Report: 6th Meeting of the International Go Rules Forum
Robert Jasiek
Report by Robert Jasiek
Venue and Dates
USA, Seattle, Hotel Hilton Sea-Tac, Board Room.
2006-07-03 + 2006-07-04.
Participants
AGA: Chris Kirschner
AGA: Terry Benson, to some extent implicitly acting as the
meeting's chairman
China: Jin Tong Shi
China: Wu, professional 9 dan
EGF: Robert Jasiek, sometimes assisting the meeting's chairman
Ing Foundation Shanghai: Ni Yao Liang
Ing Foundation Taipeh: Yang Yu-Chia
Ing Foundation Taipeh: Ying
Singapore Go Association: Tang Teng Chung
Technicalities
This report is unofficial and based only on the author's notes
and memory.
Any accidental inaccuracy in the report would be the author's
fault. Jin, Wu, and Ni are a bit under-represented in the report
because not all Chinese discussion was translated.
Translations between Chinese and English: Yang (frequent), Ying
(less frequent), Tang (scarce).
In between the 5th and the 6th meetings, some documents were
circulated by email. This includes detailed commentaries by Jasiek on
Area Scoring, Ikeda Territory I Scoring, and ko rules. During the 6th
meeting, a collection of printed reports and a CD with factual
documents of the 1st to 5th meetings is distributed.
In the Discussion section and at the beginning of a paragraph,
"China:"stands for "Jin and Wu:".
General Nature
The meeting started with a serious attempt to find out the IGF's
current opinion.
Because during earlier meetings Japanese and Korean delegates had
not or only partially shown interest in getting a general compromise
and they do not want to attend this meeting, it was voted to create
Unified Area Rules.
The major part of the meeting was an efficient discussion about
and voting on the contents of Unified Area Rules.
Ko rules are still an open issue because it was felt that a
likely compromise should be formulated and studied carefully until and
during a next meeting.
Votes
Numbers of votes are stated as Y:N for Yes versus No or as Y:N:A for
Yes versus No versus Abstention. Some of the 100% one-sided votes were
made by implicit general agreement. To ease reading, this report
translates all voting issues to let the Yes votes appear to have won,
even though the actual issue might have been formulated as its
negative during the meeting. If, in the list below, an issue uses a
negation, then a Yes vote confirms that negation.
Kirschner did not consider himself to have another AGA vote and
therefore did not vote at all while the three Ing Foundation delegates
all participated in the voting, except when Ying was absent. Because
most votes were heavily one-sided anyway, this did not matter and so a
request to adjust the actual voting system was not made during this
meeting to avoid wasting time on that technical matter.
(1) The ruleset's name is "Unified Area Rules". 8:0.
(2) Let us not talk about Ikeda Rules during this meeting. 6:1:1
(No = Jasiek, Abstention = Wu).
(3) The scoring is Area Scoring, i.e. at the end a player's score
is the number of his stones on the board plus empty intersections
surrounded by only his stones. 8:0.
(4) Suicide is prohibited. 4:3:1 (Yes = Jasiek, Tang, Jin, Wu; No
= Ying, Yang, Ni; Abstention = Benson).
(5) The players alternate. 8:0.
(6) Black starts the alternation. 8:0.
(7) The available types of moves are board-plays and pass-plays.
8:0.
(8) Pass-plays are always allowed. 8:0
(9) After 2 successive passes and if the players agree, then the
game can end. After any 2 or more successive passes, an agreement
phase is possible. After exactly 4 successive passes, the game ends.
In the previous conditions, numbers of passes in a succession of
passes are counted regardless of whether they might have been
interrupted by one or more than one agreement phase. 6:1 (No =
Jasiek).
(10) If expressed in full counting and intended for the 19x19
board, the size of komi should be 7.5 or 8 (Black wins ties). 7:0.
(11) The following ways of expressing the komi are possible:
"3.75 points [in half counting]", "7.5 points [in full counting]", "8
points (Black wins ties) [in full counting]". 7:0.
(12) Every counting procedure for Area Scoring is possible. A
tournament should announce its used counting procedure. 7:0.
(13) In case of language problems between the players, it is a
tournament organization's (but not the rules') task to suggest methods
for resignation or for an agreement phase. 8:0.
(14) Handicap stones may be placed on freely chosen
intersections. 8:0.
Open Urgent Tasks
write ruleset
bring English and Chinese language in ruleset in agreement
discuss, agree on ko rules; for long cycles find a compromise
between Chinese Ko Rules and Long Cycle Ko Rules
after the rules will be finished, submit them to superiors and
then the IGF commentary on ruleset, in particular on its ko rules
explanations of all counting procedures for Area Scoring
Discussion
(day 1, 9:00)
Ying: The AGA proposal [that the IGF accept the IGRF as the body
to suggest rules of play for IGF] was not discussed at the IGF
director's meeting.
Kirschner: At the IGF director's meeting, discussion about it was
postponed. Decisions were made before the IGF director's meeting. A
"European" letter (unofficial but probably accurate) was written with
the help of Shigeno.
Jasiek: He and Tony Atkins tried to contact the IGF directors
representing the EGF so as to get a report about the rules issues made
during the IGF meeting. The attempts were futile.
Ying: Should we reconsider the intention of the rules to be made?
a) international, b) Olympic, c) etc.?
Yang: Will China reject the World Amateur Go Championship Rules
(also called [by the IGF] International Rules) for the 2008 Olympics?
Wu: This is uncertain. It is in the Chinese gouvernment's hand.
Tang: The hosting country [of Olympics] cannot add extra sports.
