Talk:Rules of Go/Archive 1

Boundary Conditions

Does the boundary count as a liberty? If I have a piece on the boundary and it is surrounded by opposing pieces, is it always alive, or is it dead? 128.171.31.11 (talk) 12:46, 6 June 2008 (UTC)

The boundary does not count as a liberty. A stone at the edge which is surrounded on all other sides is captured. HermanHiddema (talk) 13:12, 6 June 2008 (UTC)

Miscellaneous Remarks

I delete some of my outdated remarks. Jasiek 2006-03-10

Sorry Robert - I'm going to have to treat you as a complete newbie on this. You should not delete talk page material like this, normally. The comments are not addressed just to you - if I had wanted to explain just to you what is happening here, I would use your user talk page (well, you aren't a signed-in user, but that is the correct way). So, the history on this page is to explain to everyone what has been happening here; otherwise tomorrow exactly the same discussion might occur. Charles Matthews 18:07, 22 Jun 2004 (UTC)

Robert - really:

(a) you are posting your opinions on go rules as factual, when they are just opinions;

(b) you are now posting your opinions on Wikipedia policy and etiquette as if they also mattered more.

Considering seven years of my reading your posts to rec.games.go, and I suppose vice versa, I am trying to explain things gently. But it seems to me you have to try to understand something about this large project, Wikipedia, to which 1000s of people contribute, before you begin to be helpful, rather than just making a mess which will need to be fixed up later.

Charles Matthews 18:34, 22 Jun 2004 (UTC)

The go rules page is in a process of being edited these days. I hope that afterwards the degree of facts is higher again.

As you seem to indicate, you have more experience about Wikipedia here. I was not aware of that. Since you say that talks are not deleted quickly, I may as well not delete them quickly. May I ask who deletes them or are they expected to grow larger and larger despite a 32K per page warning?

Old talk is generally "deleted" (though it remains in the history) or archived (moved to another page) on an ad-hoc basis when the page becomes too large... Evercat 19:09, 22 Jun 2004 (UTC)

OK, I was getting a bit depressed about stone counting. I think, clearly enough, we should segregate scoring/counting from the essential (outline) rules, saying that there is more than one recognised method of proceeding when the game is over, but for learning the game it doesn't matter so much. Charles Matthews 20:04, 22 Jun 2004 (UTC)

I think that's best - not to pretend to have a complete rule set, but links in particular to the ko rule section (so we don't have to distinguish superko from ko, which are equivalent in practice about 9999 times out of 10000, I suppose) and to something on scoring and counting (which ought to have diagrams). Charles Matthews 10:16, 23 Jun 2004 (UTC)

I think my solution would be to create SGFs and use the filters at http://gobase.org/software/sgf2misc/ to get images. I mean, this means in principle the same SGFs can be re-used to get different styles and standards of images. As I have not created any diagrams for Wikipedia, I'm not very confident in this area. Charles Matthews 12:25, 23 Jun 2004 (UTC)

So by counting the stones do you mean that you actually count them or is there any multiplying factors?--Grigoryev 17:51, 12 May 2005 (UTC)

Chain vs Group

This article now uses the word chain where it used to use group. The few Go books I've read used "group", and I think "group" is the more common word among Go players, but I'm not certain. Are there any objections to changing it back to group instead of chain? ~leif ☺ _HELO 11:39, 25 Oct 2004 (UTC)

Yes - it is completely confusing if you use group when you mean chain. After the introductory stage it doesn't matter so much - but initially one must be clear.

Charles Matthews 12:09, 25 Oct 2004 (UTC)

Ok, prior to just reading the Chain article at SL, I didn't realize group and chain meant different things. The definitions there are that chains are always connected, while "group" can sometimes refers to groups that are not strictly connected but just share a liberty. Perhaps this should be clarified in the article here? ~leif ☺ _HELO 12:32, 25 Oct 2004 (UTC)

One problem is that 'group' has no definite meaning, at a tactical level. Basically one assumes a group is well connected; but the usage of 'group' really is as a set of chains that cooperate well for a particular purpose. Charles Matthews 12:48, 25 Oct 2004 (UTC)

I find it interesting that on the go (board game) page and this article, it mentions the thinking times policy, both linking to the other . "See GO Rules" "See GO(boardgame)". THey both seem to have the same information on this particular point.

So what is a difference between one and another (chain/group)? Do you really have to have a "group"? In the game that I know you use grouping exclusively for the defensive purposes.--Grigoryev 17:51, 12 May 2005 (UTC)

The thing is that no one uses the term "chain" when discussing the game. So however ambiguous the word "group" is, that's what one would use to describe a strategic unit on the board. I am unable to imagine a situation in which one would confuse a "group" of connected stones with, say, a "group" that's under attack.Mimson 05:59, 15 January 2006 (UTC)

The only time you would talk about chains is when teaching a person or a computer to play go. When talking about strategy the group is what is relevant. But in teaching, the difference is important. If we use the word "group" to mean both the thing that is solidly connected (and is captured when surrounded), and also the thing that must contain two eyes, novices (who are the audience for this series of articles) may think you are saying that you must use a single solidly connected group, or chain, to surround two eyes. Whereas in fact most of the minimal groups use two chains; see diagram.

I freely admit that "chain" is not standard terminology, but then, there is no standard terminology for a set of strictly connected stones (probably because it isn't usually useful to talk about it). The AGA rules use "string" which is just as good as "chain", and so do Tromp-Taylor rules. New Zealand rules don't have a word for the concept at all. British rules say "solidly connected group" three times (i.e., there are seven white groups above), then they say just "group" to mean that same thing, then they say "These two separate spaces within the group are known as eyes", then they say "a group of stones which is unable to make two eyes, and is cut off and surrounded by live enemy groups, is called a dead group"; here they are back to using "group" in the usual strategic sense, in which there are four white groups above. This is exactly the kind of ambiguity that will mess with people learning the game. DanielCristofani 11:34, 15 January 2006 (UTC)

Can't we just eliminate the terminology and say "solidly connected groups"? I think it would be better than having new players going out and talking about "chains", which may cause unnecessary confusion. Mimson 05:17, 17 January 2006 (UTC)

On another thought, who came up with the terminology of chains? How about a citation, at least on the discussions page? Mimson 23:43, 3 February 2006 (UTC)

How remarkable -- more than 200 books have been published in English without the need for distinct terms for "group" vs. "chain." The Japanese make no distinction per SL. How is it then that we cannot seem to do without this confusion? Go is confusing enough for new players. And if there is a need for a second term, "chain" isn't it. A "chain" is "a series of objects connected one after the other" (Dictionary.com) how does that describe an eyeless clump of stones? "String" has the same problem. Face it - the word is "group." kibi (talk) 16:32, 25 August 2008 (UTC)

Most books can freely assume a basic understanding of the rules, and can use the informal "group" without confusion. Beginner introductions and official rules sets do not have that luxury. For example:

• Official AGA Rules: String
• Official EGF rules (Simplified Ing): Chain
• Official Japanese rules: Group
• Official Chinese rules: Unit

Any of these is a valid choice, group is the choice with the highest chance of confusion. HermanHiddema (talk) 19:23, 25 August 2008 (UTC)

Herman, are you disputing or agreeing? You seem to make the point that these terms are interchangeable; none of the books uses more than one for different concepts. Since "group" is the term one will encounter later on, it seems natural to use it from the start. "Chain" and "string" both imply a straight line, so I don't see how you can defend your unsupported assertion that they are less confusing than the term we all use.

kibi (talk) 17:20, 26 August 2008 (UTC)

My point above is that these terms are pretty much interchangeable, but that group is the only one that is ambiguous. I have no problem with changing the terminology to any of the options except group. I oppose the use of group in this context.

The usage in most books is meaningless, as those books assume a basic level of go understanding in their readers, an assumption we cannot make here at wikipedia. The material should be accessible to all, not just those that already know how to play go. The only books which are an exception are beginners books, where the reader is supposedly not familiar with go yet. I don't have a collection of beginners books at hand, so I do not know what terminology they use. The only English beginner book I own is Learn to Play Go, by Janice Kim, which deftly avoids using any term for this at all, not once does it use any of the words "group", "chain", "string", "unit" or any other term to indicate solidly connected stones. (The text of this book can be found here, see the section named capturing). HermanHiddema (talk) 20:01, 26 August 2008 (UTC)

Is this article necessary?

