pass too early (rules question)
Mathijs
What do the various rulesets say about this situation:
White passes, black passes. While counting, the players see that a white
groups can be killed by adding one black stone. It can be saved by
adding a white stone.
Black obviously claims it's dead. Is he right about that?
Thanks,
M.
Denis Feldmann
Certainly not. "dead" (in all rulesets using the word) means "dead
against any defense, even when the defending player starts.
Now, the rulesets are in some disagreement about the situation you
described. If you are playing formally J89 (which probably means you are
a Japanese pro playing a tournament game), and if the result of the game
would depend of the capture (or not) of White group, then *both players
have lost*. This, I believe, is *not* what you wanted to know
On the other hand, even with Chinese-like rulesets (as AGA), the
situation you describe can arise because of the usual "agreement phase"
used for quick counting ; obviously in case of dispute in this not
completely official phase, the "official" solution would be to restart
the game at the point before the last pass exchange. But this creates
some real refereeing problems (must one restart the clock ? is this not
unfair to one of the players (and which one?) etc...
So : a) If you are a beginner or DDK player, etc : dont bother too much
with all that, decide the game a draw if it was a friendly one, take
advantage of what KGS (say) propose (or suffer the loss with a smile)
otherwise
b) If you are a reasonably strong player, have to enter a competiton,
and want to know what will happen in case of disagreement (say, more
reasonably, a bent four) , ask the referee : he should know...
c) If you are a very strong player, a pro, or Robert, you must be
trolling, right ?
>
> Thanks,
> M.
Mathijs
Hi Denis,
Thanks for your answer.
> Now, the rulesets are in some disagreement about the situation you
> described. If you are playing formally J89 (which probably means you
> are a Japanese pro playing a tournament game), and if the result of
> the game would depend of the capture (or not) of White group, then
> *both players have lost*. This, I believe, is *not* what you wanted to
> know
Well, actually yes. Or... Let me tell you about the game: it's an online
tournament game. We played too fast/careless and both missed the game
deciding atari. Because it's a tournament play, the result will
influence other players (and our own chances). That's why I'm looking
for the official rule (if there's something like that in at all), so we
can solve this issue in a way we can live with. The server allows
continueing the game after a pass (or two passes).
Unfortunately, there is no referee who can decide on it and we have to
find a good solution ourselves.
So, if this happens in the Honinbo final, the game won't count?
Unfortunately, this is not a server-option...
Thanks again,
Mathijs.
Denis Feldmann schreef:
Robert Jasiek
wrote:
>it's an online tournament game.
To get an answer, don't ask for all rulesets' rulings (it would take
weeks to answer) but please specify the go server, the used rules of
play, the tournament rules, and possibly any agreements between the
two players before the start of the game!
Mathijs
http://www.littlegolem.net/jsp/games/gamedetail.jsp?gtid=go19&page=rules
"On this server is go played according to Japanese rules. Komi is:
# Go 9x9 - 5.5 komi
# Go 13x13 - 8.5 komi
# Go 19x19 - 6.5 komi"
No agreements between players.
I assume this is not really useful for a rules-purist, that's why I
didn't mention it before...
Besides finding a way to elegantly solve this issue, I'm curious about
other rulesets.
Does anyone know who this issue would be solved if it occured under:
1. Japanese rules (A: both players lose)
2. EGF: ?
3. AGA/Chinese (A: restart game before last pass exchange)
Based on those answers, I hope to make a good and fair decision.
pieter mioch
Mathijs wrote:
> Hi,
>
> What do the various rulesets say about this situation:
>
> White passes, black passes. While counting, the players see that a white
> groups can be killed by adding one black stone. It can be saved by
> adding a white stone.
You write *while counting the players see...* this reads to
me as if both players thought the white group was alive -at
the moment of ending the game by consecutive passes- I'd
count the game on that presumption (that the stones are alive).
>
> Black obviously claims it's dead.
Do you mean "obviously" or "cunningly" :-)
Is he right about that?
Do you mean "does he have the right to claim that?" or "is
his claim right?
>
> Thanks,
> M.
Denis Feldmann
>
> Hi Denis,
>
> Thanks for your answer.
