring, and then a 1/2 pt. play. Later Bernd discovered another 1/2 pt. play.\"?" "flOvermind","","2008-09-17 01:41:06","[QUOTE=Bill Spight]I have given that last a little bit of thought. Like not calling the first \"pass\" a pass, but something else. How about \"knock\"? Or \"check\"? Our experiences with coupon go show that players easily adapt to taking a card or piece of paper. Doing so rather than saying \"Pass\" might help make the distinction clear.[/QUOTE]\ \ [QUOTE=RobertJasiek]The Asians do not like physical means to express a button and they also do not like another move type. So I think the easiest way to convince them is either \"a play not made on the board\" or \"a pass not contributing to the game end and made before regular passes\".[/QUOTE]\ \ Wouldn't it be sufficient to say that three consecutive passes are needed to end the game? The first of these three would be worth a point. Now if the second player passes, too, the first player can still play a contestable point, shared dame, sente dame, whatever...\ \ That way you don't need a \"new\" move type, and it would archive the same result." "RobertJasiek","","2008-09-17 06:10:00","A succession of three passes is a solution, too, but somewhat of an overkill because it might be \ \ pass \ play \ ... \ play \ pass \ pass \ pass \ \ Unless two more passes succeed the first pass directly, a succession of three passes could be meaningful to lift a ko ban, however, passing does not lift ko bans under positional superko." "RobertJasiek","","2008-09-17 06:12:48","Here are two more rec.games.go articles: \ \ ************************************************************************* \ ************************************************************************* \ \ Bill Spight, who has been familiar with \"button go\" for years, has suggested how to solve the first pass problem. The contestable points rules may be viewed as an extra problem that can be treated separately and later by either abandoning it or by adding more rules. However, the conceptually more fundamental and therefore more urgent problem is the exact handling of the first pass. \ \ Currently the problems are caused by the fact that the \"first pass\" is both a) a pass that may contribute to the game stopping succession of passes and b) a play [besides the board] of value. Bill suggests to solve this by separating both from each other! \ \ Ikeda Area 3 Rules achieve the separation by the rule \"After the first pass, the game ends with a succession of two passes.\". The wording is a bit ambiguous and could be stated more clearly (in English) as: \"The first pass is a move of value [either 0 if Black makes it or -1 if White makes it]. After the first pass, each pass is unvaluable. Two unvaluable passes in succession stop / end the alternation.\" \ \ Button go rules achieve that by introducing a new term for the special move and move type: \"1/2 point play not on the board\", \"taking the button\", \"taking the token\", \"card\", \"check\", or whatever. Since we do not want any physical object, this is not so helpful, except as a conceptual background information. \ \ There are two types of moves: valuable moves and unvaluable moves. There are two types of passes: contributing to game ending or not contributing to game ending. There are two places: on the board or not on the board. Altogether we need three types: \ \ - valuable moves on the board not contributing to game ending (= \"plays\") \ - valuable moves not on the board not contributing to game ending (= [currently called \"first pass\" but might also be called \"special play\" or whatever] \ - unvaluable moves not on the board contributing to game ending (= \"pass\") \ \ ************************************************************************* \ \ Area Scoring modified by a not game ending first pass leads to these strategic novelties compared to Area Scoring: \ - dame become unvaluable on the strategic level \ - the \"first pass\" can be like a play that is tedomari \ - there are (so far known) four types of extremely rare positions with plays of a miai value 1/3 or 1/2 that one would want to take only after the \"first pass\" (but currently this is for rules experts only; even I do not recall those positions by memory; so we can safely ignore these tiny plays for now) \ \ ************************************************************************* \ \ Below, for the sake of clarity and simplicity, I call the special move \"button\". \ \ Example 1: \ \ . . . . . . \ # # # # # # \ . # O # . . \ # O . O # # \ O O O O O O \ . O . O . . \ \ Black to move \ \ Variation 1: (correct) \ \ . . . . . . \ # # # # # # \ 3 # O # . . \ # O 2 O # # \ O O O O O O \ . O . O . . \ \ Black 1 = button \ 4, 5 = pass \ score = 20 - 16 - 0 = 4 \ \ Variation 2: (failure) \ \ . . . . . . \ # # # # # # \ 1 # 6 # 4 5 \ # O 3 O # # \ O O O O O O \ . O . O . . \ \ White 2 = button \ 7, 8 = pass \ score = 20 - 16 - 1 = 3 \ Black 1 is a strategic mistake: It fails to take the tedomari first. \ \ Variation 3: (failure) \ \ . . . . . . \ # # # # # # \ 1 # O # . . \ # O 2 O # # \ O O O O O O \ . O . O . . \ \ Black 3 = button \ 4, 5 = pass \ score = 20 - 16 - 0 = 4 \ Black 1 is a strategic mistake: It fails to take the tedomari first. \ White 2 is a strategic mistake: It fails to take advantage of being ko monster [informally: having \"more\" ko threats]. \ \ Conclusion: \ \ If there are equal endgame plays, then then button is tedomari. If the endgame plays are related to a ko monster advantage, then the tedomari needs to be taken first. \ \ ************************************************************************* \ \ Example 2: \ \ . # . # . \ O # # . . \ . O O # # \ O . O O # \ . O . O # \ \ Black to move \ \ Variation 1: (failure) \ \ 1 # . # . \ O # # . . \ 2 O O # # \ O . O O # \ . O . O # \ \ Black 3 = button \ 4, 5 = pass \ score = 13 - 12 - 0 = 1 \ Black 1 is a strategic mistake: It fails to take the tedomari first. \ White 2 is a strategic mistake: It fails to take advantage of being ko monster. \ \ Variation 2: (failure) \ \ 1 # . # . \ 6 # # 4 5 \ 3 O O # # \ O . O O # \ . O . O # \ \ White 2 = button \ 7, 8 = pass \ score = 13 - 12 - 1 = 0 \ Black 1 is a strategic mistake: It fails to take the tedomari first. \ \ Variation 3: (correct) \ \ 3 # . # . \ O # # . . \ 2 O O # # \ O . O O # \ . O . O # \ \ Black 1 = button \ 4, 5 = pass \ score = 13 - 12 - 0 = 1 \ \ Variation 4: (correct) \ \ 2 # . # . \ 6 # # 4 5 \ 3 O O # # \ O . O O # \ . O . O # \ \ Black 1 = button \ 7 at 2, 8 at 3 \ 9, 10 = pass \ score = 13 - 12 - 0 = 1 \ \ Conclusion: \ \ If there is a privilege dame, then then button is tedomari. If the privilege dame is related to a ko monster advantage, then the tedomari needs to be taken first. \ \ ************************************************************************* \ \ Example 3: \ \ . O O # . . . \ # # O # . . . \ # # O # # # # \ . . O # O O O \ O O O # O O O \ # # # # O O . \ . # . # O . O \ \ Black to move \ \ Variation 1: (failure) \ \ 1 O O # . . . \ # # O # . . . \ # # O # # # # \ . . O # O O O \ O O O # O O O \ # # # # O O . \ . # . # O . O \ \ White 2 = button \ 3, 4 = pass \ score = 27 - 20 - 1 = 6 \ Black 1 is a strategic mistake: It fails to take the tedomari first. \ \ Variation 2: (correct) \ \ 3 O O # . . . \ # # O # . . . \ # # O # # # # \ . . O # O O O \ O O O # O O O \ # # # # O O . \ . # . # O . O \ \ Black 1 = button \ 2, 4, 5 = pass \ score = 27 - 20 - 0 = 7 \ \ Conclusion: \ \ If there is