Ying: The Chinese Olympic committee was asked if Go can be a [so
called] demonstration sport [which essentially is a side event] in
2008's IOC Olympics. It can if, e.g., Ying sponsors. His conditions he
told them at that time were: The rules are Ing Rules or "Olympic
Rules". Federations get a letter from us hopefully suggesting also
they might support it. It requires ca. $100,000 from [the Ing
Foundation] and other sponsors.
Jasiek: What is sufficient for adopting "Olympic Rules"?
Ying: It is sufficient if the IGRF agrees on a ruleset.
Kirschner: The GAISF accepts only one body per sport. This would
be the IGF. We have not seen any IGF report on its discussions and
votes yet.
Benson, Ying: Currently in effect there are two parallel paths
towards [international rules]: 1) the IGRF, 2) the IGF and the
Japanese' related "behaviour". The two are in opposition. The IGF is
still much controlled by the Japanese.
Benson: The IGRF activity can be brought to the IGF via those IGF
directors representing China, the EGF, or the AGA. He fears though
that the Nihon Kiin's support of the European Go Cultural Centre
influences Erik Puyt, who also works for it.
Kirschner: The AGA/EGF proposal was rejected [by the IGF], so the
AGA should not do the same attempt soon. Instead China should make the
next attempt. The AGA board is not really supporting him and Benson.
It should be China because it is a major Go country and because the
2008 Olympics are there. - He tells the major contents of an EGF
letter sent before the IGF meeting: There should be improved multiple
rulesets; the IGRF's work is approved. He wonders if later this EGF
opinion was tabled or influenced by Shigeno [who is known to favour
Japanese rules as international rules].
Jasiek: The official report on the IGF meeting is not available
yet [to the general public].
Kirschner: China should ask for a report.
Tang: 2008 is important for China, especially because in China Go
is a "sport". This makes things easier than if Olympics are in other
countries.
Jin: The Chinese IGF director Hua Yigong has not reported yet. A
fax was sent. The Chinese gouvernment told the Chinese Go Association
that Go as a side event in 2008 is approved.
China: They want the Chinese Go Association to take the chance to
use Chinese Rules [or similar rules]. The Chinese gouvernment pays
high attention to the "2008 mind sports event". The Chinese Go
Association organizes this. Rules should help this.
Jin: Rules should be a tool rather than a problem creator.
Jasiek: Agrees.
Tang: Go fans say: Chinese rules are good, Ing Fill-in Counting
is better.
Jin: He is famous in [Chinese language] Internet forums as a
rules expert. There also he has made Tang's statement. - Chinese
Counting counts stones and is a remainder of China once using Stone
Scoring. Contrarily Ing Fill-in Counting counts Area. This is more
fitting. So Ing Fill-in Counting should replace Chinese Counting. A
fundamental of the Ing Rules is its counting method. Ing Rules as
written in Chinese use the ancient Chinese language, which is compact.
That text consists of exactly 1,000 "words".
Ying: He agrees with Jin. The highlight Ing [Fill-in] Counting
makes irregular territories countable. Ing Rules have many other
features. Ing Chang-ki spent 10 years on ko rules. William Mann sent
him a letter [that suggested why the Superko Rule should be replaced
due to the behaviour of a few very rare particular shapes; the
author]. There are four important topics: a) komi, b) "counting
system", c) time system, d) ko rules. The Ing people don't insist on
"that" [or does he mean: "the Ing ko rules"?; the author]
(break 10:50 - 11:00)
Jasiek: Which ruleset should we develop: a) for international
professional tournaments, b) for Olympics, c) three rulesets: I) Area
Scoring, II) Ikeda Territory I Scoring, III) Traditional Territory
Scoring?
Ying: If we use the word "international", we might annoy the
Japanese and Korean fans.
Kirschner: The IOC sues improper usage of the word "Olympic" as
in "Olympic Rules".
China: The Chinese Go Association will recommend the IGRF
ruleset, if there will be any.
Ying: We should first agree on its name.
Jasiek: It is easy to find some name: E.g. "Unified Area Rules"
or "International Area Rules".
Ying, others: The word "international" is too big a topic.
[Implicitly this leaves "Unified Area Rules" as the consensus and is
listed as vote (1); the author.]
Tang: He wonders about possibly having three IGF accepted
rulesets. Our purpose is rather to come up with just one ruleset, even
if it has different applications. Singapore wants to apply unified
rules.
Yang: A sponsor has the right to decide the rules. Regardless,
progress should be made about the quality of the rules, in particular
if they are applied for international professional tournaments.
China: A neutral ruleset might be used for the Olympics 2008. If
it is an Area ruleset, then it will be used for national Chinese
professional tournaments. They will have to suggest [the unified]
ruleset to Chinese superior officials but if Ying ("big sponsor")
approves, then the superior officials' agreement will be very likely.
For international professional tournaments, [usage of the unified
ruleset] is very likely when they are sponsored by Chinese entities.
If the Olympics 2008 should use [the unified ruleset], then other
Chinese tournaments will also use it, including amateur tournaments.
One [general] case of disagreement [about removals during an optional
agreement phase] may be involved; there must not be special rulings.
Jasiek: One ruleset for all is to be preferred; to propose three
rulesets is an emergency measure.
Benson: Our aim should be to propose only one ruleset for the
Olympics.
Tang: The new ruleset should be based on Chinese / Ing / Japanese
rules.
(break 12:15 - 13:55)
Ying: Area Scoring has only advantages.
Jasiek: No. We should be honest. It does not provide the
advantage of both komi parities being equally meaningful in practice.
Besides the Japanese and Koreans would want to add that dame are not
unvaluable.
Benson: He proposes Area Scoring.
Jasiek: He proposes first to discuss both Area Scoring and Ikeda
Territory I Scoring and only then to decide.
Ying: The EGF should first test Ikeda Scoring for a few years.
Jasiek: He will ask the EGF.
Some: Why were compromise scoring methods studied at all?
