go defined

[phillip1882]

so i'm trying to think up a rule set that's clear, concise, and handles most situations if not all, with ease.
first let's define Go itself.
Go is a 2 player board game, played on a 19 by 19 board.
players play on the intersections of the squares rather than inside.
played with white and black circular pieces, called stones, with black playing first.
each side places 1 stone at a time.
the objective of the game is to surround empty intersections with your stones.
one you play a stone on the board, it cannot be moved, but it can be captured and removed.
a stone is considered connected to an ally stone if it placed either horizontally or vertically connected, called a group. a group of stones (or a singular stone) may be captured if it is surrounded on all 4 sides.
if a group of stones surrounds 1-6 intersections, we call this an eye. a group with only 1 eye can be captured, if the opponent surrounds both the outside and the inside.
if a group surrounds two separate intersections, we consider it unconditionally alive, and cannot be captured.
suicide is illegal, that is, you cannot play in such a way that your move leads to self capture. so you can only completely fill inside a group surrounding empty intersections if you have first surrounded outside.
repeating a previously played board state is illegal. when such a possibility to do so arises, it is called ko, usually by capturing back and forth.
when such a situation arises, you must play elsewhere, before recapturing, or fill the empty capture intersection.
you may also pass you turn at any point in the game.
when both players pass consecutively, the game is considered over.
any group of stones not forming two eyes, or connected to stones that do so, or cannot easily do so, are considered dead, and removed as captures.
players get 1 point for each empty intersection they surround, and 1 point for every capture. white also gets 6 1/2 free points. the player with the most points wins.


how am i doing so far?

[alejo]

Fine, but you forgot about seki... Though I can't find an easy way to explain it

[moyoaji]

I once attempted to make a concise set of rules. It was actually the reason I joined this forum - to share it and get comments on it. Here's the original thread: http://www.lifein19x19.com/forum/viewtopic.php?f=10&t=8520

The problem I found is that, if you make the rules concise enough that you don't scare off beginners, more experienced players have a lot of complaints.

Really, go has quite basic rules if you want to be concise. Here is the review section of that rulesheet:

1. Go is played with black and white stones on a grid. Stones are played on the points
(intersections). Points connected by line segments are adjacent.
2. The objective of Go is to control more of the board than your opponent.
3. A stone or group must have at least one liberty (adjacent empty point) or it is captured.
4. Rule of Ko: The board position cannot repeat.
5. The game ends when both players pass. Your score is your territory minus captured stones.

That is the essence of go, but if you define go like that then this raises a lot of questions. You can look at the old thread if you want to see the kind of discussions and questions that came from my attempt to make a concise set of rules with examples. I also discussed many of the resources I found and why I didn't use them. They all had flaws that led me to create my own rulesheet based on the rules from Sensei's Library.

I will be teaching at JAFAX 19 this year and hope to update the sheet for that. I intend to add a "life and death" explanation page because that was the most common question I got when teaching there. If you can create something better than what I have I'd love to use it.

[msgreg]

See also the AGA website's Handouts section, which obviously focuses on the AGA ruleset.

There are a variety of rules handouts including on a business card, on a half-page, and on a full-page.

[RobertJasiek]

how am i doing so far?

100 kyu.

a rule set that's clear, concise, and handles most situations if not all, with ease.

Most is not good enough. All is good enough.

Clear, concise, with ease:

http://home.snafu.de/jasiek/simple.html
http://homepages.cwi.nl/~tromp/go.html

if a group of stones surrounds 1-6 intersections, we call this an eye.

This is one of your first great mistake. Hint 1: 1-6 intersections can also be more than 1 eye. Hint 2:
http://senseis.xmp.net/?FormalDefinitionsOfEye#toc1
http://home.snafu.de/jasiek/wagcmod.html

a group with only 1 eye can be captured

This is your second great mistake: you overlook a group with exactly 1 eye that cannot be removed. (Seki!)

not forming two eyes, or connected to stones that do so, or cannot easily do so

This is your third great mistake: you fail to define "cannot" and "easily". Hint: You need to refer to hypothetical-sequences and hypothetical-strategies as in:

http://home.snafu.de/jasiek/j2003.html

It is your exercise to find all you minor mistakes.