I hate to throw cold water on so much work, but speaking as an enthusiastic (if crummy) Go player and a longtime wikipedian, I would urge people to change this article radically.

We can't make it so comprhensive that it could be used by a newcomer to learn the game. For one thing, trying to explain eyes and liberties and all that will make it FAR too long to be read on most browsers. ("The Way of Go" by Ishi Press is about as simple an introduction as you can get, and it's many pages long.) For another, and on a more general level, Wikipedia is not, and shouldn't try to be, a set of rulebooks or instruction manuals.

The role of this article (if it should be an article at all, which I doubt) is to (a) give people a sense of how Go is played and how it differs from chess and other similar games - a feel for Go, not a lesson - and (b) to explain the history of how the rules have developed and a general feel of how they differ around the world.

At least, that's my opinion! - DavidWBrooks 20:14, 26 Feb 2005 (UTC)

Chess is massively irrelevant. 'Rules of Go' means two or three things: (a) instructions on how to play, (b) 'tournament rules', i.e. practical rules to deal with potential unclarities and potential disputes, (c) the arcana, namely close reading of rules sets, the slight regional differences and implications, and rational design for sets of rules. You say (a) matters most (I would agree really), and is impossible to cover sensibly here (not really true: life-and-death is derived from rules, not part of them. Under WP general policy this article needs to indicate something about all of these. Charles Matthews 06:03, 25 Apr 2005 (UTC)

Remember that this is not an article for beginners. The article (more specifically the section "Basic rules") should be updated to more clearly indicate this. alex (talk) 09:01, 2 January 2013 (UTC)

Representing positions

I mainly created this template to play games under Wikipedia sandbox/userspaces, but it can be used in articles out of convenience as well. The talk page there documents it. For smaller sizes than 20 pixel fonts, for now you have to use the no coordinate version. Oh anyone up for a possible game of go at Wikipedia:Sandbox/Chess? Could hold one there, or from a separate page. -- Natalinasmpf 03:33, 8 May 2005 (UTC)

I am in the process of documenting a manual for using the template. -- Natalinasmpf 01:26, 9 May 2005 (UTC)

Alternatives to play with the game

Back in high/school (in Ukraine), I used to play something of this game (we call it "the estate" or "the lands"). We used pens or pencils and regular "math" paper or construction design paper, the last one - almost never. One chose either a cross or a dot.

When a player would cut another one from the adjacent intersections, "the cutee" would connect his chosen objects around the number of the opponents objects that were cut, thus marking his territory putting the surrounded opponents objects out of play. This game doesn't have a concept of Ko. Wins one who has more "estate" or "land".

Another thing, I don't consider that the size of the board depends on a level of play. It's really not such a difficult game to understand. There's a saying "all genius is simple". The size of the board depends on time you would like to "kill" while playing the game.

--Grigoryev 17:18, 12 May 2005 (UTC)

Well it depends - a larger board allows you more freedom, but the enemy as well - hence you must plan ahead. With a smaller board it's harder to recover from a debacle or a mistake, but you have to plan ahead less. A larger board can be viewed as more advanced because both sides have to plan ahead more, in contrast to the smaller one. -- Natalinasmpf 06:11, 25 September 2005 (UTC)

Area vs. Territory scoring

The previous version of the Winner rule was completely wrong; it pretended that the winner would be the player with more empty intersections surrounded by only his stones. If you do not notice what is wrong with that: either the stones on the board are missing for Area Scoring or the opposing prisoners are missing for Territory Scoring. To end the competition about which scoring to describe on the rules page, the only promisingly lasting solution is to describe them all, what I have done now. -- Jasiek

I believe that the philosophy behind Japanese scoring can be seen if it is compared to a war. First, consider each stone to be a soldier or group of soldiers. If Area scoring is used, then when a player places a stone in a dead terriritory, they have wasted a life or lives of people on their side of the war. These incongruencies only matter at the end of the game when the average scoring between white and black is similar. Obviously if one person has a huge advantage, the scoring methods are not a matter of consequence. Though I understand that the chances that white and black have very similar scores is small, I still think it is an issue.

Theoretically, a person in the lead can potentially lose using Territory scoring if he constantly puts stones in the opponents territory that are well under the opponents' control. Obviously a better player that is in the lead would unlikely do this since they've gone so far into the game and are winning. But the problem is if both players are of equal skill. If this is the case, then every point is critical. If one player thinks that trying to capture a questionable area is risky, he would attempt it under chinese rules and wouldn't attempt it under japanese rules, unless he evaluated it as profitable. If a position is profitable, then of course a player would capitalize on it, but if it isn't or if it is really hard to determine, then the scoring method matters.

If you consider yourself the General or King of all the soldiers represented by your stones, then you do not want to sacrifice lives in a war that is already won or lost. Before the endgame, with either scoring method, you are penalized for placing stones in your own territory. This isn't done explicitly, but it can be shown by you "wasting" a turn, where you could have put your stone somewhere else to capture more area or territory. During the endgame, most fights have determined a winner except for a few leftover risky manuevers. Territory scoring would penalize you for making the mistake if the risk you took was not good, and Area scoring wouldn't. This doesn't make sense.

It is in this way that the Japanese scoring method is more elegant; not because it is the tradition or that there is a "beauty of omission." Lives matter, and that is why in the game of Go, we consider things Alive or Dead. It doesn't make sense that a ruler would risk any more lives of his people if the war is over. In conclusion, Area scoring allows for riskier play and Territory scoring allows for a person to balance offense and defense, while agreeing on an end of the game sooner too.70.111.251.203 14:59, 6 February 2006 (UTC)

Further analysis of Chinese vs. Japanese scoring has lead me to believe that both are roughly equivalent. Because of their equality, Chinese scoring is superior due to it's simplicity in counting compared to the Japanese territory. The only special case where Japanese scoring is particularly different is when one player is winning slightly and has the opportunity for a big risky move. Since close games would usually occur between two equally skilled players, they should both have the same chance of making mistakes and taking varying sized risks. 71.250.68.205 15:30, 14 February 2006 (UTC)

Basically in Chinese, you get to keep going without penalty, and in Japanese you try to end the game when it seems like it is done or you could risk losing points.

AGA rules

The AGA rules are basically the same as the Chinese rules. They have modified the Japanese territory scoring so that whenever a person passes they have to give one stone to the opponent as a prisoner and also that white has to be the last one to pass to end the game. This makes both area and territory scoring equivalent. Though it is not elegant, it cancels out the need for territory scoring and keeping track of prisoners at all for AGA rules. I think that this is a great simplification but it loses the value that Japanese scoring has instantiated. 128.6.175.17 13:41, 15 February 2006 (UTC)

Komi on 9x9 and other boards

I can't find what are proper Komi values and handicape for other boards. Do they have same Komi/handicap as 19x19?

I don't know, either. But, on the Kisedo Go Server, the "default" number of handicap stones on a small board is less than those on the larger boards.

The komi is not well defined for smaller boards. - alex (talk) 08:40, 2 January 2013 (UTC)

Tibetan rules

These have been pushed onto the page, but should they really be on Go Variants? -Zinc Belief

Star point positions on smaller boards - wrong?

In the section Optional Rules -> Compensation -> Handicap there is a discussion of star points, or hoshi, which states:

Smaller boards such as the 13x13 and 9x9 also have star points. The 13x13 has 9 at the 3-3 points, 3-6 points, and the center.

It seems to me that if the 13x13 board has 9 hoshi then the right position for the side ones must be 3-7, not 3-6, as 7 is the centre line. A quick net search suggests to me that the 13x13 board commonly has only 5 hoshi so the 3-6 points should be taken out of the list, not altered. But being new to Wikipedia AND not a Go player I'm not game to change this; could someone familiar with the game check that the description is right? -Medb 17:52, 10 October 2006 (UTC)

Ing Rules

It should be Ing Rules, not ING, since they are named after a person, Ing Chang-Ki.

The article states that Ing rules use a “Super Ko” rule. However, Ing tried to create a ko rule that would avoid the need for players to keep track of whole-board positions. The result was a ko rule that classified a ko as either “disturbing” or “fighting,” and also created the term “hot stone.”

The Ing rules ( http://www.cs.cmu.edu/~wjh/go/rules/SST.html ) seem to have problems. The definitions/terminology section contains arguments in favor of his rules; the definitions/terminology section uses terms that aren't in the definitions/terminology section; some of the terms are defined in other parts of the rules; some of the definitions simply refer to a diagram which gives an example of the term being defined.