>
> > Now, the rulesets are in some disagreement about the situation you
> > described. If you are playing formally J89 (which probably means you
> > are a Japanese pro playing a tournament game), and if the result of
> > the game would depend of the capture (or not) of White group, then
> > *both players have lost*. This, I believe, is *not* what you wanted to
> > know
>
> Well, actually yes. Or... Let me tell you about the game: it's an online
> tournament game. We played too fast/careless and both missed the game
> deciding atari. Because it's a tournament play, the result will
> influence other players (and our own chances). That's why I'm looking
> for the official rule (if there's something like that in at all), so we
> can solve this issue in a way we can live with. The server allows
> continueing the game after a pass (or two passes).
>
> Unfortunately, there is no referee who can decide on it and we have to
> find a good solution ourselves.
>
> So, if this happens in the Honinbo final, the game won't count?
No, this is even worse : both players have lost, so , presumably, they
can only pretend now *both* for second place, the title goes unawarded,
and the Yomiuri will keep the first-prize money for themselves :-)
Robert Jasiek
wrote:
>>>
White passes, black passes. While counting, the players see that a
white
groups can be killed by adding one black stone. It can be saved by
adding a white stone.
Black obviously claims it's dead. Is he right about that?
<<<
>http://www.littlegolem.net/jsp/games/gamedetail.jsp?gtid=go19&page=rules
>"On this server is go played according to Japanese rules. [...]
>No agreements between players.
Of course, "according to Japanese rules" is ambiguous because there
are the Japanese 1989 Rules, the World Amateur Go Championship Rules,
and verbal Japanese rules. Since most players on littlegolem would not
know or even understand the fine print of the written Japanese
rulesets, we may as well assume that verbal Japanese rules apply.
In your case, it is a simple atari or defense against it, so we do not
need to worry about life and death definitions. We need to worry about
a ruling for unstable life and death status (verbal Japanese rules are
not clear enough to have such) and a ruling about resumption (verbal
Japanese rules are not clear enough to have such but also do not
prohibit it).
Depending on whether one allows resumption, either the one or the
other player gets the advantage. This would be unfair. We also cannot
rule unstable life and death status (not provided as discussed above),
so ruling independently of the existence of resumption issue is not
possible, either.
Thus a fair ruling has to be: It cannot be decided whether resumptions
are available. This is neither player's fault. Therefore each player
should be treated equally. There are the following options for equal
treatment: default tie, default no result, default loss of each
player, replaying the game (starting a new game afresh). Since the
tournament has not specified default no result or replay, these drop
out. Since it is neither player's fault that the rules are unclear,
default loss of each player drops out. This leaves default tie as the
probably best solution.
However, if the overlooked atari does not affect the winner (one
player is still ahead enough), there is another possibility: declare
the clearly ahead player the default winner. However, such a decision
(to made by a referee) still somewhat contradicts the fact that also
it is not the opponent's fault that the rules are unclear. In a second
instance court, I would accept a referee's default win declaration in
such a case because it is reasonable enough (and default tie is
acceptable anyway) but as a first instance referee I would go for
default tie right away.
Summary: Let the game be a default tie!
>Besides finding a way to elegantly solve this issue
All rulesets with Area Scoring provide an elegant solution: Play it
out on the board if the players don't agree.
Robert Jasiek
<pmi...@xx.xxxx.ne.jp> wrote:
>You write *while counting the players see...* this reads to
>me as if both players thought the white group was alive -at
>the moment of ending the game by consecutive passes- I'd
>count the game on that presumption (that the stones are alive).
Even under verbal Japanese rules, the players are not allowed to break
the rules. Mistakenly assigning a false life status breaks the rules.
Planar
Mathijs <mat...@ongeldig.nl> wrote:
> Well, actually yes. Or... Let me tell you about the game: it's an online
> tournament game. We played too fast/careless and both missed the game
> deciding atari. Because it's a tournament play, the result will
> influence other players (and our own chances). That's why I'm looking
> for the official rule (if there's something like that in at all), so we
> can solve this issue in a way we can live with. The server allows
> continueing the game after a pass (or two passes).
As you described it initially, I would say it's too late to claim
the stones are dead: if you are at the counting stage, the
players have already agreed about dead stones and removed them,
and everything left on the goban is alive.