[Jingliu]

The essence of Go rules is to have a simple and clear criterion to judge the belonging status of intersections. Inspired by traditional Chinese rule, there is a neat answer to this fundamental question: "A intersection's final belonging is decided by who has the ability to occupy it with a stone that his opponent isn't able to remove through actual play.". This is an once for all solution even no need to mention what is "eye", what is "live", what is "seki" or other lengthy definitions. All disputes can be solved through actual play without special clauses.

Note: There are some intersections both players are not able to occupy(e.g. the last 2 eyes of a group, the mutual liberty in a seki, or other delicate situation), this kind of intersections should not be counted in score, otherwise it will contradicts the aforementioned mantra. This may be a bit weird but it's indispensable to keep the rule logically self-consistent and as concise as possible.

here is my version inspired by traditional Chinese rule:

1. Go is a game played with black and white stones competing for controlling more intersections on a Go board.
2. Stones are played alternatively, black first and then white. Once played, stones cannot be moved unless out of liberty(see below). Players have the right to pass if necessary.
3. A stone or a group of stones must have at least one unoccupied adjacent intersection(calls "liberty") to stay on board , if not, they must be removed from board immediately.
4. A player cannot play a stone that will cause his own stone(s) out of liberty unless it can also cause some opponent's stone(s) out of liberty in the same time, in which case this play is valid and the opponent's stone(s) should be removed from board instead.
5. A player cannot play a stone that will exactly reproduce a former situation.
6. A intersection's final belonging is decided by who has the ability to occupy it with a stone that his opponent isn't able to remove through actual play.
7. When both players agree, game ends, the player controls more intersections wins.

[Mike Novack]

Seki is not a rule but a state taht is a consequence of the rules.

Nor is seki usually hard defined*. What we often mean when we say that a position is seki is that for either player to continue would mean death for his or her group and life for the opponent's. But the rules don't prohibit doing that (and it might be done as a ko threat when the life of the seki group is worth less than the ko).

* For example, thirteen stones arranged as a line of five, below that three stones arranged 1,3,5, and below that again a line of five is absolutely alive and cannot be played against even as a ko threat.

[Uberdude]

Very Simple Go Rules:

1. Lee Sedol wins.

[Bill Spight]

so i'm trying to think up a rule set that's clear, concise, and handles most situations if not all, with ease.
first let's define Go itself.

Go is defined by its rules.

Go is a 2 player board game, played on a 19 by 19 board.

Most people would think of 19 squares by 19 squares, instead of 19 lines by 19 lines.

players play on the intersections of the squares rather than inside.

You mean intersections of the lines. And you have gotten ahead of yourself.

played with white and black circular pieces, called stones, with black playing first.
each side places 1 stone at a time.

Not quite enough to define a play.

the objective of the game is to surround empty intersections with your stones.

No, it's not.

a stone is considered connected to an ally stone if it placed either horizontally or vertically connected, called a group. a group of stones (or a singular stone) may be captured if it is surrounded on all 4 sides.

First, using group this way can confuse people later on. Better to say string, chain, or unit.

Second, it is inexact. Better: A group of stones of one color that are connected horizontally or vertically is called a string. A string can also be a single stone. An empty point horizontally or vertically adjacent to a string is called a dame. If the opponent plays a stone on the last dame of a string, the string is captured and removed from the board (except as provided by the ko rule).

There are other formulations, of course.

if a group of stones surrounds 1-6 intersections, we call this an eye. a group with only 1 eye can be captured, if the opponent surrounds both the outside and the inside.

Better not talk about eyes in the rules.

when both players pass consecutively, the game is considered over.

