It is currently Thu Oct 21, 2021 1:04 am

All times are UTC - 8 hours [ DST ]




Post new topic Reply to topic  [ 100 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next
Author Message
Online
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #41 Posted: Tue Sep 28, 2021 3:17 am 
Lives with ko

Posts: 201
Liked others: 17
Was liked: 20
Rank: panda 4 dan
IGS: kvasir
jann wrote:
Now I'm not sure again if I understand the idea well, how would this prevent the closed double ko loop for example?


I don't think it would do anything about the double-ko cycle.

This was a response to positions like this one were one side seems to get multiple moves in a row. Possibly, you wouldn't need to handle this kind of a double-ko like situation with a no-ban rule, because the cycle has a different cause.

Click Here To Show Diagram Code
[go]$$B
$$ -------------------
$$ | O O X . X X O . .
$$ | O O O X . X O . .
$$ | X O O O X O O . .
$$ | . X O . X X O . .
$$ | X . X X . X O . .
$$ | X X O X X X O . .
$$ | O O O O O O O . .
$$ | . . . . . . . . .
$$ | . . . . . . . . .[/go]


=== Edit === I shouldn't have said the "cycle" has a different cause but the problem has a different cause then the cycle.

Top
 Profile  
 
Offline
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #42 Posted: Tue Sep 28, 2021 5:40 am 
Lives in gote

Posts: 380
Liked others: 0
Was liked: 32
Gérard TAILLE wrote:
BTW has somebody found a position in which the "no ban in double ko" rule associated to the "traditional" pass-for-ko (for each ko) contradicts the expected japanese result?

I have no idea since that rule is way too complicated, but one thing you may want to look at is the internal triple ko between two groups where one side has an eye and the other holds two of the three kos.

Top
 Profile  
 
Offline
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #43 Posted: Tue Sep 28, 2021 6:50 am 
Lives in sente

Posts: 821
Liked others: 14
Was liked: 37
Rank: 1er dan
jann wrote:
Gérard TAILLE wrote:
BTW has somebody found a position in which the "no ban in double ko" rule associated to the "traditional" pass-for-ko (for each ko) contradicts the expected japanese result?

I have no idea since that rule is way too complicated, but one thing you may want to look at is the internal triple ko between two groups where one side has an eye and the other holds two of the three kos.


Interesting suggestion Jann. That's allow me to clarify my proposal.

Click Here To Show Diagram Code
[go]$$B
$$ -------------------------
$$ | O O O O O O X . X O . .
$$ | . O X O . O X X X O . .
$$ | O X . X O X X O O O . .
$$ | X X X X X X O O . . . .
$$ | O O O O O O O . . . . .
$$ | . . . . . . . . . . . .
$$ | . . . . . . . . . . . .[/go]


The main idea of my proposal is to recognize a strong double ko (between a white group and a black group) and to decide that the corresponding two ko are miai ko (handled by regular ko ban). ALL other ko adjacent to these two groups are handled in confirmation phase by the "traditional" pass-for-ko.

In the position above with three ko between two groups we have only to choose two ko as being miai ko and the third one will be automatically handled by the pass-for-ko rule.
Let's assume you choose the left two ko being miai ko (handled by regular ko ban) and the third one (on the right) handled by the pass-for-ko request.

Black to play:
Click Here To Show Diagram Code
[go]$$B
$$ -------------------------
$$ | O O O O O O X . X O . .
$$ | 3 O X O 1 O X X X O . .
$$ | O X 2 X O X X O O O . .
$$ | X X X X X X O O . . . .
$$ | O O O O O O O . . . . .
$$ | . . . . . . . . . . . .
$$ | . . . . . . . . . . . .[/go]


Click Here To Show Diagram Code
[go]$$B
$$ -------------------------
$$ | O O O O O O X . X O . .
$$ | X O . O X O X X X O . .
$$ | . X O X M X X O O O . .
$$ | X X X X X X O O . . . .
$$ | O O O O O O O . . . . .
$$ | . . . . . . . . . . . .
$$ | . . . . . . . . . . . .[/go]
and white is dead because two ko are banned (the left one by regular ko ban, and the right by the pass-for-ko request)

White to play:
Click Here To Show Diagram Code
[go]$$W :w3: pass
$$ -------------------------
$$ | O O O O O O X . X O . .
$$ | 2 O X O . O X X X O . .
$$ | O X 1 X O X X O O O . .
$$ | X X X X X X O O . . . .
$$ | O O O O O O O . . . . .
$$ | . . . . . . . . . . . .
$$ | . . . . . . . . . . . .[/go]
and now it is black to play and black can kill white as previously.

Note: you can easily see that it does not matter which ko you choose to be miai ko.
Maybe my wording was a little complicated (I am not an expert in wording) but I am sure any player can very easily recognize that two ko are miai ko and have to be handled accordingly.