a 1-sided dame, then then button is tedomari. \ \ ************************************************************************* \ \ Example 4: \ \ . # # O . . . \ O O # O . . . \ O O # O O O O \ . . # O # # # \ # # # O # # # \ O O O O # # . \ . O . O # . # \ \ Black to move \ \ Variation 1: \ \ 2 # # O . . . \ O O # O . . . \ O O # O O O O \ . . # O # # # \ # # # O # # # \ O O O O # # . \ . O . O # . # \ \ Black 1 = button \ 3, 4 = pass \ score = 20 - 27 - 0 = -7 \ \ Conclusion: \ \ Regardless of colour-inversion, all the moves of value (contestable points) are taken before the first succession of passes. (Examples 3 + 4.) \ \ ************************************************************************* \ \ Bill shows similar examples for Area Scoring to underline that also there similar tedomaris occur: \ \ >> \ Here is an example with area scoring without the \"first pass compensation\". \ \ \ Show Diagram Code[go] \ $$ ----------------- \ $$ | . . . . . . . . | \ $$ | X X X X . X X X | \ $$ | . X O X X X . . | \ $$ | X O . O . O X X | \ $$ | O O O O O O O O | \ $$ | . O . O . . . . | \ $$ ----------------- \ [/go]Hide Diagram Code \ Area scoring, 5.5 komi, Black to play. \ \ \ Black error \ Show Diagram Code[go]$$ Black error \ $$ ----------------- \ $$ | . . . . . . . . | \ $$ | X X X X . X X X | \ $$ | 1 X O X X X 4 5 | \ $$ | X O 3 O 2 O X X | \ $$ | O O O O O O O O | \ $$ | . O . O . . . . | \ $$ ----------------- \ [/go]Hide Diagram Code \ W6 takes ko. \ \ Black has no ko threat. \ \ Result: B 26, W 22 + 5.5 komi. W wins by 1.5. \ \ \ Correct play \ Show Diagram Code[go]$$ Correct play \ $$ ----------------- \ $$ | . . . . . . . . | \ $$ | X X X X . X X X | \ $$ | 3 X O X X X . . | \ $$ | X O 2 O 1 O X X | \ $$ | O O O O O O O O | \ $$ | . O . O . . . . | \ $$ ----------------- \ [/go]Hide Diagram Code \ Result: B 27, W 21 + 5.5 komi. B wins by 0.5. \ \ Since playing the dame gains one point, the score difference is 2 pts. between these lines of play. By comparison, when the first pass gains 1/2 point, the score difference is 1 pt. between the comparable lines of play. \ \ \ Show Diagram Code[go]$$W \ $$ ------------- \ $$ | . O . O . | \ $$ | X O O . . | \ $$ | . X X O O | \ $$ | X . X X O | \ $$ | . X . X . | \ $$ ------------- \ [/go]Hide Diagram Code \ Altering one of your examples, here is another one from ordinary area scoring. White to play. \ \ \ White error \ Show Diagram Code[go]$$W White error \ $$ ------------- \ $$ | 1 O . O . | \ $$ | X O O 4 5 | \ $$ | 3 X X O O | \ $$ | X . X X O | \ $$ | . X . X 2 | \ $$ ------------- \ [/go]Hide Diagram Code \ B 6 takes ko. \ \ Result: W 12, B 13. Black wins by 1 pt. \ \ \ Correct play \ Show Diagram Code[go]$$W Correct play \ $$ ------------- \ $$ | 3 O . O . | \ $$ | X O O . . | \ $$ | 2 X X O O | \ $$ | X . X X O | \ $$ | . X . X 1 | \ $$ ------------- \ [/go]Hide Diagram Code \ Result: W 13, B 12. White wins by 1 pt. \ \ These kinds of positions have direct analogs in area scoring, and, I dare say, in territory scoring, as well (although more complex). They have to do with the idea of komonster. \ << \ \ Therefore he assesses that the \"first pass\" should rather be considered a \"play\" that - only in the combinatorial sense - gains 1/2 point [miai value]. He continues: \ \ >> \ One sided dame gain nothing, they are like filling in territory. Therefore the proper time to play them is after the first \"pass\", which gains 1/2 point. Therefore the first pass should have no effect upon stopping play. [...] if you treat the first \"pass\" as a play instead of a pass, it simplifies matters considerably. [...] It is the fact that a pass gains nothing that allows the interpretation that the players do not believe that there are any more points to contest. [...] Making a play that gains something does not convey any such message. \ << \ \ ************************************************************************* \ \ Summarizing, the - what is currently called - \"first pass\" should not(!) contribute to ending alternation. It should be allowed to behave like every other tedomari without exception. \ \ ************************************************************************* \ ************************************************************************* \ \ The separation of \"button\" (aka \"first pass\") from game stopping passes also needs to be considered in the context of the currently - by the Asians - planned exceptional (now by me implicitly rewritten) rule \"One may not take the button or pass when there is a basic endgame ko on the board.\". For the sake of clarity and simplicity, first let me ignore the extra contestable points rules. \ \ ************************************************************************* \ \ Example 1: \ \ . . . . . . \ # # # # # # \ . # O # . . \ # O . O # # \ O O O O O O \ . O . O . . \ \ Black to move \ \ Variation 1: (correct) \ \ . . . . . . \ # # # # # # \ 1 # O # . . \ # O 2 O # # \ O O O O O O \ . O . O . . \ \ Black 3 = button \ 4, 5 = pass \ score = 20 - 16 - 0 = 4 \ \ Conclusion: \ \ The exceptional rule forces ko connections and thereby prohibits the button to behave as tedomari. The ko monster advantage is made immaterial. \ \ ************************************************************************* \ \ Example 2: \ \ . # . # . \ O # # . . \ . O O # # \ O . O O # \ . O . O # \ \ Black to move \ \ Variation 1: (correct) \ \ 1 # . # . \ O # # . . \ 2 O O # # \ O . O O # \ . O . O # \ \ Black 3 = button \ 4, 5 = pass \ score = 13 - 12 - 0 = 1 \ \ Variation 2: (correct) \ \ 3 # . # . \ O # # . . \ 2 O O # # \ O . O O # \ . O . O # \ \ Black 1 = button \ 4, 5 = pass \ score = 13 - 12 - 0 = 1 \ \ Variation 3: (correct) \ \ 2 # . # . \ 6 # # 4 5 \ 3 O O # # \ O . O O # \ . O . O # \ \ Black 1 = button \ 7 at 2, 8 at 3 \ 9, 10 = pass \ score = 13 - 12 - 0 = 1 \ \ Conclusion: \ \ If there is a privilege dame, then the button is tedomari. The exceptional rule does not apply to hidden kos. The ko monster advantage is made immaterial for the hidden basic ko but not for the long cycle ko. The simplest sequence of plays on the board is also correct play, so one may excuse the inconsistent consequences of the exceptional rule, which demands beasic endgame ko filling but not two-sided dame filling. Whether this can be excused only in this example or in all positions deserves study. Because of the inconsistency, the exceptional rule feels awkward. \ \ ************************************************************************* \ \ Example 3: \ \ . O O # . . . \ # # O # . . . \ # # O # # # # \ . . O # O O O \ O O O # O O O \ # # # # O O . \ . # . # O . O \ \ Black to move \ \ Variation 1: (failure) \ \ 1 O O # . . . \ # # O # . . . \ # # O # # # # \ . . O # O O O \ O O O # O O O \ # # # # O O . \ . # . # O . O \ \ White 2 = button \ 3, 4 = pass \ score = 27 - 20 - 1 = 6 \ Black 1 is a strategic mistake: It fails to take the tedomari first. \ \ Variation 2: (correct) \ \ 3 O O # . . . \ # # O # . . . \ # # O # # # # \ . . O # O O O \ O O O # O O O \ # # # # O O . \ . # . # O . O \ \ Black 1 = button \ 2, 4, 5 = pass \ score = 27 - 20 - 0 = 7 \ \ Conclusion: \ \ If there is a 1-sided dame, then then button is tedomari. The exceptional rule does not affect ordinary 1-sided dame. \ \ ************************************************************************* \ \ Example 4: \ \ . # # O . . . \ O O # O . . . \ O O # O O O O \ . . # O # # # \ # # # O # # # \ O O O O # # . \ . O . O # . # \ \ Black to move \ \ Variation 1: \ \ 2 # # O . . . \ O O # O . . . \ O O # O O O O \ . . # O # # # \ # # # O # # # \ O O O O # # . \ . O . O # . # \ \ Black 1 = button \ 3, 4 = pass \ score = 20 - 27 - 0 = -7 \ \ Conclusion: \ \ The exceptional rule does not affect ordinary 1-sided dame. \ \ ************************************************************************* \ \ If the players play all ordinary endgames, fill all two-sided dame, fill all one-sided dame, and since they have to fill all basic endgame kos before the first succession of two game stopping passes, then it appears that the contestable points rules are entirely superfluous, i.e., do not have any significant strategic impact. So, just to please the Japanese / Koreans, the contestable points rules might be kept and would simply mean that players are encouraged to complete the endgame before the first succession of two game stopping passes. \ \ ************************************************************************* \ \ Although the exceptional rule may be criticised for being awkward and in principle unnecessary, its usage together with the separation of button (\"first pass\") from game stopping passes in practice simplifies the very late endgame by avoiding the unexpected types of tedomari strategy uncommon under Area Scoring or Japanese / Korean Territory Scoring. This effect is achieved while the \"first pass\" functions well and privilege dame and 1-sided dame can be played before the first succession of game stopping passes regardless of whose initial turn it is. So, as both a compromise and sufficiently reasonable rules design, this combination of rules is possible: \ \ \" \ 1. If White makes the game's first pass, then 1 point is subtracted from Black's score. \ 2. After the game's first pass, the succession of two other passes stops alternate moving. \ 3. One may not make a pass when there is a basic endgame ko on the board. \ [4. Rules about contestable points are optionally possible.] \ \" \ \ ************************************************************************* \ \ Rules theorists may worry about a formal definition of \"basic endgame ko\" (I am close to finding one, but it is too complicated to become part of an applied ruleset.), plays smaller than or of equal size as the first pass's miai value, and - theoretical danger only - ko capture followed by filling one's eye when all one has is two-eye-formations. But for practical purposes, this might be tolerated. It should be pointed out though whether a ko with nearby unfilled teire is to be considered a basic endgame ko. Defining \"basic endgame ko\" is general is easier otherwise for the theoretician. But, for the purpose of the mandatory ko filling rule, it must be clarified that filling also such a ko is required." "Bill Spight","","2008-09-17 07:17:09","[QUOTE=flOvermind]Wouldn't it be sufficient to say that three consecutive passes are needed to end the game? The first of these three would be worth a point. Now if the second player passes, too, the first player can still play a contestable point, shared dame, sente dame, whatever...\ \ That way you don't need a \"new\" move type, and it would archive the same result.[/QUOTE]\ \ One of the first modern proposals for written rules, by Yasunaga Hajime, included a three pass rule, and the original Spight rules usually ended play with three passes. The main reason in both cases was that passes lifted ko bans. However, the three pass rule does not seem to have caught on. ;)\ \ Something that I expect will also solve these problems is Ing's four pass rule, which is actually a succession of two passes. Ing allows resumption after the first two passes. It has the advantage that many pros and amateurs have played under Ing rules, and so that procedure is not so strange. :)\ \ But as for needing a new move type, the first pass is a new move type. Why obscure that fact?" "Harleqin","","2008-09-17 09:25:55","Well, the firstpass is different from the other passes, but the firstpass is rather a subtype of passing: it is impossible to \"just pass\" instead of firstpassing when the firstpass has not occured yet, and it is impossible to firstpass afterwards." "Harleqin","","2008-09-17 09:45:07","[QUOTE=Bill Spight] \ Something that I expect will also solve these problems is Ing's four pass rule, which is actually a succession of two passes. Ing allows resumption after the first two passes. It has the advantage that many pros and amateurs have played under Ing rules, and so that procedure is not so strange. :) \ [/QUOTE] \ \ The four pass rule is not understood at all by the vast majority of players. You can ask Robert about that... \ \ The four passes are strange in that the second pair of passes occurs at a time where board plays cannot occur anyway, namely after dead stone removals." "RobertJasiek","","2008-09-17 10:11:45","Harleqin, one view it as a subtype of pass because it is not a play on the board but a virtual move, as a subtype of play because it is related to gaining points, or as an entirely new type of move because it is neither a play on the board nor a move contributing to ending the move-sequence. Personally I would prefer the latter. \ \ If one wanted to realize it my means of playing material, then the clearest way would be a playing card with this text: \"The owner of this card receives 0 extra points if he is Black or 1 extra point if he is White.\" However, most go players prefer realization in the form of a virtual move. Since in Chinese (IIRC) pass is described as \"virtual play\", we can hardly expect them to invent yet another different Chinese term..." "daniel_the_smith","","2008-09-17 10:12:14","Personally, I like the AGA's \"pass stones, white must pass last\" rule. *ducks*" "flOvermind","","2008-09-17 10:12:20","[QUOTE=Bill Spight]But as for needing a new move type, the first pass is a new move type. Why obscure that fact?[/QUOTE] \ \ Because... \ \ [QUOTE=RobertJasiek]They dislike virtual plays and also dislike plays of value after one and / or two passes (and have had rules for that).[/QUOTE] \ \ I'm well aware that three consecutive passes that can't serve as ko threats is nothing else than obscuring the fact that the first pass is special. That was the intention of the proposal ;)" "Bill Spight","","2008-09-17 10:32:26","[QUOTE=Harleqin]Well, the firstpass is different from the other passes, but the firstpass is rather a subtype of passing:[/QUOTE] \ \ Not if pass is defined as relinquishing the right to play." "Bill Spight","","2008-09-17 10:37:40","[QUOTE=Harleqin]The four pass rule is not understood at all by the vast majority of players. You can ask Robert about that...[/QUOTE] \ \ Nobody understands Ing rules. :( \ \ \ [QUOTE]The four passes are strange in that the second pair of passes occurs at a time where board plays cannot occur anyway, namely after dead stone removals.[/QUOTE] \ \ That is not my, ahem, [I]understanding[/I] of the four pass rule. ;)" "Bill Spight","","2008-09-17 10:44:37","[QUOTE=flOvermind]I'm well aware that three consecutive passes that can't serve as ko threats is nothing else than obscuring the fact that the first pass is special. That was the intention of the proposal ;)[/QUOTE]\ \ Over the years I have observed that when people wish to be clever with go rules, they gravitate towards the pass, overloading its meaning and sowing confusion. :( This typically leads to unintended consequences, which are sometimes defended on the basis of simplicity, as though they were not trying to be clever. ;)\ \ (I confess to trying to be clever myself, at first.)" "Bill Spight","","2008-09-17 11:08:36","[QUOTE=RobertJasiek]The Asians do not like physical means to express a button and they also do not like another move type.