Jasiek: He explains why: Because during earlier meetings there
have been requests for such. He briefly mentions a few used methods to
study Ikeda Territory I Scoring to the effect that they can be
considered well tested on the strategic level but, of course, need
practical testing of how players actually perceive application of such
rules.
Ying: The Japanese don't want a compromise. Some Koreans want,
some don't want a compromise.
Jasiek: He agrees as far as the Koreans are concerned. The
Japanese sometimes say to want a compromise while at most times they
don't want a compromise.
Vote (2).
Jasiek: He puts the following discussion topics on the wall: 1)
rules text, 1.1) wording, 1.2) structure, 2) rules of play, 2.1)
scoring, 2.2) relation (i.e. interdependence of scoring, phases, ko
rules, numbers of passes), 2.3) ko rule(s), 2.4) phases, 2.5) numbers
of passes, 2.6) definitions of basic terms, 2.7) alternation, 2.8)
suicide, 3) rules in between rules of play and tournament rules, 3.1)
komi, 3.2) counting, 3.3) optional agreement, 3.4) resignation, 3.5)
handicap stones, 4) tournament rules (time, referee, etc.). Within
each major section, the more important subtopics are mentioned first.
He explains the topics a bit.
Ying: Rules of play and tournament rules should be separate. Let
us talk about rules of play first.
Jasiek: He clarifies what we want for scoring.
Vote (3).
Jin, Tang: No Suicide is closer to Japanese rules.
Jin, Tang, Ying: No Suicide is an unnecessary restriction.
Jin: The Chinese Go Association is strict about having Area
Scoring and a playout. At a time, Chinese Rules had Suicide, recently
they have No Suicide.
Kirschner: We should try to identify what we think is the best,
except in minor things.
Jasiek: He considered wording of the case of one's own string
being without liberty. The solutions Suicide / No Suicide are equal
options in this regard. The effect is a greater / smaller strategic
variety in the games. It is personal preference which to prefer.
Either one decides according to one's personal preference or one does
favour to the Japanese / Korean preference to allow an easier
world-wide unification of the rules later.
Jin, Wu, Jasiek: Their personal preference favours Suicide but
their votes in favour of No Suicide intend to make world-wide
unification of the rules easier.
Vote (4). [This is the only close vote.]
Vote (5).
Vote (6).
Jasiek: The available types of moves should be board-plays and
pass-plays.
Vote (7).
Jasiek: As an explanation, passes are formal and non-verbal means
to approach the end of the game. Without passes, there would be
frequent so called pass-fights.
All: [A short discussion on whether passes are always allowed
takes place.]
Vote (8).
All: [Everybody just notices that Black might pass on his first
move but this would be a strategic mistake and he would lose because
of komi. Anyway, this might happen in theory.]
Jasiek: The players have the right to make strategic mistakes,
regardless of whether it is a board-play or pass-play that they make.
All: [What should be the number of passes to end the game?]
Jasiek: How exactly do those 4 passes work, which are suggested
by Benson and Kirschner?
China: In Chinese Rules, 2 passes end the game. Just for the
rules book. In the practice of an actual game [of a
Chinese professional tournament], it is always done by informal
agreement with informal means to express that one is passing. There is
a verbal rule only: A forgotten teire may be filled even after 2
[successive] passes.
Benson: Written rules should agree to verbal practice.
Jasiek: [He poses detailed questions to China, Yang, Ying to
understand how exactly the Chinese Rules and the Ing Rules work during
an optional agreement.]
Ying: During an optional agreement under Ing Rules, removal of
strings requires the players' agreement. Only whole strings may be
removed.
Some: Agree.
(end of day 1: 17:20; day 2 starts 9:25; due to a cold, Ying misses
until lunch)
China: Since there can be players of different language: at least
2 [formal?] passes are better for mutual understanding.
Kirschner: There should be 2 passes, then the players may have an
agreement, then there should be 2 passes because this is
language-neutral, and a loop should be possible.
China: A case of disagreement should always be observed by some
referee.
Jasiek: A disagreement is so rare that, in a tournament, it
should always be done together with a referee.
Tang, Jasiek: [They show flow charts to illustrate possible game
ending procedures.]
Jasiek: He suggests the following procedure: Either there are 2
passes, the players' agreement, and then the scoring or there are 2
passes, the players' disagreement, the playout, and then the scoring.
Benson: A second possible agreement should be possible. Therefore
play should end by 4 successive passes, even if there is some
(dis)agreement phase in between. As a consequence of 4 successive
passes leading to the final end, infinite loops are impossible.
Benson, Kirschner, China: A procedural loop should be possible to
allow for temporary forgetting of removals.
Benson, Kirschner: [They suggest details for vote (9).]
Jasiek: [He asks for clarification of details for vote (9). In
particular, he clarifies and Benson confirms that the following shall
be possible: Pass, pass, agreement phase, pass, agreement phase, etc.]
Vote (9).
Benson: [If expressed in full counting,] the size of komi should
be 7.5 or 8.
Vote (10).
Benson: Is it just a matter of "wording" or of choosing the
counting procedure whether one says "7.5" or "8 (Black wins ties)"?
China: We are not authorized for everything else than 3.75 [in
half counting].
Jasiek: We should define how to state the komi.
Kirschner: Can we state it in all three ways of expressing it?
Jin: The three ways of expressing it are not strictly "equal"
because all their numbers differ.
Jasiek: Can the Ing / AGA / EGF delegates vote on "7.5" versus
"8" to unify at least that?
Wu: Let us put "7.5" in a footnote.
Tang: Let us state all three [in the rules]. [Recompensation
points for] handicap stones are tournament rules, so let us discuss
this later.
Vote (11).