I don't think there is a clear definition of “disturbing ko” or “fighting ko.” If, in fact, there is no clear definition, the result may be, in effect, that Ing rules use the “super ko” rule.

In order for Ing counting boxes to work, Ing stones were of uniform thickness. 206.53.197.12 16:24, 15 October 2006 (UTC)

The Ing Rules were named after Chang-Ki Ing, and are not an abbreviation for Intelligent Go. I started making the change from ING to Ing. SlowJog (talk) 02:22, 10 May 2008 (UTC)

The section on Ing rules stated that this rule set used "super ko." I edited it to match what was stated in the section on ko. SlowJog (talk) 17:41, 10 May 2008 (UTC)

Counting up points

Sorry but I'm a complete newbie when it comes to GO and I have a question, when your counting up points at the end of a game how do you count them?

I added to the Chinese method of scoring to describe the actual method used at the Chinese Go club in Monterey Park, California in the 1990's. It's fun to make the rectangles and count by tens, especially when you win. Larry R. Holmgren 15:50, 10 February 2007 (UTC)

Excellent article

I think this is objectively one of the best articles, even better than India, and other articles; it may not be as important, but is very well written just like Chess.

Differences section

I really like this article except for the very last section which talks about the differences bettween diffrent rule sets. I would remove it beacuse it contains repetition of information already in the article and what appears to be orgional research. First it says the two rule set are almost the the same. Besides being somewhat obvious, this is stated peviouslly in the scoring section. If it is retained it should be in the introduction to the section not at the end. The second part of the section appears to me to be conjecture about how the diffrent rulesets alter play style, and that needless extending the game is dishonerable. I think this should be verified or removed. Lotu 20:35, 5 November 2007 (UTC)

Equivalence of scoring methods

The article says:

"If the game ends with both players having passed the same number of times, then the score will be identical no matter which method is used."

But isn't this technically incorrect?

Consider a game where neither player passes and in the end they take turns in filling the dame points as usual. Under Japanese scoring the dame points don't increase the score of either player but under Chinese scoring all stones, including the ones placed at dame points, increase the player's score. Now, if there's an odd number of dame points, one of the players get one point more, which could theoretically be the difference between victory and defeat.

Or have I misunderstood something? Ossi (talk) 22:45, 25 November 2007 (UTC)

No, it's correct as I understand it: there is a difference in that case. White must end in even games for the number of stones to be equal. Furthermore, the handicap stones (if any) make another difference. But the main difference is probably with sekis, since the territory is nil in that case, but the number of stones may be widely different. Filling the dame points simplifies the count, that's all. Michelet-密是力-Me laisser un message 07:09, 26 November 2007 (UTC)

Ancient Chinese rules?

I have seen a few mentions of the ancient Chinese rules of Weiqi or Yi which are distinct from the modern Chinese rules. But I cannot find an English text for them. Apparently there was a two point tax on each living group. This suggests to me that originally one's score was the number of stones of one's color which were on the board at the end of the game. This would require you to capture all enemy stones within your territory and then fill it up with your own stones except for two one-point eyes to keep it alive. Does anyone know any more about the ancient rules? JRSpriggs (talk) 09:57, 29 February 2008 (UTC)

As far as I know, ancient rules were territory scoring like the Japanese rules, and area scoring wasn't invented in China until the 15th century AD. At its introduction, there seems to have been a period where the "group tax" applied, but this was later dropped. HermanHiddema (talk) 11:16, 29 February 2008 (UTC)

According to Ikeda, "It would have been natural and easy to start from the simple rules of placing stones alternately on the board and removing stones that became surrounded, and arrive at the conclusion that the player who could place the most stones on the board should be the winner." [1] and "A player's score is the number of that player's played stones." [2] (his Area Rules I). JRSpriggs (talk) 08:11, 11 March 2008 (UTC)

It may seem natural, but the evidence is against it. Ancient rules were territory rules, area rules were not used until Ming dynasty times (15th century AD). Stone scoring (with group tax) was only in use a short time at the beginning of that period. HermanHiddema (talk) 11:29, 11 March 2008 (UTC)

Dishonorable???

The subsection Rules of Go#Differences says "It's because of the design of the Chinese rules that players using this rule set will be more able to take greater risks than compared to playing under Japanese rules; this may draw games out a little longer and at certain times seem dishonorable." (emphasis added). I cannot see how playing by the Chinese rules could encourage dishonorable behavior. Can anyone explain this? Or is this an improper POV statement? JRSpriggs (talk) 11:02, 16 March 2008 (UTC)

Basically, that whole section is crap, I'm deleting it. HermanHiddema (talk) 15:38, 16 March 2008 (UTC)

Suicide

Currently most major rule sets forbid playing such that a play results in that player's own stones being removed from the board,
[...]
in certain circumstances a suicidal move may threaten the opponent's eye shape, yielding a ko threat.

Emphasis mine. Shouldn't that be a snapback threat? -- Jokes Free4Me (talk) 19:27, 21 April 2008 (UTC)

On second thought, shouldn't the threat be left general, with no mention of ko, snapback, atari, or other particular kinds of threats? -- Jokes Free4Me (talk) 19:29, 21 April 2008 (UTC)

All rule sets forbid suicide of a single stone; rule sets that allow suicide permit it for more than one stone. I forgot where I read that, so I added the "fact" tag when I edited the article. (Suicide of a single stone would be the same as a pass, anyway.)

As far as I can tell, there is no reason for a player to suicide his/her stones except as a ko-threat. SlowJog (talk) 02:45, 10 May 2008 (UTC)

Suicide can sometimes grant life, kill or make seki, see:

• http://senseis.xmp.net/?Suicide for an example of making seki with suicide
• http://senseis.xmp.net/?SuicidalTendencies for an example of a suicide life & death situation

HermanHiddema (talk) 09:35, 10 May 2008 (UTC)

Aga Rules

Is it me or is there an exception to territory and area count-scoring giving an equal result.--ZincBelief (talk) 15:08, 10 July 2008 (UTC)

pwaldron: I believe exceptions can occur in board positions where there is a seki involving an odd number of neutral points. — Preceding unsigned comment added by 132.246.27.86 (talk) 17:26, 27 July 2011 (UTC)

Is a stone connected to itself?

This is not an extremely serious question, since the rules can accommodate either interpretation. I have a preference for saying a stone is connected to itself. I think connectedness should be an equivalence relation. Two editors have expressed objections, or at least seem to dislike the idea. Here are the alternatives:

Definition 1. In a given position, two stones of the same color (resp. two empty intersections) are said to be connected if it is possible to pass from one to the other by a succession of stones of that color (resp. empty intersections), each one other than the first being adjacent to the one before it. Any stone (resp. empty intersection) is understood to be connected to itself.

Definition 2. In a given position, a liberty of a stone is an empty intersection adjacent to at least one stone to which that stone is connected.

Definition 3. In the final position, an empty intersection is said to be a point of a player's territory if it is not connected to any empty intersection adjacent to a stone of his opponent's color.

Definition 1'. In a given position, two stones of the same color (resp. two empty intersections) are said to be connected if it is possible to pass from one to the other by a succession of stones of that color (resp. empty intersections), each one other than the first being adjacent to the one before it.

Definition 2'. In a given position, a liberty of a stone is an empty intersection adjacent to that stone, or adjacent to any stone to which that stone is connected.

Definition 3'. In the final position, an empty intersection is said to be a point of a player's territory if neither that empty intersection nor any empty intersection connected to it is adjacent to a stone of his opponent's color.

I actually think a stone is connected to itself under Definition 1', since you can take the succession consisting of just that stone. Of course, in everyday life, when you see "two stones," you usually assume "two distinct stones" is meant.