[/QUOTE] \ \ I think that you are talking about specific Asians, the ones who are taking part in crafting the WMSG rules. Asian players are another matter. :) \ \ I thought of calling the play a knock, because people could indicate it by knocking on the table or by saying \"Knock,\" or \"I knock.\" There is no physical button, but there is a new kind of play. It is this new kind of play that produces a synthesis of area and territory scoring." "flOvermind","","2008-09-17 11:34:43","[QUOTE=Bill Spight]Over the years I have observed that when people wish to be clever with go rules, they gravitate towards the pass, overloading its meaning and sowing confusion. :( This typically leads to unintended consequences, which are sometimes defended on the basis of simplicity, as though they were not trying to be clever. ;)\ \ (I confess to trying to be clever myself, at first.)[/QUOTE]\ \ I'm not trying to be clever about go rules, I'm trying to be clever about \"selling\" go rules. The point is getting exactly what you want without the people opposing it realizing you got it :P\ \ Personally I like your \"button go\". The problem with it is that people will always tend to see it as something artificial. They are opposed to it not because they are against the rule specifically, but because they are against the new concept.\ \ From previous discussion on this forum, I got the impression there is no objective reason, there just seems to be an opposition to new concepts in general. Therefore you can't solve this issue on an objective level. You have to be \"clever\" in formulating the rule so there is nothing \"new\".\ \ Let's make a thought experiment:\ 1) Let's make a new kind of move and call it \"knock\" and it's worth 1/2 point. I'm sure many people would be against it, just because there is a new move that was not there before.\ 2) Now just say the first pass gets a value, suddenly there is much less opposition. Together with a three-pass-rule, and pass not being a ko threat, that's exactly the same as with the \"knock\" move (at least strategically, of course you would have one more pass at the end of the game ;) ).\ \ Now why is it that people are opposed to one rule and in favour (or at least not so strongly opposed) to another rule when they are in fact equivalent?\ \ Of course the part about the opposition against 1) being stronger than 2) is a speculation. But I think the fact that \"they do not like another move type\", while still giving the first pass an endgame value, effectively making it \"another move type\" without calling it so, speaks for itself..." "pwaldron","","2008-09-17 11:42:54","[QUOTE=daniel_the_smith]Personally, I like the AGA's \"pass stones, white must pass last\" rule. *ducks*[/QUOTE] \ \ I'm a big fan as well. It is both signals a pass unambiguously and drives home to people used to Japanese-style counting that playing a dame is better than passing. It also allows funny situations to be played out without resorting to a lot of special cases." "RobertJasiek","","2008-09-17 11:53:49","flOvermind, discussion is far beyond that, also in Asia. Unawareness is not given, except for jokes. \ \ EDIT: \ \ Bill, although no one understands Ing Rules, I am close to understanding how they should have been designed :) My advantage is that Ing did not even make any attempt to classify fighting kos and that - now - distinguishing types of disturbing kos is straightforward (at least on the conceptual level). More somewhat later. His subclassification was not good. I will create a new, more useful classification. In particular, I shall introduce Dead Kos. (Eh, I invented the term many years ago, but now I can also define it.) - Do you know what a Closed Death is? Take Basic-Fixed-Ko Rules and study ordinary triple ko :) BTW, now I wonder why the hell a triple ko is called triple ko - it contains only two kos at a time...!" "flOvermind","","2008-09-17 12:03:51","Then I'm even more confused. How can anyone be against \"new move types\" when fully being aware that it is strategically equivalent to the current ruleset, minus a lot of unwanted special cases?" "RobertJasiek","","2008-09-17 12:05:28","Good question." "Harleqin","","2008-09-17 12:31:09","[QUOTE=pwaldron] \ \ [quote=daniel_the_smith] \ Personally, I like the AGA's \"pass stones, white must pass last\" rule. *ducks* \ [/quote] \ \ I'm a big fan as well. It is both signals a pass unambiguously and drives home to people used to Japanese-style counting that playing a dame is better than passing. It also allows funny situations to be played out without resorting to a lot of special cases.[/QUOTE] \ \ Yeah, well, pass stones are just a method. The method of pass stones can be applied to area rules with last move compensation in the following way: \ \ - pass stones, \ - white must pass last, \ - if white is first to pass, he doesn't pay a pass stone for that pass. \ \ This way, in most games there is not even a need to exchange pass stones, and territory counting can be used." "flOvermind","","2008-09-17 12:47:21","Except that you have problems with one-sided dame again ;)" "Bill Spight","","2008-09-17 13:43:42","[QUOTE=flOvermind]I'm not trying to be clever about go rules, I'm trying to be clever about \"selling\" go rules.[/QUOTE] \ \ Yes, I realized that. Sorry, I should have acknowledged it." "Harleqin","","2008-09-17 16:46:58","[QUOTE=flOvermind]Except that you have problems with one-sided dame again ;)[/QUOTE] \ \ Well, you need to make the firstpass not count towards the game stop condition." "SpongeBob","","2008-09-18 14:02:35","Sorry to interrupt - I find this discussion rather interesting, although I don't understand that much. :cool: \ Let me ask a stupid question, please ... \ \ [QUOTE=Harleqin]Yeah, well, pass stones are just a method. The method of pass stones can be applied to area rules with last move compensation in the following way: \ \ - pass stones, \ - white must pass last, \ - if white is first to pass, he doesn't pay a pass stone for that pass. \ \ This way, in most games there is not even a need to exchange pass stones, and territory counting can be used.[/QUOTE] \ That sounds great. :) \ \ \ [QUOTE=Harleqin]Well, you need to make the firstpass not count towards the game stop condition.[/QUOTE] \ But what does that mean ?? :o" "Harleqin","","2008-09-18 16:25:06","[QUOTE=SpongeBob] \ \ [quote=Harleqin] \ Well, you need to make the firstpass not count towards the game stop condition. \ [/quote] \ \ But what does that mean ?? :o[/QUOTE] \ \ The firstpass is different from all other passes in that it has a value, compensating White if Black had the last move, resp. not compensating White if White had the last move. \ \ The game stop condition is the condition when the game will stop. Usually, this is two passes because passing is interpreted as the player believing that nothing worthwhile is left on the board. However, the firstpass has a value, so it cannot be interpreted in this way. That is why it should not count towards the two passes that lead to a game stop." "SpongeBob","","2008-09-19 03:36:19","Ah, thanks Harleqin. \ \ I'm just wondering about this, however: \ \ [QUOTE=Harleqin]This way, in most games there is not even a need to exchange pass stones, and territory counting can be used.