Life In 19x19

beginsA291 wrote:distinction of legally alive but with dead shape.

Hi beginsA291,

Again, to be clear: in Go, the terms 'alive' and 'dead' have very specific meanings.
In the case above, you meant to say "legally sitting on the board (with at least 1 liberty)", but dead -- it is not alive.

skydyr wrote:experience is probably the best teacher

You ask a very good question.
And as you can see from the discussions so far,
it is not easy to answer. As skydyr suggests, you need to play many more games to gain the experience necessary to understand some of these nuances.

EdLee wrote:

beginsA291 wrote:distinction of legally alive but with dead shape.

Hi beginsA291,

Again, to be clear: in Go, the terms 'alive' and 'dead' have very specific meanings.
In the case above, you meant to say "legally sitting on the board (with at least 1 liberty)", but dead -- it is not alive.

skydyr wrote:experience is probably the best teacher

You ask a very good question.
And as you can see from the discussions so far,
it is not easy to answer. As skydyr suggests, you need to play many more games to gain the experience necessary to understand some of these nuances.

I think that beginsA291 is in a situation where he cannot play a game because he cannot tell which stones are dead or not. In that case, switching to AGA rules would allow him to play and get experience. (As would playing the Capture Game. )

It sounds like, with beginsA291 frustration with the forum, that he may not be back.
It may also be possible that with his poor introduction to the rules and confusion with the complicated explanations, that he may not come back to the game either.

Both would be very sad.


I like to explain territory-based rules and scoring as merely an agreed upon shortcut to its "cousin", area-based rules/scoring. While its not technically an accurate description, I think it serves the purpose quite nicely for beginners. Because effectively, the result of the game should not change, regardless of which ruleset you use (also, not technically accurate, but close enough for beginners)

Bill is right. You should switch to using area-base rules until you gain more experience. I think it'll make things easier to understand. And then you don't have to worry about changing the score with more or less prisoners (because prisoners don't count), or filling in your own territory, because you can continue playing, removing all the "dead" stones without changing the final score/result.

Its really a great game. Don't give up on it (or us), yet.

edit: we all actually love answering/discussing these types of beginner questions.

edit: we all actually love answering/discussing these types of beginner questions.

Except that most players forget that in the beginners forum, you have to explain in a beginner-frindly way. Was it nessesary to mention at least 5 different rulesets (most of which are theoretical and have never been used for actual play)? Plus, I assume that the position was supposed to be just one dead stone somewhere in the opponent's territory, not asking about this specific position leading to a seki with the possibility of a ko fight.

beginsA291, please don't give up. People who have been playing for a long time tend to forget how confusing every game (including go) is untill you get the hang of it.

If I am allowed to slightly modify the position to simplify:

Click Here To Show Diagram Code

[go]$$Bm11
$$ ----------
$$ | . . O X . . |
$$ | . . O X O . |
$$ | . . O X . . |
$$ | . . O X . . |
$$ | . . O X . . |
$$ | . . O X . . |
$$ | . . O X . . |
$$ -----------[/go]

The rule is that "stones that cannot avoid being captured" are removed after two consecutive passes and become prisoners (just as if they were captured during the game). If white claims that his/her lonely stone can avoid capture, and black claims it cannot, the situation is played out on a another board. While there are many possible sequences for white to try, the result will always be the same: black will capture all white stones in his territory.

Click Here To Show Diagram Code

[go]$$Wm1
$$ ----------
$$ | . . O X 6 . |
$$ | . . O X O 8 |
$$ | . . O X 1 3 |
$$ | . . O X 2 4 |
$$ | . . O X 5 . |
$$ | . . O X 7 . |
$$ | . . O X 9 . |
$$ -----------[/go]

Click Here To Show Diagram Code

[go]$$Bm10
$$ ----------
$$ | . . O X X . |
$$ | . . O X . X |
$$ | . . O X . . |
$$ | . . O X X X |
$$ | . . O X O 1 |
$$ | . . O X O 2 |
$$ | . . O X O 3 |
$$ -----------[/go]

One of many possible sequences, they all lead to white being captured.

So now it is clear white's lonely stone cannot avoid capture. Players return to the original board and remove the stone (having proven it cannot avoid capture), and add it to any other prisoners captured during the game. Now the prisoners and territory of both players are counted to find out the result of the game.

beginsA291 wrote:Argghhhh. I have lost my carefuly thought out post because the forum timed out and logged me out with no warning before I'd finished. Time and brain power lost! Very annoyed at the forum at the moment!