HermanHiddema proposes a change to Definitions 1', 2', 3'. Since this would necessitate changes in a number of places (including commentary on the rules), I am temporarily reverting the change, but I am certainly amenable to the amendment if after discussion it appears most editors are favorable to it. 136.152.224.6 (talk) 21:28, 30 August 2008 (UTC)

Although the concept of stones being connected to themselves is mathematically elegant, I fear it will confuse too many readers, and I think the usage can be avoided. I've proposed a change which would probably work. An alternative is to define a new term like chain or string for a set of connected stones (including a single stone), and define that all members of such a chain share their liberties and that a liberty is an empty point adjacent to any member. This is an approach which many rule sets follow. HermanHiddema (talk) 13:13, 1 September 2008 (UTC)

At least six rule sets exist, some of which have been developed over a period of centuries. None of them seems to require stating that a stone be "connected to itself." It is sufficient to say that stones are connected to the liberties they or connected stones share. This concept looks like "original research" to me, and does not belong here. Herman is right. The rules are confusing enough. BTW what does the abbreviation "resp." mean? kibi (talk) 13:27, 2 September 2008 (UTC)

"Chain" or "string" don't work either. They imply stones lined up in a row, but in fact the term must also include any clump, or "group", that is connected by the lines of the board. kibi (talk) 13:27, 2 September 2008 (UTC)

Chain or string not working is your point of view, with which I disagree (my POV). Both these terms, as well as "unit" and "group", have been used in actual rule sets, and usage would thus be verifiable and hence acceptable for wikipedia. HermanHiddema (talk) 20:53, 2 September 2008 (UTC)

Resp. means "respectively" and is a standard abbreviation. I have explained the notion of "chain", but in the commentary on the rules rather than in the rules themselves.

As for the question about original research, I don't think this would be original research, since the definition of "connected" is given only to allow a convenient statement of these rules. No claim is made about the way the word connected is used elsewhere. On the other hand, if this contradicts the way the word is used elsewhere, then that is a practical problem and a legitimate objection.

It does appear that the Simplified Ing Rules view a stone as perhaps not being connected to itself, since they give this definition:

• A region consists of an intersection and any intersections connected to it.

(Their definition of connected doesn't make it clear, however.)

In any case, I'm prepared to concede this. I simply didn't realize this would be confusing. I had previously assumed that a stone was always connected to itself, but I realize now that this contradicts the way some people (at least) think about it. 67.150.254.54 (talk) 22:52, 3 September 2008 (UTC) (same contributor as 136...)

Sidebar on "resp:" The usage here is incorrect. According to Wiktionary, "respectively" means "In a relative manner; often used in lists, to clearly indicate that two (or more) separate items or lists correspond to each other, in terms of order." For instance one might say that "Go players use two sets of stones, black and white respectively, to play." But the whole phrase about "vacant intersections" seems unclear and unnecessary, the sentence works better without it. kibi (talk) 13:44, 4 September 2008 (UTC)

Obviously, the ideal thing would be to find a good reference for how resp. should be used, but here is an example:

• Let I (resp. S) be the initial population of the inner city (resp. the suburban area). [3]

It avoids the need to repeat an entire sentence whose structure would be parallel, except for one or two words. That is exactly what is needed here because connected empty intersections are defined in a way analogous to connected stones. 67.150.245.8 (talk) 23:48, 4 September 2008 (UTC)

OK 67.160.245.8, here are references for how resp. should be used, none of which supports your usage: Wiktionary

Dictionary.com

Merriam-Webster

Encarta

Cambridge Dictionary

American heritage Dictionary

Please do not put it back unless you have more to back you up than an example of misuse on another web page. kibi (talk) 15:14, 12 September 2008 (UTC)

Kibi, this use of resp. falls under the "in the order given" definition in the Merriam-Webster. You may disagree with this use of the abbreviation resp., but you have yet to suggest an alternative. What you have written in the article leaves out the definition of connected empty intersections altogether, which is far worse than the supposedly incorrect use of resp. So if you understand what is meant (as exemplified by the citation you have called incorrect), then please find a way of saying it that agrees with your idea of linguistic correctness. Otherwise, the lesser of the

two evils here is leaving in the version with resp. 67.150.255.98 (talk) 00:29, 16 September 2008 (UTC)

Sorry, but the [phrase "two distinct stones of the same color (resp. two distinct empty intersections)" does not list more than one item in a given order. Your usage is incorrect and as far as I can tell, unnecessary. The statement "two distinct stones of the same color are said to be connected if it is possible to pass from one to the other by a succession of stones of that color, each one other than the first being adjacent to the one before it" is fully self-explanatory. Or, to be simpler, "Two stones of the same color are connected if there is an unbroken line of stones of that color from one to the other." If you're unhappy with this, there are at least six widely validated rule sets out there; if you can't find something there that says what you're trying to say, isn't that kind of weird?

Frankly, the whole emphasis of your statement seems confusing to beginners, since the point is not that any two stones are connected but that they are all connected. I don't think we want novices trying to figure out if this stone here is connected to that stone over there; just that it has a connection to a living group. Seems like a silly thing to have a revert war about, but if you insist. . . . To others, reading this thread, anyone else have a thought? kibi (talk) 15:09, 17 September 2008 (UTC)

I do not think removing this usage of "resp." is the answer. The rules as stated on the page require the concept of emtpy intersections being connected. This information is now gone from the article. If you have a problem with grammar, try to fix it while keeping the intended meaning of the text. Just deleting stuff and thereby removing information from the article is not helpful. HermanHiddema (talk) 08:33, 18 September 2008 (UTC)

Kibi, the usage is not incorrect. It is encountered widely in mathematical texts, at least. Here is an example from the web of an explanation of how this is used:

• A (contrived) example would be: "The square of 2 (resp. 3) is 4 (resp. 9)." That could also be stated: "The squares of 2 and 3 are 4 and 9 respectively." [4]

If you feel strongly about it, we can repeat the entire sentence with "empty intersections" in lieu of "stones of the same color," as Herman has done. This is not my preferred option as it adds to length, and deemphasizes to some degree the analogy between connected stones and connected empty intersections. However, since I'd rather avoid an argument about English usage, I'm quite satisfied to leave in Herman's version.

In your proposed version, I don't think the meaning of the words "unbroken" and "line" is clear. The rule sets you are referring to sometimes use language that would be difficult to interpret if it were not for the fact that a lot of people already know what the rules are trying to say.

As for the issue of whether we should speak of two or several connected stones, what you're implicitly talking about is the notion of equivalence class. There might be some roundabout ways of describing equivalence classes without first defining the corresponding equivalence relation, which relates only two stones. One would be to give this definition:

• The chain of a given stone is the smallest set of stones of that stone's color which: 1) contains that stone; and 2) contains any stone of the required color adjacent to a stone in the set.

You can try proposing a better definition of a chain and then we can discuss its merits and weaknesses. 67.150.253.118 (talk) 19:38, 18 September 2008 (UTC)

Thanks for your thoughtful response, 67.150.253.118. (May I call you "67" for short?) Now I understand better what's happening. I'm not a mathematician so I was not aware of this special usage of "resp." I think Wikipedia is intended for the general public, so this usage will be confusing to many others non-math people as well. Still don't understand why the sentence doesn't work by merely removing the "resp." references, but it seems like I'm alone on that . . . kibi (talk) 13:36, 25 September 2008 (UTC)

Yes, of course, please call me 67. All my friends do. I had always assumed that "resp." was widely used this way in non-mathematical contexts as well. Perhaps I was wrong. As for why it couldn't just be removed, look at the old and new versions. They mean the same.

• In a given position, two distinct stones of the same color (resp. two distinct empty intersections) are said to be connected if it is possible to pass from one to the other by a succession of stones of that color (resp. empty intersections), each one other than the first being adjacent to the one before it.
• In a given position, two distinct stones of the same color are said to be connected if it is possible to pass from one to the other by a succession of stones of that color, each one other than the first being adjacent to the one before it. Similarly, empty intersections are connected if it is possible to pass from one to the other by a succession of adjacent empty intersections.

Without the part about connected empty intersections, the following definition is no longer meaningful, because nobody has said what the word "connected" means in it:

• In the final position, an empty intersection is said to belong to a player's territory if neither it nor any empty intersection connected to it is adjacent to a stone of his opponent's color. 67.150.255.7 (talk) 04:25, 26 September 2008 (UTC)

Just a quick note: I am neither supporting nor opposing the usage of "resp." Since I am not a native English speaker, I have no idea whether it is correct or not. My position was merely that the answer to grammar mistakes is not to remove information from the article, but to fix the mistakes by rewording the text while keeping its intended meaning. HermanHiddema (talk) 08:50, 26 September 2008 (UTC)

Split proposal

In the last few days, I have expanded the article a good deal, in large part by adding examples and diagrams. Most of that is directed at people looking for a basic idea of what go is. The page may be too big now, so I suggest splitting it into two articles, one for the basic rules and one focused on differences between rule sets. I can't think of good names, but two possibilities that come to mind are Basic rules of go/Rules of go and Rules of go/Variation in the rules of go.