[/QUOTE] \ If white passes first, and then black passes, black has to pay a pass stone because he doesn't know if white will resume, right? \ \ So in my understanding, if the handover of a pass stone is used to indicate the pass, there should always be the need to exchange pass stones." "Harleqin","","2008-09-19 08:17:54","[QUOTE=SpongeBob] \ If white passes first, and then black passes, black has to pay a pass stone because he doesn't know if white will resume, right? \ \ So in my understanding, if the handover of a pass stone is used to indicate the pass, there should always be the need to exchange pass stones.[/QUOTE] \ \ Formally, yes. Informally, a lot of things happen, and in most games, there will just be no difference :) Since we are talking about rules here, I should not mention informal deviations, though. I'm sorry." "Bill Spight","","2008-09-19 08:36:04","[QUOTE=SpongeBob]If white passes first, and then black passes, black has to pay a pass stone because he doesn't know if white will resume, right? \ \ So in my understanding, if the handover of a pass stone is used to indicate the pass, there should always be the need to exchange pass stones.[/QUOTE] \ \ Well, if you use pass stones, you can exempt White from handing over a pass stone on the first pass (if White makes it instead of Black)." "Nate","","2008-09-22 12:01:39","[QUOTE=TMark]Ithink that Robert is, or has become, the Large Hadron Collider of Western Go, searching for smaller and smaller particles of definition within the rules. Unfortunately, one word constantly misused by him is \"strategic\" which has no place within the ever diminishing minor nit-picking details of possible ko captures to the nth degree. \ \ Read them at your peril! Best wishes.[/QUOTE] \ \ Hilarious." "snorri","","2008-09-22 19:28:53","[QUOTE=Bill Spight]The first pass, so-called, is not really a pass. The rules say, \"Playing alternately is the right of both players, but either player may give up the right to make a play and this is known as a pass.\" The first \"pass\" does not give up the right to make a play. It is a play that gains 1/2 point.[/QUOTE] \ \ I'm willing to accept the idea that under area scoring, that sometimes playing a dame is better than taking or filling a ko. That's fairly easy to understand and explain. But this idea of using passes to gain points is offensive. It is also disgusting to leave an open (single) ko and claim in the scoring phase: \"that point is really mine.\" \ \ A pass should mean that the player making the pass can no longer alter the score. It should not be some kind of weird parity game for the last meaningful point. I would indeed rather have button go, and these WMSG rules actually make it more attractive. :D" "flOvermind","","2008-09-23 02:26:20","That's exactly the point. A pass should mean \"I don't have any meaningful moves\". So overloading the pass with an additional meaning creates a contradiction ;) \ \ Button go, and \"first pass does not contribute to game end\" are actually equivalent solutions to this problem. The only difference is the terminology..." "SpongeBob","","2008-09-24 12:02:19","[QUOTE=TMark]Ithink that Robert is, or has become, the Large Hadron Collider of Western Go, searching for smaller and smaller particles of definition within the rules.[/QUOTE] \ He get's almost as much publicity and criticism for his research as the high energy physicists. :) \ \ [QUOTE=TMark]Unfortunately, one word constantly misused by him is \"strategic\" which has no place within the ever diminishing minor nit-picking details of possible ko captures to the nth degree.[/QUOTE] \ I thought the same. What about 'Late endgame strategy'?" "RobertJasiek","","2008-09-24 16:26:46","About 2 persons have provided noteworthy constructive criticism of the Japanese 2003 Rules version 35a, i.e., the so far final version. This is above average. 0 persons have pointed out the remarkable editing mistakes (in particular, some diagrams are missing!) in the online version of my so far final commentary on the Japanese 1989 Rules. My earlier endless lists of flaws of some particular ruleset caused more replies. Only rather recently a few people have praised my overall work of creating rulesets. So, in the eye of the beholder, it is not totally impossible that the quality has risen indeed. (I assure you: it has. Now that the quarks of go rules and go terminology theory have been discovered, it has become much easier to create quality reasonably quickly.) \ \ So a comparison to public curiosity about high energy physicists is reasonable. More so than comparing me with the LHC; I am not the machine used for research but the theoretician postulating the higgs particle. \ \ :)" "RobertJasiek","","2008-09-26 18:39:23","There has just been an announcement of 1st WMSG - Supplements to \ Regulations for Go Tournaments. It contains this additional ko rule: \ >> \ For controversies over half-point ko before the endgame, the owner of \ the ko must end the ko and make the last play, despite the outcome of \ the ko. \ << \ \ \ This rule suffers from bad translation, wrong contents, doubtful \ terms, and ambiguous semantics. So what might it want to express? \ \ \ The most obviously, \"endgame\" is a wrong translation and should be \ \"game end\". \ \ \ Under WMSG Rules, \"half-point ko\" is a doubtful assessment of values. \ A basic endgame ko can assume the deire values 1 or 3 when there are \ or are not any dame. Dividing 1 by 2 gives \"half-point\" but why should \ one divide by 2? With which model? Even then, it might also be the \ other case of 3 divided by 2. \ \ Speaking of \"half-point ko\" makes some sense (simplistic 50% \ likelihood model of either ko connection or ko capture and connection) \ under Territory Scoring but WMSG Rules do not use Territory Scoring. \ \ Maybe we should forget about values and assume that \"half-point ko\" is \ just an informal term for what might also be called \"basic endgame \ ko\". \ \ \ Who is the \"owner of the ko\"? The player for whom the ko is open? The \ player who can win the ko? Something else? Let us assume the simplest \ guess: The player for whom the ko is open. (I.e., the one who can \ connect the ko stone by filling the ko.) \ \ \ Why the last play? There can be several basic endgame kos on the \ board. Does the rule apply only if there is only exactly one basic \ endgame ko left? This would be a dangerous ruling because one would \ have to conclude that the rule did not apply otherwise - when there \ are more than one basic endgame ko still on the board. Let us be \ pragmatic and assume implicitly that the rule has consequences for \ several basic endgame kos: a transformation to the case of only one \ basic endgame ko is expected. \ \ \ What is \"end the ko\"? Probably just \"fill the ko\". For basic endgame \ kos, the only reasonable ways to end something is a) to fill the ko or \ b) to pass. Since the rule does not mention \"pass\", presumably (a) is \ meant. \ \ \ \"must [...] make the last play\": Why \"last\"? There might be, e.g., \ several 1-sided dame for the opponent. Ok, let us not be so devious \ but consider a more ordinary case: The ko has been part of a dame ko \ fight, the opponent has just filled the second last dame, the player \ has run out of ko threats and so has to fill the ko (otherwise he \ loses the ko), then the opponent plays the last dame - i.e., the \ player cannot fulfil the rule's