(break 10:33 - 10:45)
Jasiek: If a sponsor insists, then let the counting procedure
depend on the sponsor, otherwise allow every counting procedure for
Area Scoring.
Jin: Every counting procedure for Area Scoring should be
permitted.
Ni: Ing Fill-in Counting should be used.
China, Yang: This is too specific; we should not raise passions.
Tang: Every tournament should choose its counting procedure,
which has to be suitable for Area Scoring.
Jasiek: Every tournament's organization should specify its used
Area Scoring counting procedure.
Kirschner: We should provide explanations for every possible
counting procedure.
Vote (12).
Yang: Ying has said that we will not insist on Ing Ko Rules. How
does the Chinese Go Association handle ko rules?
All: [Discussion about Superko. Almost it is adopted until
everybody becomes aware that the Chinese Rules do not mean [Superko]
when they say so.]
Jasiek: The standard Superko Rule is: "A board-play may not
repeat an earlier position."
Jin: The Chinese Rules use [the Positional Superko Rule].
Wu: The Chinese Rules use [the Situational Superko or the Natural
Situational Superko Rule; it remains unclear which he means; the
author].
Jin: In practice, Chinese Rules use No Result due to referee
ruling. [He explains a bit more about what the Chinese referee ko
rules might mean.]
Terry: We should use [the Positional Superko Rule].
Jin: We do not know all ko shapes, so we should allow a referee
to decide. This is easily put into rules but difficult for the referee
to apply.
Kirschner: The first player to notice repetition may inform the
referee.
China, Yang: Superko is too difficult strategically.
Ni: The Ing Foundation has a different opinion but does not
insist during this meeting.
China: Ko rules are difficult. The Chinese Go Association has not
reached a unified conclusion yet. So always No
Result is concluded.
Benson: What about the Olympics, where one needs a winner? Can
the winner be decided by nigiri if a long cycle occurs?
Jasiek: Life and death with its many variations and long
move-sequences is much more difficult than recognizing a repeated
whole board position with shapes like, say, Eternal Life.
China: We [or all Chinese professionals?] disagree. - If under
Chinese Rules a referee decides due to a long cycle, then he may
declare either a tie, a No Result, or a replayed game. He may not
declare a playout of the position.
Jin: Rules must be executable and reasonable. Under Chinese
Rules, the referee may apply one of the aforementioned long cycle
outcomes also just by the shape on the board, i.e. without any moves
being played to create a long cycle.
All: [The discussion becomes even hotter.]
China, Yang: We dislike Superko.
Benson, Kirschner, Jasiek: We like Superko.
All: [Although a vote is not done, it becomes clear that pure
Superko drops out as a possible ko rule for the unified ruleset.]
Jasiek: Chinese professionals are worried about which player gets
a $400,000 prize money. Now, in case of a long cycle, China wants a
referee to decide who shall get the prize. Is that not contradicting?
China: We want to let the referee decide nevertheless.
All: [Most delegates are about to suggest cutting discussion and
voting on adopting the Chinese Ko Rules while Jasiek wants to propose
an improvement on them. The lunch break comes just in time to avoid a
possibly too rash, insufficiently reflected vote.]
(break 12:28 - 14:15)
Kirschner: Concerning the power of the referee, there appears to
be a cultural difference. US culture: first come the law / the rules,
then the referee applies them. Eastern culture: The power of the law
is in the referee as a person. The West is comfortable with letting
the rules decide - the East is comfortable with letting the referee
decide. - The referee should have at least the option to have a
Superko playout by the players.
Wu: Chinese long cycle tournament rules allow a replayed game
when the old game's remaining thinking times of the players are still
valid.
Benson: The referee's choices should be given due to rules.
Jasiek: Why should there ever be No Result?
Jin: Chinese Rules about long cycle kos allow the following
referee options: 1) "No Result" with the additional meaning of
replaying the game using the old rest thinking times, 2) "replayed
game" with full new thinking times for both players, 3) "No Result"
with the additional meaning of not having any replay but considering
the outcome as "both players lose", 4) "drawn game", i.e. a tie.
Before every tournament, it is announced which of the choices the
referee will make.
Jasiek: During the last meeting, the Chinese delegates have said:
"We want to learn more about ko rules!". So why not continue to
discuss ko rules now?
China: Because ko rules are too difficult. Therefore we should
discuss them later: during the next meeting.
Benson: When does the full-board repetition rule actually apply
in the Chinese Rules?
All: [Discussion continues a bit.]
Jasiek: There is an easy classification [as a means of improving
the Chinese Ko Rules]: During a cycle, there is either a zero or a
non-zero difference of removed black and white stones. For each case,
one can use a rule: Either the game is a tie [or the player with fewer
removed stones wins]. This is easy, applicable, and the professionals
do not need to worry about losing due to a too difficult ko rule.
Benson: During an agreement phase, should we use some sort of
pass stones to overcome language problems between the two players?
Vote (13).
Benson: The next topic is Fixed versus Free Handicap.
Vote (14).
Jasiek: As an explanation why we have chosen Free Handicap, it
was very good for the development of Go in Japan when they started all
their even games without any handicap-like opening pattern.
All: [In emails, we should work out a rather complete rules text
in English, then translate it.]
All: [Ko rules discussion continues.]
Jasiek: Yang has said that we were in search for the best rules.
So why should we not do this also for ko rules instead of taking just
some?
Ying: He suggests to have an experts' meeting in China lasting 3
days about ko rules.
All: [Everybody agrees but before ko rules discussion shall
continue by emails.]
China: The referee should be involved in long cycle incidents.
Tang: He asks for writing of a completion of the current Chinese
Ko Rules.
Jasiek: The zero / non-zero removals difference is such a
possible completion.
China: There is room for improvement of the Chinese Ko Rules.
China, Ying: We will check and discuss also Jasiek's suggestion.