The first article would contain:

• The statement of the basic rules.
• The explanation of the basic rules. (This part is probably over half the length of the current article.)
• Something along the lines of the current section "Comparative features of the basic rules" but with links to the appropriate sections of the second article.

The second article would contain:

• A link to the first article in the opening paragraph.
• (Possibly) a brief statement of the basic rules.
• Complete information about differences in rule sets.

I would like input on this, including suggestions for the names of the articles if the split seems like a good idea. 136.152.224.59 (talk) 08:21, 2 September 2008 (UTC)

Excellent idea! I think that Rules of Go is the most appropriate name for the first article, explaining the basics. Perhaps the article dealing with the differences between various rule sets could be called Go rule sets? That would allow for easy and nice links to sections like Go rule sets#Chinese, Go rule sets#AGA, etc. HermanHiddema (talk) 20:49, 2 September 2008 (UTC)

Actually, I imagine that the bulk of the second article will be made up of headings like "komi", "handicap stones", "suicide", "prevention of repetition", "territory scoring", "end-of-game rules/confirmation phase", etc. There's no point reexplaining situational superko repeatedly under New Zealand, AGA, etc. (Of course, the individual rule sets should also have their own sections.) This is why I thought "Variation in the rules of go" might be a more appropriate title than "Go rule sets".

Instead of linking to specific rule sets from the basic article, just link to the appropriate "issue" in the second article. As an exception, top priority problems with the basic rules, like the fact that playing the game out is a bad idea under Japanese rules, should be mentioned explicitly in the basic article.67.150.245.243 (talk) 01:42, 4 September 2008 (UTC) (same person as 136...)

I agree strongly to this suggestion. As it is now Rules of Go is very intimidating to new players. Since this page is quite likely the first encounter for a beginner to the rules of go it is highly important that we make it clear and concise. I suggest we move the current page to Rules of Go details and restart Rules of Go from a clean sheet. Betelgeuse11 (talk) 08:11, 19 October 2008 (UTC)

(For the benefit of Betelgeuse, I'd like to repeat that I am the same editor as 136.) Actually, my proposal would be to have the current sections "Basic rules" and "Explanation of the basic rules" in the first article, and the last few sections of the current article in a second article devoted only to those topics on which rule sets disagree. 67.150.246.211 (talk) 05:35, 27 October 2008 (UTC)

Self-capture

This whole concept, presented as "Step 3" of the turn, is out of whack. There is no "Step 3." If suicide is permitted, the player committing suicide does not remove his/her own dead stones; they become the opponent's prisoners in the opponent's next turn. I started to clean this up, then decided to post this and see if anyone supports keeping this unnecessary, misleading and erroneous section. kibi (talk) 16:49, 4 September 2008 (UTC)

I'm sorry, but those are rather strong statements. Do you have anything to back them up? Because AFAIK all the common rule sets that allow suicide specify it like that, eg:

• New Zealand rules: A play consists of placing a stone (of that player's own colour) on an unoccupied intersection, then removing any of the opponent's stones that then have no liberties (if any), and then removing any of that player's own stones that then have no liberties (if any).
• Official Ing: Removal: Breathless stones are taken off the board by the player who eliminated their last breath, whether the stones belong to that player or his opponent. This is called removal. When the stones of both sides become breathless simultaneously, the player removes his opponent's stones.
• EGF Simplified Ing: A play removes stones of every region without adjacent empty intersection. Removing opposing regions takes precedence over removing own regions.
• Tromp-Taylor: A move consists of coloring an empty point one's own color; then clearing the opponent color, and then clearing one's own color.

I am not aware of any rule set that does removal of suicide stones on the opponent's next turn. The page as it was was correct. I also feel it is rather insulting to the author, who has done a lot of hard, good work on this article, to baselessly call this "wacky", "misleading" and "erroneous". HermanHiddema (talk) 18:30, 4 September 2008 (UTC)

Let me address some of the changes by Kibi. I think some of them should not be kept.

• The rules are essentially the "elementary rules" given in the reference. These allow suicide.
• On the mechanism of suicide:

• As Kibiusa has revised the rules, they no longer make sense if Optional Rule 7A is removed, since the "suicidal" stones stay on the board after a move.
• If the stones are removed on the opponent's next turn, that may screw up the non-repetition rule.
• Since area scoring is used, it is not necessary to view the stones as being "captured". What to do with captured stones is properly addressed in the section devoted to "territory scoring", where it can be explained that all stones of one color that have been removed in the course of the game become prisoners of the other player. It does not matter who physically removed them from the board.
• In any event, I agree with Herman that this is the scheme adopted by major rule sets.

• The elementary rules of James Davies do not have a komi. Including one defeats the purpose of keeping the rules simple, and relegating discussion of variants to another section. Virtually all sets of rules taught to beginners ignore the issue of komi. Even the "Simplified Ing Rules" treat komi as a matter for tournament directors, and not for the rules themselves, although this is perhaps going a bit far. Saying in the rules that "five to seven" stones are given is a bit too specific, and in any case makes little sense since scoring does not involve prisoners.
• It is necessary to define "connected empty intersections" in order for the definition of "territory" to make sense. The Simplified Ing Rules do it this way: Likewise, empty intersections are connected if they are adjacent or if there is a chain of adjacent empty intersections between them.
• I point out the fact that adjacent intersections must be distinct, because if it is not pointed out, one might say, "Well, a point is certainly connected to itself by a line, and there is no other intersection between that point and itself". This is done for clarity. Kibi, you were arguing before that stones should not be connected to themselves. Now, you are arguing that there should not even be clarification that an intersection is not adjacent to itself. I'm afraid this is not clear from the definition that has been left. (You say that "two intersections are of necessity distinct." When you say "the two numbers √9 and 3 are equal", are you saying that they're distinct? Or are you saying that I should have said "the number √9 and 3 is equal"?)
• I'll add that suicide of a single stone (if it weren't prohibited by the superko rule), would have the same effect as a pass, except with regard to the end-of-game rule. What Kibi seems to be missing is that there are no prisoners in the basic rules. (This is just as in the Chinese rules, and, importantly for us, as in Davies' rules.) This is not a "wacky piece of original research". 67.150.255.245 (talk) 02:45, 5 September 2008 (UTC)

I accept that my attempt to address the self-capture issue has left some discrepancies in this article that must be addressed. In addition, in view of the reported research I withdraw my statement that there is "no step 3." Thank you 67.150.255.245 for bringing those facts into the discussion. I still insist however that "step 3" is a needless complication that does not appear in any major rule set. When is the last time that you played under New Zealand, Ing or Tromp-Taylor rules? In the Ing rules the player removes all captured black and white stones at the same time. There is no separate step -- that is an original idea. What happens to those stones? One infers that they go back in the bowl, which is fine since there are no prisoners in Ing rules (or Chinese rules). In the absence of further objection, I will do my best to revise the article so that it's all consistent without this needlessly complicating step. The question remains, what is the simplest way to present the rules? The current description of "basic rules" here makes it seem way more complicated than it needs to be. We need to boil it all down to the essential elements. I own a go shirt that tells you everything you need to play in about 100 words and 2 diagrams. Five-year-old children can play this game, but they could never begin to grasp all the twists and turns of our "basic rules." Explaining why a stone is "connected to itself" or an intersection is "adjacent" to itself does nothing to advance anyone's understanding. Let's aim for the normal person who can think clearly, not try to clarify every silly thought that a person could have. kibi (talk) 14:38, 5 September 2008 (UTC)

Leaving aside your sarcasm, what I understand from what you've written is that you believe the rules in the statement are too precise, which adds to their length.

You say "if there are no further objections". I do object to removing that step. And I would say that what Herman wrote above counts as an objection. Removing it will in any case do relatively little to reduce the length of the rules. Rules like "Go is played by two players" and "Go is played on a board" were here before I touched the article, and I didn't see any need to remove them.

There are three of us who have edited the article recently. What I propose is that we discuss your objections one by one and decide collectively what to do. 67.150.245.122 (talk) 21:21, 5 September 2008 (UTC)

I've reverted most of your changes to the last version without inconsistencies in the rules, until agreement can be reached. We should address issues one at a time.