Since you are talking about territory scoring, this provision leads to problems. Every practical set of rules allows play to resume after two passes.

[Bantari]

you fail to define "cannot" and "easily"

Hmm... I understand the need to define "easily", but I thought "cannot" is pretty self explanatory. Its the opposite of "can". As in "I can do this" vs "I cannot do this." Or is this nitpicking?

[Mike Novack]

a stone is considered connected to an ally stone if it placed either horizontally or vertically connected, called a group. a group of stones (or a singular stone) may be captured if it is surrounded on all 4 sides.

First, using group this way can confuse people later on. Better to say string, chain, or unit.

It is much worse than that. A group does not necessarily consist of connected stones! For example, two subgroups (that are each a connected string) that between them have two separate empty spaces, constitute a live group, even though the subgroups are not, and cannot be connected.

[DrStraw]

you fail to define "cannot" and "easily"

Hmm... I understand the need to define "easily", but I thought "cannot" is pretty self explanatory. Its the opposite of "can". As in "I can do this" vs "I cannot do this." Or is this nitpicking?

"Can" is ambiguous in the vernacular. It means both "am able" and "am allowed". Very different.

[Bantari]

you fail to define "cannot" and "easily"

Hmm... I understand the need to define "easily", but I thought "cannot" is pretty self explanatory. Its the opposite of "can". As in "I can do this" vs "I cannot do this." Or is this nitpicking?

"Can" is ambiguous in the vernacular. It means both "am able" and "am allowed". Very different.

Linguistically, you are certainly right.
With respect to actions you take or not - the result is the same: You ain't gonna do it.

[RobertJasiek]

I thought "cannot" is pretty self explanatory. Its the opposite of "can". As in "I can do this" vs "I cannot do this." Or is this nitpicking?

"Can" and "cannot" are abused as informal abbreviations for "can force" and "cannot force" in, or implied by, the following extract of

http://home.snafu.de/jasiek/j2003.html

"For a given player, a _hypothetical-sequence_ is an imagined sequence of hypothetical-moves that

- starts from the final-position with the given player,
- lets the players alternate hypothetical-moves,
- either has a finite number of hypothetical-moves and ends with the pass succeeding a pass or has an infinite number of hypothetical-moves and does not have a pass succeeding a pass, [...]

A _left-part_ of a hypothetical-sequence is either the whole hypothetical-sequence or a part that consists of one or more than one successive hypothetical-moves of it and starts with its first hypothetical-move.

A player's _hypothetical-strategy_ is a set of one or more than one left-parts of hypothetical-sequences so that

- each left-part starts with a hypothetical-move of his,
- each left-part ends with a hypothetical-move of his,
- there are not two left-parts so that they without their last hypothetical-move are equal, and
- the above is not true for the set together with any left-part not in the set.

A hypothetical-sequence is _compatible_ with a hypothetical-strategy of a player if each left-part that is of the hypothetical-sequence and ends with a hypothetical-move of the player is in the hypothetical-strategy.

A player _can force_ something if there is at least one hypothetical-strategy of his so that each compatible hypothetical-sequence fulfils that something."

[Bantari]

I thought "cannot" is pretty self explanatory. Its the opposite of "can". As in "I can do this" vs "I cannot do this." Or is this nitpicking?

"Can" and "cannot" are abused as informal abbreviations for "can force" and "cannot force" in

Hmm... not being a rule theorist, I have to admit most of that goes above my head. But from what I understand, what you talking about, the different between "can force" and "cannot force" is more of the strategic issue. Where rules are concerned, I think more in terms of "can do" and "cannot do", as in "i can play here" (i.e. the rules allow it) or "i cannot play here" (the rules do not allow it.)

In either case, the words "can" and "cannot" are pretty much obvious in meaning, I think.
Just like "do allow" and "do not allow". I think anybody who reads it will pretty much understand what it means.