Benson: When we will have a draft of the [ko?] rules, it should
be discussed among the delegates by email. The Chinese delegates
should then present it to the Chinese Go Association for approval.
Jin: Ok.
Kirschner: The IGF will make an application of recognition to the
IOC in December. So we should have ready our proposal before that
time.
Benson: The Chinese IGF director should then submit the IGRF
rules proposal to the IGF.
All: [It is mentioned that there will have to be tournament rules
for the Olympics 2008. Now it is still not the time to
discuss their contents.]
All: [The next meeting's date is not fixed. First let us exchange
the necessary emails about ko rules and rules text.]
(end 16:01, one hour before the scheduled time)
[Citing parts of the or the whole report is allowed. In particular,
journalists may use it.]
--
robert jasiek
Barry Phease
> 6th Meeting of the International Go Rules Forum (IGRF)
>
> Report by Robert Jasiek
very interesting.
> Because during earlier meetings Japanese and Korean delegates had
> not or only partially shown interest in getting a general compromise and
> they do not want to attend this meeting, it was voted to create Unified
> Area Rules.
While a unified international area rule specification is very desirable,
it does seem a shame that the Japanese and Korean delegations should have
dropped out. Is it likely that they will rejoin for a future meeting?
> The major part of the meeting was an efficient discussion about
> and voting on the contents of Unified Area Rules.
> Ko rules are still an open issue because it was felt that a
> likely compromise should be formulated and studied carefully until and
> during a next meeting.
Of course ko rules are hardest (once you have decided on area counting).
>
> Votes
>
> Numbers of votes are stated as Y:N for Yes versus No or as Y:N:A for Yes
> versus No versus Abstention. Some of the 100% one-sided votes were made by
> implicit general agreement. To ease reading, this report translates all
> voting issues to let the Yes votes appear to have won, even though the
> actual issue might have been formulated as its negative during the
> meeting. If, in the list below, an issue uses a negation, then a Yes vote
> confirms that negation.
It may ease reading, but allows for a considerable amount of confusion.
In particular (with the closer votes) it is not clear whether
the proposer hoped that the yes vote would win.
>
> Kirschner did not consider himself to have another AGA vote and therefore
> did not vote at all while the three Ing Foundation delegates all
> participated in the voting, except when Ying was absent. Because most
> votes were heavily one-sided anyway, this did not matter and so a request
> to adjust the actual voting system was not made during this meeting to
> avoid wasting time on that technical matter.
I would hope that the question of representation could be sorted out
before getting to points where it does matter.
>
> (1) The ruleset's name is "Unified Area Rules". 8:0.
> (2) Let us not
> talk about Ikeda Rules during this meeting. 6:1:1
> (No = Jasiek, Abstention = Wu).
Ikeda rules are only necessary if a compromise is sought with Japanese and
Korean rules. By not talking about them you are implicitly ruling out
such a compromise.
> (3) The scoring is Area Scoring, i.e. at the end a player's
score
> is the number of his stones on the board plus empty intersections
> surrounded by only his stones. 8:0.
> (4) Suicide is prohibited. 4:3:1 (Yes = Jasiek, Tang, Jin, Wu; No
> = Ying, Yang, Ni; Abstention = Benson).
A close vote: Ing versus the possibility of including the Japanese and
Koreans. Not important enough to fight over I hope. (I would have voted
NO).
> (5) The players alternate. 8:0.
> (6) Black starts the alternation. 8:0.
> (7) The available types of
> moves are board-plays and pass-plays.
> 8:0.
> (8) Pass-plays are always allowed. 8:0
> (9) After 2 successive
> passes and if the players agree, then the
> game can end. After any 2 or more successive passes, an agreement phase
> is possible. After exactly 4 successive passes, the game ends. In the
> previous conditions, numbers of passes in a succession of passes are
> counted regardless of whether they might have been interrupted by one or
> more than one agreement phase. 6:1 (No = Jasiek).
This sounds rather cumbersome, and I doubt that it would be followed in
professional games. Can you give your reasons for your no vote?
> (10) If expressed in full counting and intended for the 19x19
> board, the size of komi should be 7.5 or 8 (Black wins ties). 7:0.
> (11) The following ways of expressing the komi are possible:
> "3.75 points [in half counting]", "7.5 points [in full counting]", "8
> points (Black wins ties) [in full counting]". 7:0.
I wonder that this needed to be so explicitly stated. I would have
preferred to see it specified as komi should be 3.5 (7 points in full
counting) but that white win ties at the discretion of the event
organiser. If these rules are to be used in swiss tournaments the ties
should not be ruled out, even if they are undesirable in a knockout.
> (12) Every counting procedure for Area Scoring is possible. A
> tournament should announce its used counting procedure. 7:0.
OK; In tournaments I organise we will count by throwing a pair of dice. :)
I understand why they didn't want to list the available methods, but they
should at least have specified what features a counting system should have.
> (13) In case of language problems between the players, it is a
> tournament organization's (but not the rules') task to suggest methods
> for resignation or for an agreement phase. 8:0.
> (14) Handicap stones may be placed on freely chosen
> intersections. 8:0.
>
> Open Urgent Tasks
>
> write ruleset
> bring English and Chinese language in ruleset in agreement discuss,
> agree on ko rules; for long cycles find a compromise
> between Chinese Ko Rules and Long Cycle Ko Rules
> after the rules will be finished, submit them to superiors and
> then the IGF commentary on ruleset, in particular on its ko rules
> explanations of all counting procedures for Area Scoring
--
Barry Phease
Robert Jasiek
wrote:
>the Japanese and Korean delegations should have
>dropped out. Is it likely that they will rejoin for a future meeting?
It is their own decision. I cannot predict that. I have expressed and
continue to express my hope though.