On the issue of when the last time I played by New Zealand rules was, I can only say that I do not live in New Zealand, but that that is not an argument against them in an international encyclopedia. Their rules are mentioned on virtually every page listing major rule sets. See for example the British Go Association: [5]. The New Zealand rules are often cited by authors as a model. See for example [6], written by the author of a book on go for beginners.

I think that our job is not to present a dumbed down version of complicated rule sets, but to present a simple one that makes sense on its own, and leave complications for other sections of the article, or even another article. The "Elementary Rules" of James Davies are good for this, but leave some things undefined or only vaguely defined, like "surrounded". That is why I think it is desirable to borrow some ideas from other rule sets, like the New Zealand and Simplified Ing Rules, that were intentionally designed to be simple and rigorous. 67.150.252.182 (talk) 02:28, 6 September 2008 (UTC)

I am sorry for the slow reply, but I currently have extremely little time to spend on wikipedia. I will try to quickly give my opinion on some issues. First of, the basic rule should be: This is wikipedia, which means you do not invent new terms or methods, you make sure everything you write can be backed up with quotations, etc. So:

• Try to avoid the use of "stones being connected to themselves" and "points being adjacent to themselves", as far as I am aware there are no rule sets that describe this, so wikipedia is not the place to introduce such a concept. As I suggested before, the use of "string" or "chain" is an alternative that is in use in existing rule sets.
• The last time I played with Ing rules was this summer, when over 700 people played with them at the European Go Congress, the largest tournament outside Asia. They are the official rules for several EGF organized tournaments.
• I would prefer to include suicide as part of a three step process, this is how it is described in existing rule sets. Ing rules do not actually "remove all captured black and white stones at the same time", because it is impossible for a move to both capture enemy stones and be suicide, if a move captures stones it automatically creates at least one liberty for itself. The Ing rules shortcut the three-step process into a sort of two step process, where the second step contains a conditional. This is a valid alternative to the three step rule, and I have no objection to using it instead, though I prefer the three step process for clarity.
• Kibi removed the the mention of black and white having "unlimited stones", but as far as I am aware, there is no rule set in use that limits the number of stones that may be played (otherwise, we would see things like: In the 5th game of the Honinbo final, white ran out of stones and lost). Keep in mind that unlimited is not the same as infinite, the players are given an adequate number of stones (usually about 180 each), but if they do run out, they are provided with extra stones. The phrasing "adequate" might be an alternative here, it is also used in the official AGA rules.

HermanHiddema (talk) 10:14, 10 September 2008 (UTC)

I never thought an intersection should be adjacent to itself, and I gave up on stones being connected to themselves. Now the disagreement seems to center on whether it should be pointed out that these concepts apply only to distinct intersections/stones. On "unlimited" vs. "adequate", either is fine for me, but I like them better than Kibi's suggestion "ample". Finally, I agree with Herman that a three-step process is clearest. 67.150.245.44 (talk) 17:45, 11 September 2008 (UTC)

"Adequate" seems fine.

I'm glad we seem to have settled the question of whether a stone is connected to itself, adjacent to itself, or otherwise in some weird parallel universe. Many other things to discuss however. One of the great beauties of go is its elegant simplicity, which is completely lost in the current article, which amounts to an inventory of every possible misunderstanding a reader could possibly reach. "Precise" is good but there is such a thing as too much information, cf. Blink by Malcolm Gladwell. Perhaps we should give up the fantasy that go can be learned by reading the rules. It really requires actually interacting with other players. kibi (talk) 14:59, 12 September 2008 (UTC)

"Territory" section

This section as originally written contains several important errors. A is not a neutral point; it is Black territory under "Area" rules. Even more clearly, c can end up as nothing but White territory. Herman reverted these changes on the basis that they refer to "the current situation rather than a hypothetical end situation." However, there is nothing "hypothetical" about it unless Black fails to see that g is atari. On the same note, Black could miss the atari at e, allowing White to start a ko with i; so h is not really 100% Black territory yet either. By the way, the footnoted document does not contain this example diagram, so I have removed that incorect footnote as well. —Preceding unsigned comment added by Kibiusa (talk • contribs) 16:36, 29 September 2008 (UTC)

Kibi, a beginner needs to know what would happen if both players passed, and the game ended that way. Otherwise, he doesn't know why the players need to make the moves you're talking about. 128.32.238.145 (talk) 22:48, 29 September 2008 (UTC)

In any case, the reader is warned that good players wouldn't pass in this position, by the following paragraph:

• In order to understand the definition of territory, it is instructive to apply it first to a position of a kind that might arise before the end of a game. Let us assume that a game has ended in the position below (even though it would not normally occur as a final position between skilled players).

As for the footnote, the example game is exactly the one in the rules. Obviously, those rules are written with the participants in mind, who are all dan-level, so it doesn't bother explaining what would happen if a game ended in the middle that way. I chose to use a point before the end of the game to illustrate what would happen. Since only the game, and not the diagram, comes from the source, perhaps I should have written "example of a game" instead of "example". However, the real final position is exactly as in the WMSG rules.

I'm reverting your changes for the reasons I've explained. 128.32.238.145 (talk) 23:17, 29 September 2008 (UTC)

Just a note to Kibi to acknowledge that you were right about the game not appearing in the rules. They appear in an appendix to the rules. I've corrected the link now. Thanks for pointing this out. 128.32.238.145 (talk) 23:30, 29 September 2008 (UTC)

Also, this section is only defining territory, so there is no need to speak of territory rules vs. area rules in this section. Though most rules with area scoring systems use the word territory, I chose to adopt the distinction between territory and area that appears in the AGA rules. The reason is that in the non-basic sections, this article needs to discuss both concepts, so different words are needed for clarity.

Territory scoring can be mentioned in the section "Score". However, only the existence of something called "territory scoring" should be mentioned there, because in order to understand it fully, a whole array of topics would have to be explained. These should be done in the section "Scoring systems", which is not part of the explanation of the basic rules. My opinion is that differences between the basic rules and other rule sets should only be mentioned in passing in the section "Explanation of the basic rules", and developed fully in other sections of the article. (I think the prohibition of suicide should be an exception, though.)

The topics for territory scoring include prisoners, and, most difficult of all, a confirmation phase. A simple rule like "the game ends when both players have passed consecutively" simply doesn't work any more, because that would force Black, for example, to play in his own territory to capture a dead white stone, and that would cost him points. So life and death then have to be defined somehow, so that you can come back to the position before the game was played out. That is a complicated task, and even the "advanced" parts of this article are very incomplete on the topic at the moment. Obviously, this is overload for a simple set of rules. That's also why the basic rules adopt area scoring. 128.32.238.145 (talk) 00:16, 30 September 2008 (UTC)

Thanks for correcting the footnote. Now I see the problem -- you did not quote the source by reproducing the original diagram. Instead, you have omitted the last several moves, then tried to imagine how beginners might count the game if they think it is over. You're trying to teach beginners how to count a game that isn't finished, how does that make sense?

If the goal is to define territory, then it is perfectly appropriate, in fact necessary, to describe the two basic concepts of what territory is. "Area" and "territory" cannot be omitted from the discussion.

Your analysis is also simply incorrect. Ifa is not black territory because it is connected to b, the same can be said for all of the points marked X, which is of course ridiculous. And under no circumstances can c and d be considered anything but white territory. Sorry, but your whole approach here violates Wikipedia's "original research" policy. If you cannot accept a proper description of the state of affairs in this example, you probably need to redo from scratch with a record of a game that is actually finished. I'm undoing your reversion and in the absence of better support, will continue to do so. I revised the discussion of a and g slightly, since that's not what actually happens in the completed game record; less confusing for someone who actually checks the original source.

kibi (talk) 18:02, 1 October 2008 (UTC)

This is an explanation of the rules of the game for people who haven't seen them before. If we don't explain what would happen at an intermediate stage of the game, then there is no way to understand why the moves you're talking about are the right ones.

What you're calling "ridiculous" is exactly what would happen if two unskilled players passed at that stage. Why would skilled players not pass in that position? In large part, precisely to prevent all that territory from remaining neutral. So if nobody tells a beginner what happens in a case that wouldn't really happen, they don't know why they need to make the moves that are made.

Also, your discussion of strategy is probably excessive at this point, when people are just trying to find out what territory is, not how to get it.

About territory and area, this section only explains territory, not area. The next explains area. The section "Score" says that a player's score is the number of points in his area, but that there are other scoring systems, for which the reader can refer to "Scoring systems". I wrote at length above to explain why it is not feasible to have a detailed discussion of territory scoring in the section "Explanation of the basic rules".