In thanks to the two players who replied to my post I will write it again, but I've gone off this forum now.

I feel your pain. I recommend Lazarus: Form Recovery, a browser add-on, and it will never happen again.

tiger314 wrote:

edit: we all actually love answering/discussing these types of beginner questions.

Except that most players forget that in the beginners forum, you have to explain in a beginner-friendly way.

Good point On the other hand, the surprising variations that Kirby and others showed in which we see that the white stone may not be as dead as it looks, offer a wonderful and important lesson: When discussing go, the specifics of a position always matter. Naturally, a beginner may not have suspected these intricacies, and you were right to offer a position that better served to clarify the question at hand, but I don't think that the answers previously given were out of line. On the contrary, they demonstrate that go players are rightfully cautious about making general pronouncements.

“he may not be back” & similar … I'm back.

I admit to frustration with the game over this issue though. This otherwise excellent introduction http://www.playgo.to/iwtg/en/count.html where I got my example from, says “Thus black will not make such moves.” But there is no “thus” because they do the opposite of what you would think without the knowledge of the ending procedure of testing life and death. Many introductions are similar.

It was meant to be a simple toy example that put the issue I had into a nutshell. But it turned out a 5x2 area is enough to require analysis from you experts!

Actually looking again I see that wikipedia Go and Rules_of_go do address this, but only after I know what to look for. Territory counting does seem to lead to headaches (the rules not covering every situation, tribunals, different treatment of ko). Ambiguity and the need for special cases in the rules (like topazg' bent four in the corner and “the awkward bit for which I have no answer”) is off-putting, makes further traps for beginners and generally detracts from the elegance of the game.

There does seem to be folk knowledge and gentlemens' agreements involved that is not explicitly included in the rules. Or does that go too far?

- Question: Wikipedia says under territory scoring “If the players reach an incorrect conclusion, then they both lose”. But how do we know the conclusion is incorrect? Who says? Also does it mean the disputed territory is not counted to either player, or that neither wins the game? (The latter seems harsh!)

- Question: Does area counting avoid all these difficulties? Why isn't this method (or other rule choice mentioned) adopted by all if so?

Also : “you have to explain in a beginner-friendly way”
For me, the problem was the absence of important information (about territory scoring), everything said here is understandable and interesting. Thank you all.

beginsA291 wrote: - Question: Does area counting avoid all these difficulties? Why isn't this method (or other rule choice mentioned) adopted by all if so?

Area scoring is used, but once you get used to territory scoring, it is quite fast. It just takes a bit of time to get used to and preferably have someone show you.

The odd positions like bent 4 show up once in a while, but they're not hard to take into account.

beginsA291 wrote:- Question: Wikipedia says under territory scoring “If the players reach an incorrect conclusion, then they both lose”.

As a general statement for all those rulesets having territory scoring, this is wrong. A both lose rule occurs for the Japanese 1989 Rules, but the actual rule says something else. Therefore, if your citation is accurate (I have not checked it), this information on Wikipedia is misleading or possible wrong. As a beginner, you do not get any advantage by studying the 1989 Rules, as you can notice by having (not more than) a quick glance at

http://home.snafu.de/jasiek/j1989c.html

But how do we know the conclusion is incorrect? Who says?

In theory, the ultimate source is the complete knowledge of all legal sequences and strategic decisions; this concept is sometimes called "the omniscient player". In practice and for practically occurring positions to be judged by arbitration, the strongest (available) player (should) have the knowledge and skill to make the same judgement that the omniscient player would make. Alternatively, it sometimes is possible to make mathematical proofs about the existence of sequences and related decision-making, although such does not appear in real world territory scoring rulesets.

Also does it mean the disputed territory is not counted to either player, or that neither wins the game?

Forget about these "both lose" ideas until you want to become a rules expert.

- Question: Does area counting avoid all these difficulties?

Yes.

Why isn't this method (or other rule choice mentioned) adopted by all

Because in some countries, clubs or tournaments, the existing tradition of using territory scoring prevails over simplicity.

beginsA291 wrote:“he may not be back” & similar … I'm back.

Glad you are still with us.

Territory counting does seem to lead to headaches

I learned the Japanese rules that came with the go set I bought in Jackson, MS, many years ago, which were probably as crummy as the rules that came with your set, but I never had any trouble. I know people who did have headaches, though.