>It may ease reading, but allows for a considerable amount of confusion.
>In particular (with the closer votes) it is not clear whether
>the proposer hoped that the yes vote would win.
Do you think that something particular about the votes in my report is
confusing?
>I would hope that the question of representation could be sorted out
>before getting to points where it does matter.
Well, during the first meeting, a voting system had been agree upon.
However, that was when also votes for Japan and Korea were assigned.
So it is not absolutely clear whether everybody still considers the
same vote distribution fair. I do.
>Ikeda rules are only necessary if a compromise is sought with Japanese and
>Korean rules. By not talking about them you are implicitly ruling out
>such a compromise.
Not exactly. We have decided not to talk about them DURING THIS
MEETING. Later we might reconsider them again.
>> (9) After 2 successive
>> passes and if the players agree, then the
>> game can end. After any 2 or more successive passes, an agreement phase
>> is possible. After exactly 4 successive passes, the game ends. In the
>> previous conditions, numbers of passes in a succession of passes are
>> counted regardless of whether they might have been interrupted by one or
>> more than one agreement phase. 6:1 (No = Jasiek).
>
>This sounds rather cumbersome, and I doubt that it would be followed in
>professional games. Can you give your reasons for your no vote?
Because I think that rules should be simple. My proposal (see report)
would have made it simpler. We two seem to agree here, but it was the
losing vote.
>> (11) The following ways of expressing the komi are possible:
>> "3.75 points [in half counting]", "7.5 points [in full counting]", "8
>> points (Black wins ties) [in full counting]". 7:0.
>
>I wonder that this needed to be so explicitly stated.
Maybe experience will tell whether it should. Presumably it goes into
a footnote of the rules.
>I would have
>preferred to see it specified as komi should be 3.5 (7 points in full
>counting) but that white win ties at the discretion of the event
>organiser. If these rules are to be used in swiss tournaments the ties
>should not be ruled out, even if they are undesirable in a knockout.
Certainly a good point. Maybe it will be considered later when
tournament rules (and in particular possible tournament systems) will
be discussed.
>> (12) Every counting procedure for Area Scoring is possible. A
>> tournament should announce its used counting procedure. 7:0.
>
>OK; In tournaments I organise we will count by throwing a pair of dice. :)
>I understand why they didn't want to list the available methods, but they
>should at least have specified what features a counting system should have.
The rules' wording will be bothered with later. So far we have made
only the basic decisions about their contents. The essential criterion
here is: "for Area Scoring". So throwing a pair of dice does not fit
:)
--
robert jasiek
Bill Spight
Robert:
>> (1) The ruleset's name is "Unified Area Rules". 8:0.
>> (2) Let us not
>> talk about Ikeda Rules during this meeting. 6:1:1
>> (No = Jasiek, Abstention = Wu).
Barry:
> Ikeda rules are only necessary if a compromise is sought with Japanese and
> Korean rules. By not talking about them you are implicitly ruling out
> such a compromise.
At least for the moment. However, either Button Go or an Ikeda style
area rule remains a compromise between area and territory scoring.
Starting from an agreed upon set of area rules makes the compromise easier
from a technical standpoint. Excluding the Japanese and Koreans may make
the politics more difficult.
Best,
Bill
Simon Goss
>OK; In tournaments I organise we will count by throwing a pair of dice.
>:)
As you have just recommended 7 komi, is this actually useful? :)
--
Simon
UK Go Challenge for schools
http://www.ukgochallenge.com
Bill Spight
Barry:
>>OK; In tournaments I organise we will count by throwing a pair of dice.
>>:)
Simon:
> As you have just recommended 7 komi, is this actually useful? :)
Seven is the proper komi for a dice throw. ;-) If you don't want ties,
throw the dice again if a seven comes up.
Best,
Bill
-
Rule #1: Eight individuals can determine "the International
Go Rules" for the remainder of the world, and for all nations.
Rule #2: Korea and Japan need not apply for any "rules input."
Rule #3: Discussion about Ikeda Rules is -prohibited- here.
Rule #4: A non-voting member will be allowed to attend the
IGRF, give others input, but not be responsible for voting.
Rule #5: One-eyed points in seki will be considered territory
because the committee has adopted "Area Scoring." This
precludes any need "to demonstrate life in life shapes," in
order to establish -qualifications- to possess territory, and
imposes a _particular_slant_ upon the interpretation of Go.
Rule #6: Three members plus one abstention consider that
suicide cannot be prohibited. Does the narrow majority, not
really a conclusive vote, accomplish prevention by sedation?
Rule #7: Robert Jasiek is alone in the world, who thinks that
game ending conditions ought not to be made definitive.
Rule #8: Handicap stones may not be confined to star points.
Means that IGRF forces such future international tournaments
to adopt only the standards that they wish to dictate about this.
Rule #9: It takes only one person (Ying) to declare that "Olympic
Rules" requires merely some agreement by the IGRF.
Rule #10: Rules based in territory scoring are (erroneously)
called "Japanese Rules" even though Sidney Yuan has
observed that territory scoring rules had originated in China.
Rule #11: The word "international" annoys Japanese and Koreans.
Rule #12: After lunch, Robert Jasiek can mention "Ikeda Territory
Scoring" even though the IGRF had precluded this before lunch.
Rule #13: Everyone appears to agree that referees decide when
disagreement occurs, yet Chris Kirschner can tell us what "US
Culture" is, to the contrary.
Rule #14: Sensible alternatives to the IGRF process will -not- be
considered. Such as "The `Tournament Director / Referee'
shall announce the rules before Round One at each general
introductory player meeting. Questions will be answered then."
Rule #15: Bill Spight's "button go" proposal had not been discussed.
A lot of rules. Appears that IGRF doesn't understand Go very well.