As you have left the article at the moment, it is not even coherent with itself. Even if it were, I don't think the direction you're going is the right one. At, the moment, of three active editors, you're also the only one who thinks that what was there before was incorrect or misleading. I'm going to revert and ask you to gain consensus for the type of changes you want to make. 67.150.253.52 (talk) 22:17, 1 October 2008 (UTC)

I give up. Assuming that there are actually two of you, not just one person logging in from two different IP addresses -- since neither of you has an editing account here -- I accept that I have been outvoted by two people, so I guess the page will just have to confuse people. Fortunately there are other resources for them. But neither of you will ever be able to show that c and d are anything but White territory, as it stands, under any rule set.

kibi (talk) 14:54, 2 October 2008 (UTC)

Kibi, I didn't mean to imply that 67 and 128 are different people. I am the same person, and I thought you'd guessed. (For example, as 128 I apologized to you for having gotten the link wrong, something I'd done as 67.) The third person I'm referring to is HermanHiddema, who reverted your territory changes here: [7]. 67.150.255.236 (talk) 21:29, 2 October 2008 (UTC)

And I'd just like to remind you that the article says this to tell readers what they're seeing:

• Let us assume that a game has ended in the position below (even though it would not normally occur as a final position between skilled players). 67.150.253.87 (talk) 22:14, 2 October 2008 (UTC)

"Original research" template

Kibiusa considers the section Rules of Go#Territory, most of which I contributed (variously as 136, 67 and 128), to contain original research.

Kibiusa has previously removed the disputed content, but was reverted by a third editor, HermanHiddema, here.

The section contains an example in which the players pass earlier than good players would, with the result that huge swaths of neutral territory are left on the board in the final position. Based on Kibiusa's previous comments on the talk page, I think this is what Kibiusa's objection is to.

The diagram explains why a and similar points would be neutral territory in this situation. The statement that if the players passed in this position, all of that territory would be neutral, is not original research; it is a straightforward application of the rules, which are referenced. (Please see WP:NOTOR#Obvious deductions.)

Kibiusa's request for a citation for the fact that a is neutral territory is like asking for a citation from a published source establishing that 4,687 × 5,672 = 26,584,664. In one case we make a straightforward deduction from a previously published algorithm for multiplication; in the other, we make an obvious deduction from previously published rules of go.

Though the charge of original research here is absurd, it might be argued by Kibiusa that it's simply a bad idea to talk about things that would never really happen. Here is a hypothetical situation that would never occur in a game between good players:

Before		Black plays		After removal

Though this position would never really occur, it is useful in illustrating the operation of the capture rule. In the same way, the hypothetical premature game end is used to illustrate the definition of territory.

Kibiusa cannot make the argument that we are misleading readers either, because they are warned with this sentence that good players would not play this way:

• Let us assume that a game has ended in the position below (even though it would not normally occur as a final position between skilled players).

I'd appreciate comment from other parties so we can remove the absurd "original research" template. I don't think I can do it myself without establishing consensus. 128.32.238.145 (talk) 04:31, 29 October 2008 (UTC)

I agree that this in no way constitutes original research. The added citations are fine, and I see no reason to keep the OR banner there. HermanHiddema (talk) 10:18, 29 October 2008 (UTC)

Agree, this looks fine to me.--ZincBelief (talk) 10:54, 29 October 2008 (UTC)

I agree as well, that territory is still neutral by derivation from go rules. However, I agree that multiple ISPs are confusing to follow through one discussion; consider using a account if you can. /Ninly (talk) 14:32, 29 October 2008 (UTC)

Kibiusa has removed the banner. Thanks to all of you for your input. 128.32.238.145 (talk) 02:04, 31 October 2008 (UTC)

About article quality

From the editing and the discussion so far, I get the feeling that Kibi feels that the current article is overly technical, will not help a reader unfamiliar with the rules in learning those rules, and may in fact put them off. That may be the case (I am not passing judgment here, just considering the option), but the right course of action in that case is to discuss that on the talk page, listing you concerns and how you suggest to address them. Accusations of "Original Research" are not the way to go in such a case.

It is perfectly possible to write an extremely complicated and technical article on go rules without any OR in it, because most formal rules text are very technical, and sometimes extremely complicated. From a Wikipedia point of view, such an article would be fine. As long as everything is neutral and backed up by citations, Wikipedia has no problem with such content. As go players, we may consider the additional goal of attracting the attention of potential new players. That goal is completely irrelevant by Wikipedia standards, but as long as the content of the text is correct by Wikipedia standards (WP:NPOV, WP:NOR, WP:VER), there is no objection to such an additional goal. In fact, Wikipedia encourages well written text (Refreshing Brilliant Prose, as it was once called).

So please, lets work together on how we can improve this article, ok? HermanHiddema (talk) 10:18, 29 October 2008 (UTC)

Obviously, there is a trade-off between precision and simplicity in some cases. The appropriate balance to strike between these two, or ways to achieve both at once, should be discussed without resorting to the sort of hyperbolic language that has been used here.

When disagreement has arisen, it has also often felt to me as if Kibiusa simply ignored my arguments. I don't know what it means when you write several times that you are describing a situation in which both players have passed and the game has ended, and that fact is never even acknowledged as a serious argument by the other editor in discussion. You end up going round in circles. 128.32.238.145 (talk) 02:50, 31 October 2008 (UTC)

Hi 128, sorry if I haven't seemed responsive to your arguments, so let me clarify that. I believe the best way to teach is to show correct situations and explain them, not to show incorrect situations and try to make sense of what is not sensible. If the position you offer appeared in a real game between real beginners, according to AGA rules they would play it out and decide among themselves who won. There can be no "correct" or "precise" explanation of an incorrect situation, like the one you offer.

And if it's a choice between "precision" and "simplicity" with beginners, to me the choice is obvious . . .

BTW others have asked that you creat a login account as an editor so people can tell who you are when you log in from different URLs, I also request that you do so, perhaps you could explain why you haven't?

kibi (talk) 13:13, 31 October 2008 (UTC)

According to the AGA rules, if the players both passed, then couldn't agree which stones to remove, and then passed twice again, then the game would end exactly as in the paragraph. So it would take four passes. I agree that the players would be playing badly if they did that, but that is what would happen under the rules.

What matters, though, are Davies' and Tromp-Taylor's rules, because they are the basis of the rule set we are explaining. It's clear that if players pass just twice in a row under those rules, the game ends and is scored using that position.

So if what you mean by "incorrect" is that the rules are being applied incorrectly, that is wrong.

If what you mean by "incorrect" is that it's a bad idea to pass in that situation, that's true. But it's impossible to teach a game by showing only good moves - you have to tell people what the consequences would be, according to the rules, for bad ones, like passing here.

For example, the fact that, before the white stone is played at the top in the second diagram, all the points marked X would not yet be White territory if the game ended there, gives a good idea of why White should play where he does in that second diagram. Besides explaining the definition of territory, that's what I think is good about showing what would happen if the players ended the game too soon.

As for precision vs. simplicity, I think the article Go (game) already gives a summary of the rules, so the main point of these rules is the in-depth explanation, not the brief statement. I think there's room here for the rules to be entirely accurate (without leaving terms like "connected" and "surrounded" undefined or only vaguely defined), while remaining reasonably accessible. Their bare statement may be a difficult read, but my hope is that the detailed sections with examples won't leave things unclear. Ucbear (talk) 01:41, 1 November 2008 (UTC)

Thanks for creating an identity, ucbear. BTW we are actually using AGA rules, not Tromp-Taylor, at least those are the rules you cite to justify your assessment of the position. Going with that, you are right, the position given is valid under AGA rules. Perhaps a better word than "incorrect" would be "incomplete", since any experienced player can see that the position is unfinished. I strongly disagree with your premise that one must explain bad moves (except a game in which bad moves were actually played). If the position isn't finished, and you want to xplain something, explain how it should be finished!

I would agree that there is room for at-length explication of the rules, for experienced players who are interested in the differences, fine points, etc. But a detailed analysis of an incomplete position is not going to teach any beginner anything useful. kibi (talk) 17:12, 1 November 2008 (UTC)

The AGA rules are cited for the definition of "neutral points". Your problem isn't so much with the definition of neutral territory as the meaning of the "end of the game". Here, the Tromp-Taylor rules and Davies' (the ones followed in the article) are simplest: the players pass twice and the game ends.