Since it's an online game, I presume you are in fact still at
the agreement stage, and the computer is counting
incrementally (displaying what would be the score at any
time if the agreement stage were to stop there).
You say the server allows continuing the game after two passes,
That's what you should do. Who plays first? I guess the
computer will decide this for you. Too bad for the other
player, but he shouldn't have passed. Getting your mistakes
punished is exactly what Go is about.
--
Planar
remove .invalid from my address to send me mail
"Surprise. Then just use AGA rules and that's it." - Robert Jasiek
Nick Wedd
<feldmann.den...@neuf.fr> writes
>Mathijs a écrit :
>> Hi Denis,
>> Thanks for your answer.
>>
>> > Now, the rulesets are in some disagreement about the situation you
>> > described. If you are playing formally J89 (which probably means you
>> > are a Japanese pro playing a tournament game), and if the result of
>> > the game would depend of the capture (or not) of White group, then
>> > *both players have lost*. This, I believe, is *not* what you wanted to
>> > know
>> Well, actually yes. Or... Let me tell you about the game: it's an
>>online tournament game. We played too fast/careless and both missed
>>the game deciding atari. Because it's a tournament play, the result
>>will influence other players (and our own chances). That's why I'm
>>looking for the official rule (if there's something like that in at
>>all), so we can solve this issue in a way we can live with. The
>>server allows continueing the game after a pass (or two passes).
>> Unfortunately, there is no referee who can decide on it and we have
>>to find a good solution ourselves.
>> So, if this happens in the Honinbo final, the game won't count?
>
>No, this is even worse : both players have lost, so , presumably, they
>can only pretend now *both* for second place, the title goes unawarded,
>and the Yomiuri will keep the first-prize money for themselves :-)
Unusually, I find myself disagreeing with Denis.
He is I assume referring to Article 13.1 of the J1919 rules, which
reads, in English translation,
" Article 13 (both players lose)
" 1. After the game stops according to Article 9, if the players find
" an effective move, which would affect the result of the game, and
" therefore cannot agree to end the game, both players lose".
But in this case, Article 13.1 does not apply. It is not the case that
"the players .. cannot agree to end the game". They have agreed to end
the game. They just want to know how to score it.
Article 7.1 provides the definitions we need to do this. "Stones are
alive if they cannot be captured... Stones which are not alive are said
to be dead."
In this case we have an unsettled white group. By Article 7.1 it is not
alive, and is dead. The white stones should be removed from the board.
Nick
--
Nick Wedd ni...@maproom.co.uk
T Mark Hall
<jas...@snafu.de> writes
However, there is a difference here; *both* players agreed that the game
was over and then later found a sequence to change the status of that
agreement. The players should be able to resume if there are unplayed
liberties, a ko that should have been connected or stones that needed to
be captured by a further necessary move. But, spotting that further
moves *could* be made after they have agreed that the game was over,
does not fall into this category. The players had passed, the game was
over and, whatever the score, that must remain the result of the game.
Best wishes.
--
T Mark Hall
Honorary Vice-president, British Go Association
http://www.gogod.co.uk/index.htm
http://www.gogod.co.uk/NewInGo/NewInGo.htm
Robert Jasiek
<tm...@gogod.demon.co.uk> wrote:
>*both* players agreed that the game
>was over
Sure, but the question is what they were meaning to do by agreeing on
a game end. It does not necessarily need to be the same as a J1989
"game end".
>The players should be able to resume
Why? If the server generally allows resumption, one may either
consider this feature to be an implicit tournament rule or argue that
"Japanese rules" leave it unclear whether resumption is available.
>if there are unplayed
>liberties, a ko that should have been connected or stones that needed to
>be captured by a further necessary move.
Why should the presence of such by itself be enough cause for
resumption? I mean, how do you conclude this from "Japanese rules are
used"?
Denis Feldmann
"and therefore" is the all important point, imho
>
> But in this case, Article 13.1 does not apply. It is not the case that
> "the players .. cannot agree to end the game".
Up to "anf therefore", the article applies perfectly
They have agreed to end
> the game.
No, they have agreed to stop it :-)
They just want to know how to score it.