- regards
- jb
=============================================
Robert Jasiek <jas...@snafu.de> wrote:
> 6th Meeting of the International Go Rules Forum (IGRF)
> Report by Robert Jasiek
> ...
edse
Robert Jasiek skribis:
> The essential criterion here is: "for Area Scoring". So throwing a pair of dice does not fit
> :)
Well, then let's count both the number of dices and the points thrown
out.
:))
--
edse
Robert Jasiek
> Rule #3: Discussion about Ikeda Rules is -prohibited- here.
Would anybody want to apply that rule? I don't.
> Rule #7: Robert Jasiek is alone in the world, who thinks that
> game ending conditions ought not to be made definitive.
If you should not just be satirical, this is wrong. My preference for
that has been as in
http://home.snafu.de/jasiek/siming.html
> Rule #8: Handicap stones may not be confined to star points.
It is prohibited (to try) to pile 9 stones one over another on the
same star point.
> Sidney Yuan has
> observed that territory scoring rules had originated in China.
This is an interesting theory.
> Rule #15: Bill Spight's "button go" proposal had not been discussed.
Not during the 6th meeting, but during the 3rd I mentioned it.
--
robert jasiek
-
Robert Jasiek <jas...@snafu.de> wrote:
> Would anybody want to apply that rule? I don't.
(2) Let us not talk about Ikeda Rules during this meeting.
6:1:1 (No = Jasiek, Abstention = Wu).
I see now that your report of voting preceeds discussion of it.
I rescind Rule #12. :-)
>> Rule #7: Robert Jasiek is alone in the world, who thinks that
>> game ending conditions ought not to be made definitive.
> If you should not just be satirical, this is wrong.
> My preference for that has been as in
> http://home.snafu.de/jasiek/siming.html
There was no mention of "game end" in the above URL.
No specification that scoring or result produces "game end."
-----------------
Jasiek: [He asks for clarification of details for vote (9).
In particular, he clarifies and Benson confirms that the
following shall be possible: Pass, pass, agreement phase,
pass, agreement phase, etc.]
(9) After 2 successive passes and if the players agree, then
the game can end. After any 2 or more successive passes,
an agreement phase is possible. After exactly 4 successive
passes, the game ends. In the previous conditions, numbers
of passes in a succession of passes are counted regardless
of whether they might have been interrupted by one or more
than one agreement phase. 6:1 (No = Jasiek).
The intent of (9) was to -discourage- "pass, pass, agreement
phase, pass, agreement phase, etc." etc. etc. etc.
>> Rule #8: Handicap stones may not be confined to star points.
> It is prohibited (to try) to pile 9 stones one over another on the
> same star point.
I did not see that Rule. Nor have I seen more than a half
dozen regulation-style stones piled on top of each other,
irrespective of placement on a star point or non-star point.
>> Sidney Yuan has
>> observed that territory scoring rules had originated in China.
> This is an interesting theory.
It is more than an interesting theory. It states that we are being
incorrect to label a Ruleset according to modern geographical or
nationalistic preference. A discussion, for IGRF purposes, concerns
the "Area Rules" or "Territory Rules"; not "Japanese Rules" or
"Korean Rules" or "Chinese Rules" or "Timbuktu Rules."
>> Rule #15: Bill Spight's "button go" proposal had not been discussed.
> Not during the 6th meeting, but during the 3rd I mentioned it.
I suppose that suffices, then?
Rule #16: AGA funds might be paying for Terry Benson's travel costs.
Rule #17: More travel costs could be forthcoming as the years drag on.
- regards
- jb
-------------------------------------------------------------
Germany Must Pay Black Drug Dealer
http://www.nationalvanguard.org/story.php?id=9551
-------------------------------------------------------------
Robert Jasiek
>>> Sidney Yuan has
>>> observed that territory scoring rules had originated in China.
If so, where is the historical evidence?
>>> [button go]
> I suppose that suffices, then?
I would have liked to see discussion following my short explanation of
it.
--
robert jasiek
Makrai Jozsef
> (10) If expressed in full counting and intended for the 19x19
> board, the size of komi should be 7.5 or 8 (Black wins ties). 7:0.
Why must the komi be fixed, and not left open for tournament rules?
> Jin, Tang: No Suicide is closer to Japanese rules.
> Jin, Tang, Ying: No Suicide is an unnecessary restriction.
> Jin: The Chinese Go Association is strict about having Area
> Scoring and a playout. At a time, Chinese Rules had Suicide, recently
> they have No Suicide.
> Kirschner: We should try to identify what we think is the best,
> except in minor things.
> Jasiek: He considered wording of the case of one's own string
> being without liberty. The solutions Suicide / No Suicide are equal
> options in this regard.
I'd consider not only wording but the number of logical concepts:
1) strings without liberties are removed as prisoners
2) removal of opposing strings has priority
Both concepts are necessary for a simple ko capture (even if suicide is
forbidden). No further concept is necessary for suicide. A third
concept is necessary for no suicide.
(OC, I understand political issues are more important.)
> Tang, Jasiek: [They show flow charts to illustrate possible game
> ending procedures.]
> Jasiek: He suggests the following procedure: Either there are 2
> passes, the players' agreement, and then the scoring or there are 2
> passes, the players' disagreement, the playout, and then the scoring.
> Benson: A second possible agreement should be possible. Therefore
> play should end by 4 successive passes, even if there is some
> (dis)agreement phase in between. As a consequence of 4 successive
> passes leading to the final end, infinite loops are impossible.
Here double ko seki may need attention.
> All: [Discussion about Superko. Almost it is adopted until
> everybody becomes aware that the Chinese Rules do not mean [Superko]
> when they say so.]
;o)
> China, Yang: We dislike Superko.
> Benson, Kirschner, Jasiek: We like Superko.