About bad moves, actually, game commentaries for amateurs continually look at variants that would never be played by professionals, if only to explain why not to play that way. You can't explain what a game is by showing only good moves. The question "Why shouldn't I do this?" is an important one.

The thing that a person should take away from the example is that, for instance, if White neglects to play a stone for the rest of the game, then the territory in the top right is not his. That explains why he plays the stone where he does in the next diagram.Ucbear (talk) 04:33, 2 November 2008 (UTC)

Connected stones

Regarding this revision, I would like to obtain consensus on which edit is better for the article.

The first of the two edits, reversion to "...and that is not connected to any other stones", is a source of confusion for me. I am a novice Go player. Rather than explaining why I am confused (it's hard for some who doesn't understand something to explain why he doesn't understand it!), I want to focus everyone's attention on this phrase and double-check that it is as clearly written as possible?

For the second edit, Ucbear's observation that "'same color' is superfluous" is correct but in my opinion not sufficient reason to re-delete the phrase. My argument is that the intention and function of the phrase "... is the same as saying that" is exactly to be superfluous. In addition, the "same color" phrase reminds the reader that one should not ever refer to two stones of different colors as connected with each other. Bsoo (talk) 04:32, 22 April 2009 (UTC)

Regarding the first argument, I think the point is to show that in this example:

That B1 and B2 do not form a separate chain. So B1-B2-B3 is a chain because it is not connected to any other stones, but B1-B2 is no a chain because it is connected to other stones (and the same goes for B2-B3). I do think the wording can be improved here though, the current wording is somewhat formal, mathematical, one might say, and may indeed lead to confusion. HermanHiddema (talk) 11:03, 22 April 2009 (UTC)

On the first point, does either of you have a suggestion for how to define "chain"?

About the second edit, Bsoo, I've made a change which I hope you will find to be an improvement. The statement as it was originally worded was of the form "A is equivalent to B", where A is "The stones are connected," and B is "The stones belong to the same chain." I believe what bothered you was that the only case truly of interest is when C, "The stones are of the same color," is true, since it is well known to the reader that otherwise, both A and B are false.

I don't think that "A is equivalent to (C and B)," which is what you had, really clarifies things for the reader, although it is technically true. It does, as you say, have the advantage of reminding the reader that C is a prerequisite for A. However, C is also a prerequisite for B, and writing "C and B" instead of just "B" could mislead the reader into questioning whether that's the case.

Instead, I've changed the passage to read "If C, then A is equivalent to B," since we don't really care about cases where C is false. I hope that's okay. Ucbear (talk) 07:34, 30 April 2009 (UTC)

Thanks, Ucbear. I think that does the job!

Revisiting the first point, I have a suggestion now. First, to quote the sentence in question again: "A chain is a set of one or more stones of the same color that are all connected to each other and that are not connected to any other stones."

Suggestion: "When a set of stones are connected to each other, they are collectively referred to as a chain. A chain has the following properties:
- A single stone is itself a chain when it is not connected to any other stones.
- The term "chain" always refers to all members of the chain collectively, and never a subset. Therefore a chain cannot be said to be connected to any other stone (or chain)."

First of all, I think we can remove references to colors since it is already stated in the definition of connected. I think referring to colors adds confusion, as if we were focusing on a same-color "case" of connected stones. Secondly, I think this rewording requires less mental effort to comprehend but still paints a complete picture - I would think this is how most people actually understands this definition in their head - it is certainly how I understand it now. --Bsoo (talk) 14:42, 19 July 2009 (UTC)

Concerning your point about colors, I don't have strong feelings about that. I will make a change and you can make a change yourself if you think it can be further improved.

About your suggested definition, I don't find it easy to understand for two reasons: a) you first give one definition of "chain," but then modify it, by restricting it further; b) the definition of how it is restricted further makes reference to itself, since you say that the word "chain" is used only to refer to all members of a "chain." The last part does say that no proper subset of a chain should be a chain, but it takes a good deal of reasoning to piece together what a chain is from this statement of what a chain may not be.

The way I think of a chain intuitively is this: it's a set of stones that are all connected to each other, and that is not contained in any larger set with the same property. I think this is what your formulation is probably getting at. (For people familiar with math, this can be formulated concisely by saying that a chain is a maximal element of the set of connected sets of stones, partially ordered by the relation of inclusion.)

However, I think that even though "not contained in any larger set with the same property" (i.e., consisting of connected stones) is probably what we think of visually, it is not particularly easy to understand when written down. In my opinion, "not connected to any other stones" is better for this purpose, because it requires consideration only of other stones, and not of bigger sets of stones, which are a more complicated idea. I think any definition close to the one you are proposing would require us to think about sets of stones other than the one we are claiming to be a chain. Ucbear (talk) 01:56, 3 August 2009 (UTC)

The oldest method???

" The oldest counting method is as follows: At the end of the game, all white stones are removed from the board, and the players use black stones to fill the entirety of the black territory. Score is determined by counting the black stones. Since the board contains 361 intersections, black must have 181 or more stones to win. This method is still widely used in mainland of China. " - Where does this come from? This is not the oldest known method. The oldest known method is something else entirely, and there has been speculation that the oldest method was to simply count the number of stones on the board (thus resembling area scoring with group tax), but this is by no means known. So where does this statement come from? I don't see any references. What's being described in the quote is a method for using area scoring, written quite hastily (it's not very thourough, and it's also not necessary to remove white stones). Area scoring - as described above - popped up in the 20th century when the chinese term of "land" (= "place where you can put a living stone") was replaced with the japanese term "territory" (= "an empty point surrounded by your own stones"). This is wrong so it should either be sourced or removed. 83.142.0.60 (talk) 19:53, 28 November 2009 (UTC)

Tromp-taylor rules?????

Who gives a crap about the tromp-taylor rules? The rules that should be described in the lead and the "explanation of basic rules" should be either japanese rules or chinese rules (or even better: Both). Here they are: Chinese rules in english: http://www.cs.cmu.edu/~wjh/go/rules/Chinese.html ; More recent chinese rules in chinese: http://go.yenching.edu.hk/chirule.htm ; and japanese rules in english: http://www.cs.cmu.edu/~wjh/go/rules/Japanese.html . Here the tromp-taylor rules are presented first and then the article goes on to mention japanese, chinese, AGA, Ing and New Zealand rules as a curiosity - it should be the other way around. In fact, I don't see why the tromp-taylor rules need to mentioned at all - then we might aswell start to describe every little rule set along with toroidal boards ("daoqi") and so on; it's hardly notable enough in comparison to all the other major rule sets. The same goes for "simplified ing rules" and "logical rules". 83.142.0.60 (talk) 20:06, 28 November 2009 (UTC)

grrrrrrrrrr —Preceding unsigned comment added by 76.236.79.23 (talk) 02:28, 30 May 2010 (UTC)

stone scoring

Stone scoring shouldn't be identified with area scoring. Please get that right. No matter if SC ever was used or how short. After all it's the most logical of all and should be at the top. RP —Preceding unsigned comment added by 86.111.232.133 (talk) 11:11, 10 August 2010 (UTC)

Done, though I've put it it at the bottom rather than the top. Although I agree it's the most logical, and it was likely the original method, but it seems to me to be too obscure. If you'd like to move it up, you should probably add more (referenced) detail on the history and theory.Bennetto (talk) 15:47, 11 January 2012 (UTC)

Fit for Wikipedia?

This article feels morel like a how to than an encyclopedia article. You wouldn't include a game guide in an article for Super Mario Bros., so why include it in this.

The wording is even outside of the voice of an encyclopedia. for example:

"For simplicity, we will illustrate the rules mostly using 5 × 5 boards"

I have never seen "we" used on wikipedia. This article needs serious overhall TylerRDavis (talk) —Preceding undated comment added 05:05, 7 September 2011 (UTC).

• I agree with this. This areticle reads more like an expositional essay than an encyclopedi article. Reyk _YO! 07:49, 4 November 2012 (UTC)

Mistake in reconciling scoring system

In the section "Attempts at reconciling the scoring systems", if Black passes last, area counting will yield 1 more point for Black than territory counting because White did not have the opportunity to give a passing stone to balance the score. AGA rules forces White to pass last for this very reason. This should be made clear in the article. 199.180.97.5 (talk) 07:42, 7 November 2012 (UTC)