(the rules not covering every situation, tribunals, different treatment of ko).

The Ing rules may or may not cover ever situation, but they are area rules. The Japanese rules (territory) have a certain ambiguity in the definition of life and death. Different rules treat kos and other repetitive situations differently, but that is a modern development. I don't know about tribunals. {shrug}

Ambiguity and the need for special cases in the rules (like topazg' bent four in the corner and “the awkward bit for which I have no answer”) is off-putting, makes further traps for beginners and generally detracts from the elegance of the game.

The ambiguity of the 1989 Japanese rules concerning life and death has never been a problem in practice. But it's only been 26 years. Maybe something will come up. The 1949 Japanese rules had special cases, but that is a non-issue now. (BTW, I was shodan before I knew what Bent Four in the Corner was. )

As for the "awkward bit for which I have no answer", all rules have problems with such positions. The basic problem, as I pointed out, is that the players should not have stopped play at that point. However the rules handle such positions, there will be an argument against what they do.

There does seem to be folk knowledge and gentlemens' agreements involved that is not explicitly included in the rules. Or does that go too far?

Go was played for centuries, perhaps millenia, without written rules. So, yes, everything was folk knowledge and gentlemen's agreements. Occasionally some rules problem arose, which the players usually took to a stronger player for adjudication. There are a couple of famous examples before the 20th century. Got that? A couple. Not much of a problem.

Then, in a team match in Japan in 1928, one of the players, Takahashi Shigeyuki, violated one of those gentlemen's agreements and refused to end the game as directed by the referee. (His teammates encouraged him in this.) For more, if you can stand it, see http://senseis.xmp.net/?TenThousandYearKo%2FrulesCrisis . One interesting thing about this rules dispute was that nobody could say whether the move was a right or an obligation. There were no passes in those days. The pass in go is a modern invention.

Nearly all go rules today end play by a succession of passes. But without passes, how did play end in those days? By agreement -- a gentlemen's agreement, if you will. (Note: Many no pass games reach a point where further play does no good and the result can be determined without play. That is a natural stopping point, and the players normally stop play by agreement. )

- Question: Wikipedia says under territory scoring “If the players reach an incorrect conclusion, then they both lose”. But how do we know the conclusion is incorrect? Who says? Also does it mean the disputed territory is not counted to either player, or that neither wins the game? (The latter seems harsh!)

Wikipedia is not quite correct. If play stops at an "awkward bit" position and the players do not resume play, then both players lose. That is harsh, but the Japanese 1989 rules were written by and for professionals. AFAIK, that rule has never been invoked.

- Question: Does area counting avoid all these difficulties?

No. However, the players can avoid or resolve many end of game disputes by play under area scoring, because filling your own territory costs nothing. Edit: Unless you fill an eye you need for life.

Why isn't this method (or other rule choice mentioned) adopted by all if so?

Mostly because the potential problems for territory scoring hardly ever arise. Except for beginners, unfortunately. But in the main places where territory scoring is used, beginners usually play against more experienced players who can resolve and explain life and death questions at the end of the game.

In the West, beginners often play beginners, which is why I recommend area scoring for beginners.

----

Actually, there is a form of go, called Button Go, which synthesizes territory and area scoring and, IMO, is better than either. But that is another discussion.

beginsA291 wrote: There does seem to be folk knowledge and gentlemens' agreements involved that is not explicitly included in the rules. Or does that go too far?

In theory the results follow from the rules, but the situations can be complicated as to why, and most people don't know the specific rulesets well enough to understand it. So, sort of. It's really much cleaner with area scoring, even though that was a later development.

- Question: Wikipedia says under territory scoring “If the players reach an incorrect conclusion, then they both lose”. But how do we know the conclusion is incorrect? Who says? Also does it mean the disputed territory is not counted to either player, or that neither wins the game? (The latter seems harsh!)

This only applies to the current Japanese rules, and is something you should ignore unless you plan to play in tournaments in Japan. Possibly only professional ones.

There's a bit of commentary on it here: http://www.cs.cmu.edu/~wjh/go/rules/Japanese.html. As I understand it, by the rules if you request to resume play, your opponent gets to play first. In those examples in the commentary, it means that whoever asks to resume play gives away the good move to their opponent, losing the game. Since each player would lose the game by requesting resumption, neither one does. However, the game is really uncountable because the status of the group is not determined. As such, each player is determined to have not taken the move needed to win, therefor they lost.