requirement to be the one to make the \ last play. But let us see below for further arguments. \ \ \ Here the condition \"despite the outcome of the ko\" comes in. \ (\"outcome\" is a bit ambiguous but we might pretend to understand it: \ either a player fills or the opponent captures and the fills. The \ outcome is about who might fill.) The condition, I am guessing, might \ be interpreted somehow like this: The ko is not fought. If there are \ ko threats, they are ignored and the players may not play them. \ Instead the last basic endgame ko is filled without fighting the ko. \ If there are dame (regardless of whether there might be a dame ko \ fight), the dame are filled and then the open player fills the ko. (If \ he gets the last two-sided dame, the opponent gets the first pass, and \ then the player gets the \"last play\", where - I haven't mentioned this \ further guess yet - approach moves to supposedly dead stones do not \ count as possibly being last plays.) \ \ \ What are \"controversies\"? Maybe \"ko threats\", \"dame as ko threats\", \ \"1-sided dame\", etc. Maybe verbal disagreements? One cannot tell. \ \ \ Maybe the rule means much less: After a player has won the ko, the \ opponent passes (likely the game's first pass), and then the player \ has to fill the ko. \ \ \ Guessing again, dead kos or kos in other disturbing kos are not meant \ to be \"half point kos\". \ \ \ It is really tough to guess the intended contents of the rule. All the \ above is pretty much nothing but guesswork. So what is the rule meant \ to say? \ \ \ Judging from hearsay, my guess of what the rule wanted to express is: \ \ \"A player may not pass if there is at least one _fighting ko_ on the \ board and he has at least one legal play other than filling his own \ liberty of a two-eye-formation.\" \ \ Here, \"fighting ko\" is meant in the rough sense of being the opposite \ of \"disturbing ko\". Note that I am close to defining fighting ko, but \ this requires about half a dozen or dozen of pages of definitions. \ Defining a more specialized type (basic endgame ko) is yet more \ difficult. But if we assume that the rule's authors didn't care about \ formalized theory and that they overlooked the \"he has at least one \ legal play other than filling his own liberty of a two-eye-formation\" \ side condition, it becomes: \ \ \"A player may not pass if there is at least one half point ko on the \ board.\" \ \ \ Apparently the only way to - hopefully - find out - at least part of - \ the intention is to hear some explanation at the WMSG... \ \ In particular, it is still an open question whether the first pass may \ serve as a tedomari." "pwaldron","","2008-09-26 18:45:43","Looks like the intention is to require that thousand-year kos get filled before the game ends. Situations in which only one player can connect are easy to imagine, so I suppose that player would be the 'owner' in some sense. It's also a less common situation, which might be why nobody thought to cover it until the rules supplement was issued.\ \ Poorly worded, though." "Bill Spight","","2008-09-26 21:17:30","[QUOTE]For controversies over half-point ko before the endgame, the owner of the ko must end the ko and make the last play, despite the outcome of the ko.[/QUOTE] \ \ I agree that \"endgame\" means the end of the game. The Korean rules speak of [I]Jongguk[/I], which is the point after all dame are filled, and I think that that is the general idea. \ \ The \"owner\" of the ko is, I think, a player who can win the ko, even if the opponent plays the ko first. A ko may not have an owner. \ \ I think that requiring the ko owner to \"end the ko and make the last play\" means that the ko owner is not allowed to pass with the ko unresolved (because passing means giving up the right to play). I am not sure. Perhaps there will be a commentary. \ \ Examples, with a very tentative interpretation: \ \ [go] \ $$ ---------- \ $$ | . . . . | \ $$ | X X X X | \ $$ | 1 X O X | \ $$ | X O . O | \ $$ | O O O O | \ $$ | . . . . | \ $$ ----------[/go] \ \ W2, B3 = pass. \ \ The passes are OK because there is no ko owner. The ko is contestable, so its points are shared. End result: Black +1, as White gets a point for the first pass. \ \ [go]$$W \ $$ --------------- \ $$ | . . . . . . | \ $$ | X X X X X X | \ $$ | X X O X 3 4 | \ $$ | X O 2 O X X | \ $$ | O O O O O O | \ $$ | . O . O . . | \ $$ --------------- \ [/go] \ \ W1 = pass, W5 takes ko back. \ \ Black protests. The game is returned to W1, and White is forced to fill the ko instead of passing. \ \ [go] \ $$ ------------- \ $$ | . X . X . | \ $$ | O X X . . | \ $$ | . O O X X | \ $$ | O . O O X | \ $$ | . O . O X | \ $$ ------------- \ [/go] \ \ B1, W2 = pass. \ \ [go] \ $$ ------------- \ $$ | a X . X . | \ $$ | a X X . . | \ $$ | a O O X X | \ $$ | O . O O X | \ $$ | . O . O X | \ $$ ------------- \ [/go] \ \ The [I]a[/I] points are contestable, and are shared." "RobertJasiek","","2008-09-27 01:34:18","Bill, can you support your concept of \"owner of a ko\" by facts or are you also only guessing this?" "Bill Spight","","2008-09-27 07:20:03","[QUOTE=RobertJasiek]Bill, can you support your concept of \"owner of a ko\" by facts or are you also only guessing this?[/QUOTE] \ \ Neither. I have no references in the literature to \"owner\" of a ko. But I am not guessing, either. If either player can win the ko, neither player can be said to own the ko, by the ordinary meaning of the term. A player may then be said to own the ko if he can prevent the opponent from winning the ko. The player may not be able to do so physically if the opponent plays first, but may be able to do so as a practical matter by winning the ko fight. To interpret the ko fight we run into the normal assumption of correct play." "RobertJasiek","","2008-09-27 11:36:22","Your \"owner\" concept can be derived from my \"fighting ko\" concept or defined by similar means. Simply speaking, fighting ko will be about like this: \ \ - a player moving first can force a two-eye-formation of his on the ko intersections, \ - the opponent moving first can force a two-eye-formation of his on the ko intersections, \ - from the ko stone captured position, the player moving first can force a two-eye-formation of his on the ko intersections, and \ - from the ko stone captured position, the opponent moving first can force a two-eye-formation of his on the ko intersections. \ \ However, the additional ko rule wants more: It also speaks of the last play and probably means the last play of CGT value greater than zero. This aspect of meaning intersects with the contestable points concept. \ \ Then there are the 1-sided dame asymmetry problems. \ \ Currently the additional rule(s) miss a clear concept and lack conceptual completion. \ \ It would be nicer if instead these additional rules would be used (under the assumption that somebody - Japanese and Koreans - needs to be pleased with addditional rules at all): \ \ - Dissolution of all fighting kos: A player may not pass if there is a fighting ko on the board unless the player would have to fill a liberty of a two-eye-formation. \ - Perfect pass: A player may not make an unvaluable pass if his play can improve the score for him. [Formally: If he can force score-improvement, which needs to be defined in its obvious way.] \ - Acknowledging that the first pass is a move of value: Ikeda style game stop by passes = after the first pass, the succession of two other passes stops the game. \ \ BTW, also