But obviously, they are of different "opinion" of what the score is, so
they cannot score, so the game is not ended. If I recall correctly thez
rules, scoring *is* done by the players, and is part of the game. To go
back to one of your own articles, the question of what must be done if
the players are not able to count (innumeracy) or made the same counting
mistake, declared the wrong winner, and discovered after registering the
win they were mistaken are hard ones, which obviously are not covered
by J89, but...
>
> Article 7.1 provides the definitions we need to do this. "Stones are
> alive if they cannot be captured... Stones which are not alive are said
> to be dead."
>
> In this case we have an unsettled white group. By Article 7.1 it is not
> alive, and is dead.
Antiseki is not far away...
T Mark Hall
<jas...@snafu.de> writes
I conclude nothing from Japanese rules. What I conclude is what I
consider common sense. I said that the players *should* be able to
resume; this would apply whether the players are at a tournament, club
or on a server, especially if the game has not been properly concluded.
What appears to have happened here is that the players saw that further
play could be made (not *must* be played to complete the game) after the
normal number of passes to finish the game. I can foresee the situation
where the players agree the game is over by passing, then one player
spots something that he/she had previously missed and he/she demands
that the game be resumed because he/she considered that the game had not
been completed. On some servers, repeated undos are treated as
resignation. Are we to have the same (or opposite) problem with repeated
requests to resume?
The players passed; one then saw further moves but had already agreed
the game was over. A group that is dead as it stands (not requiring any
further move to kill it) should be removed, but all other groups should
remain on the board and be counted. I leave it to the rules zombies to
phrase it in a way that fits tournament and server rules. :)
-
Don't understood why this is a difficult question for Go Players.
"Mathijs" wishes to ask what "various rulesets" have to say, but the
problem concerns what is feasible on a particular Go Server. In
a dispute situation the region is regarded as "seki" unless one of
the players can demonstrate otherwise. After "pass-pass", under
resumption rules, a player to ask for "kill" must allow the defending
player to mover first. If a defending player asks for "live" (instead
of seki) then the player to kill is allowed to move first. Otherwise
they agree on "seki" and score the game. Yet a particular Server
might not conform some sensible resolution, and must be repaired !
None of the Rulesets are purposed to make Go unenjoyable.
"Robert Jasiek" <jas...@snafu.de> wrote:
> Even under verbal Japanese rules, the players are not allowed to break
> the rules. Mistakenly assigning a false life status breaks the rules.
I am surprised, at this late date, that rules maven "Robert Jasiek"
labors under misconceptions. The status of a group is assigned by
player agreement. If outside parties intervene then the players are
advantaged by additional perspectives to properly assign the status.
If the players have agreed then there is no mistaken assignment of
status with regards to any group on the board.
"Nick Wedd" <ni...@maproom.co.uk> wrote:
> Article 7.1 provides the definitions we need to do this. "Stones are
> alive if they cannot be captured... Stones which are not alive are said
> to be dead."
>
> In this case we have an unsettled white group. By Article 7.1 it is not
> alive, and is dead. The white stones should be removed from the board.
More disinformation from "Nick Wedd." Resumption rules stipulate
that a player to capture first allows a defending player the first move.
Unsettled groups are "seki" unless they are demonstrated otherwise.
"T Mark Hall" <tm...@gogod.demon.co.uk> wrote:
> ... But, spotting that further
> moves *could* be made after they have agreed that the game was over,
> does not fall into this category. The players had passed, the game was
> over and, whatever the score, that must remain the result of the game.
"T Mark" (Honorary Vice-president, British Go Association) has
completely denied players the perogative of request for resumption.
It's very surprising to find so many Go Players in abysmal ignorance.
In his subsequent remark, "T Mark" illustrates his own ignorance of
protocol for request of resumption of play: other player moves first.
I don't understand how we're going to obtain the correct information
unless parties to this discussion first agree to become serious.