- Question: Does area counting avoid all these difficulties? Why isn't this method (or other rule choice mentioned) adopted by all if so?

It does for the most part. As for why it's not adopted everywhere, a variety of historical and political reasons. For one thing, there is no "ruling body of go" or anything like it. Different countries have different organizations. Area scoring was developed relatively late in China (possibly to help prevent cheating when gambling) after go had already taken hold in other countries like Japan and Korea. Counting method changes had already occurred in Japan and Korea by that point as well. Now that each has established go organizations with a high level of play, there aren't enough scoring disputes and issues to make it worthwhile for them to change. To look at another example, the innovation of starting with an empty board, instead of a preset starting position, dates back maybe 500 years or so to Japan. But it's only in the last century that each country agreed that standard go started with an empty board, and go with the various set starting positions were variants. In international tournaments, the rules or ruleset is generally specified beforehand. Even then, if two players mutually counted the score via territory scoring, say, when the rules specified area, I doubt anyone would care or notice unless there were a dispute or it was one of the top boards.

Second, in Asian countries, most people don't really worry about the rules at a high level like this at all. They are often passed orally between casual players or picked up through osmosis rather than by reading a document.Traditionally, for example, a triple ko is just a voided game, or no result, and the more complicated situations, like moonshine life, many players have never heard of or seen. In many rulesets, there is now a superko rule that determines in which cases a player may take each ko and in which they may not, but this is frequently ignored in casual play.

To take chess as an example here, most people learn the rules through a friend or family member, who guides them through the process rather than explaining all of the intricacies of stalemates and drawn games and such right away. No one not playing in a tournament of some sort worries that they've hit the 50-move rule.

In the west, go came to most places via Japan, so the base of players, small as it may have been, used traditional Japanese rules and scoring (territory scoring). The physical act of counting the game is quite different between area and territory scoring, so there's a lot of resistance to change in that respect. To avoid this, many western rulesets, like the AGA one, use area scoring behind the scenes and have some sort of trick to allow games to be counted in the traditional (territory) manner while coming up with the same score as an area-counting method.

Bill Spight wrote:Occasionally some rules problem arose [...] Got that? A couple. Not much of a problem.

One of those rules problems was the problem in Europe for a few centuries to understand what (besides its playing material) Go is by trying to understand what the (verbal Japanese) rules are.

One of those rules problems was the problem in Europe for a few centuries to understand what (besides its playing material) Go is by trying to understand what the (verbal Japanese) rules are.

When I learned go I used Japanese rules. Not 1949 Japanese rules, not Japanese 1989 rules, not 2000 Japanese rules, not Japanese rules with a twist of lemon. It was just rules as used in Japan, which may have been fuzzy around the edges once in a blue moon, but I never saw a blue moon. I only ever encountered a very slight difficulty once, when a 2-dan opponent wasn't aware that bent four is dead. He accepted his loss graciously once other strong players convinced him that was the rule - no explanation , just "that is the rule." And games ended just as graciously with a gentleman's agreement - just as they usually do now in most clubs.

Nowadays, thanks to the rules mavens, in the very occasional games I play I have absolutely no idea whether bent four is dead, whether I have to pass, or whether points count in a seki, and this is not at all to do with the fact I usually don't know what day of the week it is. The mavens have created confusion. They have added nothing to the game. They have taken away much. Too many cooks have not just spoiled the broth; they have pissed in it.

Which divinity ever said go has to be describable by perfect rules? Go is still mostly a social game. Society decides the rules.

John Fairbairn wrote:Which divinity ever said go has to be describable by perfect rules?

Whether unambiguity of rules of play is desirable depends on intentions. E.g., a yin-yang approach makes imperfection desirable according to Sakai Takeshi; automatic computer - computer play makes perfection desirable.

Go is still mostly a social game. Society decides the rules.

Go is also a study object and a computer game. There have been more than one society involved (such as Chinese and Japanese).

quoting myself, quoting tiger314, from another thread...

xed_over wrote:

tiger314 wrote:Generally speaking, Japanese rules assume both players play perfectly. The authors probably saw the fact that this is not the case as irrelevant

hahahaha... that is seriously, probably the best explanation of Japanese rules that I've ever read!