Ing ko rules knew a dissolution of all fighting kos rule (The outcome of a ko must be clear.) OC, fighting ko was undefined for 23 years (my earlier New Ko Rules' definition was a substitute rather than a solution). Only now that I have two-eye-formation as a defined term, I am about to have a chance to reduce the fighting ko definition to it." "flOvermind","","2008-09-27 14:07:36","[QUOTE=RobertJasiek]- Perfect pass: A player may not make an unvaluable pass if his play can improve the score for him. [Formally: If he can force score-improvement, which needs to be defined in its obvious way.][/QUOTE] \ \ Isn't this somehow like saying that a player is prohibited from making an endgame mistake? A rule prohibiting suboptimal play looks questionable to me... \ \ What would happen when a player simply overlooks the last single point play? Normally, I would expect that this player would just lose one point. With your rule, what would happen? Would he be \"forced\" by the rules to gain an additional point, therefore correcting a mistake? (Ignoring the fact that the opponent is unlikely to complain :p)" "Harleqin","","2008-09-27 14:31:34","[QUOTE=RobertJasiek] \ Currently the additional rule(s) miss a clear concept and lack conceptual completion. \ [/QUOTE] \ \ My guess is that this is exactly what is happening there - a bunch of officials from japan, korea, china, and china, trying to change what they have no intent to comprehend, and partly succeeding. They add one little bit of nonsense after the other, until the final text is completely inapplicable chaos. Whoever the person currently in charge of the rules is, he has apparently not enough political weight or stubbornness to keep the rules text clean. \ \ I fear that we can only hope for a better WMSG in four years." "RobertJasiek","","2008-09-27 15:45:07","flOvermind: Perfect pass rule: Yes. It is (formulated differently) part of the Korean 1992 Rules. Of course, I have always been against such a rule (or partial, similar rules like the contestable points rules) because it is the players' right to make strategic mistakes during the game. This should be so without exception. \ \ But a perfect pass rule is at least much clearer than the current mess of contestable points rules, asymmetrical treatment of 1-sided dame, and partial perfect pass rule as part of the additional ko rule. \ \ Harleqin: right. - During the International Go Rules Forum meetings, I maintained these few things from that I would not retreat: Rules must be clear so that they can be understood easily by everybody and rules must be logical + complete so that they can always be applied, It is ironic to observe that, now that I am not at the first front, a rules consensus could be found and the rules are lacking exactly my favoured aspects. I doubt though that this gap does the rules anything good. Compromise is a fine thing and if the compromise means to have simple latest endgame strategy, so be it. But why not get it right immediately? Partial last minute patching of bugs does not convince but confuses the players, referees, and apparently even the rules writers themselves. \ \ Bill, when reflecting my sketched fighting ko definition, I am worrying about local threats. Do you think that using ko master, ko monster, or something similar is necessary for such terms after all?" "TMark","","2008-09-28 02:46:59","[QUOTE=RobertJasiek] \ During the International Go Rules Forum meetings, I maintained these few things from that I would not retreat: Rules must be clear so that they can be understood easily by everybody and rules must be logical + complete so that they can always be applied, It is ironic to observe that, now that I am not at the first front, a rules consensus could be found and the rules are lacking exactly my favoured aspects. I doubt though that this gap does the rules anything good. Compromise is a fine thing and if the compromise means to have simple latest endgame strategy, so be it. But why not get it right immediately? Partial last minute patching of bugs does not convince but confuses the players, referees, and apparently even the rules writers themselves. \ [/QUOTE] \ \ Would it have been at all possible to find out what the other delegates would accept before you make statements like the above? Despite many years of work, it appears that you have not yet established a set of rules to satisfy your own requirements and yet you seem deliberately to offend those very people from whom you need to gain agreement. You may, in your own view, win the argument but you lose the vote. \ \ Best wishes." "RobertJasiek","","2008-09-28 06:39:48","TMark, during every meeting, attempts to find out what every delegate wanted were made but, apart from default opinions (e.g., \"Simplicity is not all.\" or \"We will accept whichever ruleset...\" or \"Local is beautiful.\"), it was more often than not difficult to find out more. There was much discussion but little statements of moved opinion. Process was like the stereotype myth of Asian meetings: Sit together for two years - then suddenly treaties are being made. Things changed during the last meeting, when only the Area Scoring faction was present. \ \ There are numerous rulesets satisfying myself. Some of them are pretty old. But satisfaction is always for particular purposes. That, for other purposes, I search for other rulesets does not mean that I would be dissatified with some of the already given rulesets for the purposes I presume for them. \ \ All: \ \ According to Thomas Hsiang, the following wording suggested by me \ expresses about the intended meaning: \ \ \ \"When there are no two-sided dame on the board and there is exactly one basic endgame ko on the board, the player who can connect the ko stone has to do so on his next move, which therefore may not be a pass.\" \ \ \ Note that for the purpose of this rule, \"basic endgame ko\" should be understood in a somewhat broad sense. In particular, a basic ko nearby teire is also considered a basic endgame ko aka \"half-point ko\", regardless of the ko's value." "Bill Spight","","2008-09-28 07:05:40","[QUOTE=RobertJasiek]Bill, when reflecting my sketched fighting ko definition, I am worrying about local threats. Do you think that using ko master, ko monster, or something similar is necessary for such terms after all?[/QUOTE] \ \ No, I don't. They are good for analysis and perhaps for understanding the import of such definitions, but I would not include them as part of the definitions. :)" "snorri","","2008-09-28 09:45:32","[QUOTE=RobertJasiek] \ \"When there are no two-sided dame on the board and there is exactly one basic endgame ko on the board, the player who can connect the ko stone has to do so on his next move, which therefore may not be a pass.\" \ [/QUOTE] \ \ This is what I would have guessed the update to mean. (Only a rules wonk would think \"own\" means \"theoretically can win.\" :-)) \ \ So with this change, you can't just pass on the last move with an open ko and then claim it as a non-contestable point. It seems like it's constructed to fix the flaw you describe in section this section of your post here: \ \ 4) Exceptional basic endgame kos / study if passing is allowed when basic endgame kos are unfilled. \ \ So my question is: does this \"patch\" really fix that problem?" "Bill Spight","","2008-09-28 10:06:44","[QUOTE=snorri](Only a rules wonk would think \"own\" means \"theoretically can win.