- regards
- jb
-----------------------------------------------------------------
Twilight of the Psychopaths
http://www.agoracosmopolitan.com/home/Frontpage/2008/01/02/02073.html
-----------------------------------------------------------------
Mef
wrote:
> Nick Wedd a écrit :
>
>
>
> > In message <fm2h9e$6d...@netfinity.fr>, Denis Feldmann
> > <feldmann.denis.asuppri...@neuf.fr> writes
Unless I'm mistaken, the "...and therefore..." is in reference to
9.3, which gives a provision for either player to resume. There is
the implicit assumption in 13.1 that neither player is willing to
resign the game by giving their opponent the first move via
resumption, therefore they both lose. As I understand it, at the
point the game stops and a gainful play is found there are 3
possibilities: 1) It affects the score, but not the outcome, so the
player who is winning will call for the game to resume in order to
claim his victory. 2) It affects the outcome, and one of the two
players decides to "take one for the team" so to speak, and allow his
opponent to claim the win via resuming, or 3) Neither player is
willing to give the win to the opponent, so they both lose.
Cheers,
Mef
Robert Jasiek
<tm...@gogod.demon.co.uk> wrote:
>What I conclude is what I consider common sense. I said that the players
>*should* be able to resume
I do not consider it an aspect of common sense but of the nature of
the game: It is a player's right to make strategic mistakes!
From that I conclude more specifically
a) It is a player's right to make strategic mistakes during
board-plays.
b) It is a player's right to make strategic mistakes by passing.
So I think that the players should not be able to resume for the
purpose of correcting strategic mistakes.
***
(For reference only: The Korean 1992 Rules prescribe perfect passes.
The WAGC Rules have partially perfect passes. J1989 have abandoned
such.)
Robert Jasiek
> If the players have agreed then there is no mistaken assignment of
> status with regards to any group on the board.
Some people dig out a skeleton. They agree that that human being is
alive. By virtue of their agreement of it being alive, it is alive. Is
it?
Mike
leaves an 'enclosure' not complely enclosed. Now there are no "dead" stones to be removed, but if the enclosure
nominally contains for example 6 points of the form:
WWWWB
W WB
W B
W WB
WWWWB
then:
1) with no more moves, W presumably cannot claim any of the 6 points as they are not truely 'enclosed' by white.
2) with a white move
WWWWB
W..WB
W..WB
W..WB
WWWWB
w can get 6 points, and
3) if B moves first
WWWWB
W..WB
W.WBB
W..WB
WWWWB
then w gets 5 points.
Mike
-
> jazze...@hotmail.com (-) wrote:
>> If the players have agreed then there is no mistaken assignment
>> of status with regards to any group on the board.
Robert Jasiek <jas...@snafu.de> wrote:
> Some people dig out a skeleton. They agree that that human being
> is alive. By virtue of their agreement of it being alive, it is alive. Is it?
As I have stated, the players may receive the benefit of external
advice. And, having received such advice, when the players agree
on the status of their groups there is no mistaken assignment of status.
- regards
- jb
---------------------------------------------------------------------------------
Raud the Strong is honored by many each January 9 for his refusal
to abandon his Old Gods in favor of the Christian faith imposed at
swordpoint by Norwegian King Olaf for political reasons
http://www.wvwnews.net/story.php?id=2888
---------------------------------------------------------------------------------
T Mark Hall
<jas...@snafu.de> writes
>On Wed, 9 Jan 2008 18:24:22 +0000, T Mark Hall
><tm...@gogod.demon.co.uk> wrote:
>>What I conclude is what I consider common sense. I said that the players
>>*should* be able to resume
>
>I do not consider it an aspect of common sense but of the nature of
>the game: It is a player's right to make strategic mistakes!
>
>From that I conclude more specifically
>a) It is a player's right to make strategic mistakes during
>board-plays.
>b) It is a player's right to make strategic mistakes by passing.
>
>So I think that the players should not be able to resume for the
>purpose of correcting strategic mistakes.
>
have quoted me, and your following remarks, looks as if I am saying the
opposite.
Juha Nieminen
> White passes, black passes. While counting, the players see that a white
> groups can be killed by adding one black stone. It can be saved by
> adding a white stone.
Where I play (using verbal japanese rules) some people use the
informal rule that if a teire passes unnoticed by both players and the
game ends, and then they notice it, the teire is filled like it was a
dame. In other words, the owner of the group loses one point, his
opponent gets nothing besides that (because he didn't notice it during
regular play).
Sure, there may be situations where this doesn't work very well (eg.
if one move is not enough to defend the group), but it has worked in
practice.