> All: [Although a vote is not done, it becomes clear that pure
> Superko drops out as a possible ko rule for the unified ruleset.]
I don't think my ko rules would be appreciated here, but I'm curious
what else could be used for even a remote approximation of Chinese ko.
> Jasiek: There is an easy classification [as a means of improving
> the Chinese Ko Rules]: During a cycle, there is either a zero or a
> non-zero difference of removed black and white stones. For each case,
> one can use a rule: Either the game is a tie [or the player with fewer
> removed stones wins]. This is easy, applicable, and the professionals
> do not need to worry about losing due to a too difficult ko rule.
But how about moonshine life? Is it possible that it will be allowed
by the new ruleset?
József
dnc AT fw DOT hu
-
Almost 12 years ago makes it historical evidence for this newsgroup:
"The 'Japanese Rules' were invented by the Chinese who introduced
them into Japan during the Nanbei Dynasty (420-589 AD)." - Mr. Yuan
http://groups.google.com/group/rec.games.go/msg/a52f214a6c68bb5b?hl=en&
"Makrai Jozsef" <nos...@freeweb.hu> wrote:
> Why must the komi be fixed, and not left open for tournament rules?
They're trying to nail down "International Go Rules" for a Go Olympics.
- regards
- jb
------------------------------------------------------------------
Russia Rejects UN Authority on Kosovo
http://www.nationalvanguard.org/story.php?id=9550
------------------------------------------------------------------
Robert Jasiek
wrote:
>> (10) If expressed in full counting and intended for the 19x19
>> board, the size of komi should be 7.5 or 8 (Black wins ties). 7:0.
>Why must the komi be fixed, and not left open for tournament rules?
The basic idea is to set the standard komi for all standard
tournaments: professional, world championships, Olympics,... I guess
that amateurs may still have fun in specialized tournaments for
special purposes.
>I'd consider not only wording but the number of logical concepts:
>
>1) strings without liberties are removed as prisoners
>2) removal of opposing strings has priority
>
>Both concepts are necessary for a simple ko capture (even if suicide is
>forbidden). No further concept is necessary for suicide. A third
>concept is necessary for no suicide.
In a sense, there is a point to this, but can you tell in general
which systems of "basic" logical concepts could exist? Presumably,
yours is not the only possible system.
>> Benson: A second possible agreement should be possible. Therefore
>> play should end by 4 successive passes, even if there is some
>> (dis)agreement phase in between. As a consequence of 4 successive
>> passes leading to the final end, infinite loops are impossible.
>
>Here double ko seki may need attention.
Yes.
>I don't think my ko rules would be appreciated here, but I'm curious
>what else could be used for even a remote approximation of Chinese ko.
Let us know when you come up with something noteworthy.
>But how about moonshine life? Is it possible that it will be allowed
>by the new ruleset?
Surely allowed, but I think that it becomes a long cycle tie now :)
--
robert jasiek
Makrai Jozsef
>> I'd consider not only wording but the number of logical concepts:
>>
>> 1) strings without liberties are removed as prisoners
>> 2) removal of opposing strings has priority
>>
>> Both concepts are necessary for a simple ko capture (even if suicide is
>> forbidden). No further concept is necessary for suicide. A third
>> concept is necessary for no suicide.
>
> In a sense, there is a point to this, but can you tell in general
> which systems of "basic" logical concepts could exist? Presumably,
> yours is not the only possible system.
Certainly, I wouldn't claim that. I just feel there are some elemental
concepts that are hard to replace with something else, even in a
different system. I would expect the above two to be present in most
logical rulesets (if maybe in a less explicit form, hidden behind
wording).
For a truly different system to exist, we should be able to find some
other concept that is neither decomposable into smaller concepts, nor
1:1 replaceable with something else, and is at least partially
redundant/overlapping with earlier "elemental" concepts.
>> I don't think my ko rules would be appreciated here, but I'm curious
>> what else could be used for even a remote approximation of Chinese ko.
>
> Let us know when you come up with something noteworthy.
Since I still think the optimal area model is "finite stone supply"
plus "ko-challenging", it would be hard for me to look for anything
else. ;o) And because of the politics involved, I wouldn't try to
suggest anything, I'm merely curious about the outcome.
Anyway, in your report superko playout is mentioned, among other
things. Was there a complex suggestion of having a normal ko game plus
a superko playout, or did I misunderstand?
>> But how about moonshine life? Is it possible that it will be allowed
>> by the new ruleset?
>
> Surely allowed, but I think that it becomes a long cycle tie now :)
Would the Chinese delegate accept this? Even C88 lists it as a
precedent, after all. ;o)
Robert Jasiek
wrote:
>in your report superko playout is mentioned, among other
>things.
Which text are you referring to?
>Was there a complex suggestion of having a normal ko game plus
>a superko playout
Not really.
>Would the Chinese delegate accept this? Even C88 lists it as a
>precedent, after all.
They are aware that their own ko rules are flawed although they are a
bit hesitating about admitting it too often:) They are proud of the
Chinese rules being without special rulings while they cannot quite
admit to themselves yet that their referee ko rules ARE special
rulings. In other words, they will move but I do not know exactly how
far. Mr. Yang was more worried about Moonshine-life than the Chinese
delegates though.
--
robert jasiek
Makrai Jozsef
>> Was there a complex suggestion of having a normal ko game plus
>> a superko playout
>
> Not really.
(Later.) This may worth a closer look. For example:
area scoring, "finite stone supply" (or similar), and
"superko in resumption only, with passes lifting bans".
This would get at least the basic rulings right (cycles, moonshine,
send2) - quite an achievement. ;o) The problem is superko is dangerous
even in resumption. There are cases where it could be abused, even with
passes lifting bans (eg. bent-4 + double ko seki seems to break).