\" :-))[/QUOTE] \ \ Oh, really? If you were watching a go game and somebody told you, \"Black owns this ko,\" what would you think they meant?" "RobertJasiek","","2008-09-28 12:12:00","Whether the patch really fixes the problem is both still unknown and a matter of perception. \ \ Unknown: Nobody has enough time to really find out just a few minutes before the WMSG. \ \ Perception: It depends on what one perceives as \"half-point ko\". \ \ [go] \ $$ . O . . O . O . O # . \ $$ . O . . O # O O # # # \ $$ O O O O O O # # # . . \ $$ # # # # # # . # . . . \ $$ O O O O O . O # . . . \ [/go] \ \ White to move \ \ Thomas Hsiang perceives this as \"half-point ko\". I don't but would call it a fighting ko (or - on a more informal level - ko fight). \ \ [go] \ $$ . O . . O . O . O # . # . . # . \ $$ . O . . O # O O # # O # . . # . \ $$ O O O O O O # # O O # # # # # # \ $$ # # # # # # . # O . O O O O O O \ $$ O O O O O . O # O # . # # # # # \ $$ # # # # # # # # O O O O O O O O \ $$ . . . . . . . # O . . . . . . . \ [/go] \ \ I do not know anybody who would call this a \"half-point ko\". (Maybe Thomas? :) ) \ \ As you can see, it is a matter of definition of what is a \"half-point ko\"." "Bill Spight","","2008-09-28 12:41:54","Robert, I think that this is what you meant for your diagrams. :) \ \ [go]$$ \ $$ ---------------------- \ $$ | . O . . O . O . O X . | \ $$ | . O . . O X O O X X X | \ $$ | O O O O O O X X X . . | \ $$ | X X X X X X . X . . . | \ $$ | O O O O O . O X . . . | \ $$ ---------------------- \ [/go] \ \ BTW, I would call this a 1/3 pt. ko. It's a sente ko for Black. \ \ \ [go]$$ \ $$ -------------------------------- \ $$ | . O . . O . O . O X . X . . X . | \ $$ | . O . . O X O O X X O X . . X . | \ $$ | O O O O O O X X O O X X X X X X | \ $$ | X X X X X X . X O . O O O O O O | \ $$ | O O O O O . O X O X . X X X X X | \ $$ | X X X X X X X X O O O O O O O O | \ $$ | . . . . . . . X O . . . . . . . | \ $$ -------------------------------- \ [/go] \ \ This I would call a 7 pt. ko." "RobertJasiek","","2008-09-28 13:29:45","This is miai values. However, Thomas is saying that values and the \"half-point\" in the title don't matter for what is a \"half-point ko\" in the additional ko rule. However, I guess we need to wait to learn more about the intention of what shall be (not) a half-point ko." "Hsiang","","2008-09-28 15:25:39","I would like to make some comments about the misrepresentation of my view by Robert Jasiek. \ \ The comment I made (to Robert in private) about the specific example that Robert posed was that the intent of the new ruling (II-2 in the Supplement) is to enforce such end-game ko situation, IRRESPECTIVE of the value of the ko. \ \ Therefore W is required to finish the ko before passing, even though it is obvious that the ko is worth more than \"half point\". I think my response to Robert was very clear on this point. I have never attached the term \"half-point ko\" to this situation; I am surprised that Robert chose to misrepresent my view in public. \ \ I should also state that, to the organizers, my opposition for the II-2 ruling is well known. But at the same time I understand they have a HUGE tournament to run, a lot of people to please, and everyone is doing his/her best. Let's all be kind and appreciative. \ \ Thomas Hsiang \ \ [QUOTE=RobertJasiek] \ \ Perception: It depends on what one perceives as \"half-point ko\". \ \ [go] \ $$ . O . . O . O . O # . \ $$ . O . . O # O O # # # \ $$ O O O O O O # # # . . \ $$ # # # # # # . # . . . \ $$ O O O O O . O # . . . \ [/go] \ \ White to move \ \ Thomas Hsiang perceives this as \"half-point ko\". I don't but would call it a fighting ko (or - on a more informal level - ko fight). \ \ [go] \ $$ . O . . O . O . O # . # . . # . \ $$ . O . . O # O O # # O # . . # . \ $$ O O O O O O # # O O # # # # # # \ $$ # # # # # # . # O . O O O O O O \ $$ O O O O O . O # O # . # # # # # \ $$ # # # # # # # # O O O O O O O O \ $$ . . . . . . . # O . . . . . . . \ [/go] \ \ I do not know anybody who would call this a \"half-point ko\". (Maybe Thomas? :) ) \ \ As you can see, it is a matter of definition of what is a \"half-point ko\".[/QUOTE]" "Javaness","","2008-09-28 15:49:39","[QUOTE=Hsiang]I would like to make some comments about the misrepresentation of my view by Robert Jasiek. \ \ The comment I made (to Robert in private) about the specific example that Robert posed was that the intent of the new ruling (II-2 in the Supplement) is to enforce such end-game ko situation, IRRESPECTIVE of the value of the ko. \ \ Therefore W is required to finish the ko before passing, even though it is obvious that the ko is worth more than \"half point\". I think my response to Robert was very clear on this point. I have never attached the term \"half-point ko\" to this situation; I am surprised that Robert chose to misrepresent my view in public. \ \ I should also state that, to the organizers, my opposition for the II-2 ruling is well known. But at the same time I understand they have a HUGE tournament to run, a lot of people to please, and everyone is doing his/her best. Let's all be kind and appreciative. \ \ Thomas Hsiang[/QUOTE] \ \ Robert probably thought is was a strategic misrepresentation. :)" "RobertJasiek","","2008-09-28 22:14:10","Thomas, I have not intended to misrepresent your statements. Please excuse me for possibly having done so. \ \ I had understood and still understand you so that, \ - in an \"end-game ko situation\", the additional ko rule II.2 shall be applied regardless of the value of the ko, \ - the term \"half-point ko\" in the additional ko rule II.2 would be used regardless of the value of the ko and while not interpreting the \"half-point\" in the name as a value of the ko. \ \ Now you say that you have never attached the term \"half-point ko\" to this situation of this example. I guess there I have misunderstood you. I thought you would want to apply the additional ko rule II.2 also to this example, when - in email - you said \"W is required to finish the ko according to the II-2 rule.\". Our confusion arises from the additional ko rule II.2 usage of the phrase \"half-point ko\". Let us replace that for the moment by \"end-game ko situation\" to get rid of the \"half-point\" phrase misinterpretations. So then the additional ko rule II.2 would - according to your interpretation - demand that White is required to finish (or closer to the rule: to end) this \"end-game ko situation\". \ \ If White connects his three stones to the left, this does not end the ko. I guess that the rule did not consider this example position but - had the author of the rule reflected it in time - would have preferred a softer ruling like \"may not pass yet\". \ \ Similarly we may assume that the author (or the translator) did not think carefully enough about the phrase \"half-point ko\". Your phrase \"end-game ko situation\" is more liberal and might even include my bigger ko fight. This is a pragmatic phrase. More pragmatic than my suggestion of \"fighting ko\" elsewhere, which demands for a definition because it has previously been used in Ing ko rules, etc. \ \ So the current pragmatic interpretation is: \ \ \"Roughly the following shall apply to prevent the game's first pass when there is still an end-game ko situation: When there are no two-sided dame on the board and there is exactly one end-game ko situation on the board, the player having the turn may not pass.\" \ \ Please feel free to improve on this to possibly yet better interpret the additional ko rule II.2."