Top
 Profile  
 
Offline
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #44 Posted: Tue Sep 28, 2021 2:51 pm 
Lives in gote

Posts: 380
Liked others: 0
Was liked: 32
kvasir wrote:
This was a response to positions like this one were one side seems to get multiple moves in a row.

If your target are those reinforcement problems, then you may want to look at their original (from Davies I think):
Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | . . X X X X O X . .
$$ | X X O X O X O X . .
$$ | X O . O . O O X . .
$$ | O . O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]

Top
 Profile  
 
Offline
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #45 Posted: Tue Sep 28, 2021 3:03 pm 
Lives in gote

Posts: 380
Liked others: 0
Was liked: 32
Gérard TAILLE wrote:
Maybe my wording was a little complicated (I am not an expert in wording) but I am sure any player can very easily recognize that two ko are miai ko and have to be handled accordingly.

I never complain or nitpick about wordings, but you also seem to have conceptual problems here.

Quote:
That's allow me to clarify my proposal.
Click Here To Show Diagram Code
[go]$$B
$$ -------------------------
$$ | O O O O O O X . X O . .
$$ | . O X O . O X X X O . .
$$ | O X . X O X X O O O . .
$$ | X X X X X X O O . . . .
$$ | O O O O O O O . . . . .
$$ | . . . . . . . . . . . .
$$ | . . . . . . . . . . . .[/go]

In the position above with three ko between two groups we have only to choose two ko as being miai ko and the third one will be automatically handled by the pass-for-ko rule.

This does not seem obvious. Above you have three overlapping double ko pairs. As you now clarified the player(s) have to choose (randomly) one of the three pairs and apply your rule for that, and at the same time they must not apply the rule for the other two (potential) pairs. In reality this is a moonshine life where the double ko pair is shifting continuously among potential pairs.

Top
 Profile  
 
Online
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #46 Posted: Wed Sep 29, 2021 3:16 am 
Lives with ko

Posts: 201
Liked others: 17
Was liked: 20
Rank: panda 4 dan
IGS: kvasir
jann wrote:
kvasir wrote:
This was a response to positions like this one were one side seems to get multiple moves in a row.

If your target are those reinforcement problems, then you may want to look at their original (from Davies I think):


For your or Davies reinforcement problem, it is really not affected in anyway by limiting the number of pass-ko bans. I don't think just repeating some explanation for what I mean by limiting the number of pass-ko bans is helpful right now (it is probably a faulty explanation or we miscommunicated something), so I will leave that for later.

=== My understanding of this position===

In j89 (only?) white is dead in status confirmation because black can play like this:

Click Here To Show Diagram Code
[go]$$B :w4: pass-for-ko
$$ ---------------------
$$ | . . X X X X O X . .
$$ | X X O X O X O X . .
$$ | X O 1 O 3 O O X . .
$$ | O 2 O 5 O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]


It is interesting to compare it to this one when black took in the double-ko before status confirmation because now white is alive in j89.

Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | . . X X X X O X . .
$$ | X X . X O X O X . .
$$ | X O X O . O O X . .
$$ | O O O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]


Note that black is dead in this case too
Click Here To Show Diagram Code
[go]$$W
$$ ---------------------
$$ | . 1 X X X X O X . .
$$ | X X . X O X O X . .
$$ | X O X O . O O X . .
$$ | O O O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]


If black had a stone on :w1: it appears that we would have no way around the double-ko pass-ko cycle. Could we actually ask if black needs to reinforce :scratch:


=====
So, are we to conclude j89 is not "not bad" now? I don't know, but would you expect someone to protect in the first position as white?

At any rate, this is consistent with j89 intention to not allow one side to defend this kind of a ko shape. To explain what I mean, take this shape when both are alive and we ask if white can leave the kos unfilled in case white can recycle ko threats elsewhere.

Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | X . . . . X X X
$$ | . X X X X X O X . .
$$ | X X O X O X O X . .
$$ | X O . O . O O X . .
$$ | O . O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]



====
I am interested how these positions are identified as "strong double ko", "weak double ko" and so on and if the no-ban rule would apply.

Top
 Profile  
 
Offline
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #47 Posted: Wed Sep 29, 2021 6:21 am 
Lives in sente

Posts: 821
Liked others: 14
Was liked: 37
Rank: 1er dan
jann wrote:
Gérard TAILLE wrote:
Maybe my wording was a little complicated (I am not an expert in wording) but I am sure any player can very easily recognize that two ko are miai ko and have to be handled accordingly.

I never complain or nitpick about wordings, but you also seem to have conceptual problems here.

Quote:
That's allow me to clarify my proposal.
Click Here To Show Diagram Code
[go]$$B
$$ -------------------------
$$ | O O O O O O X . X O . .
$$ | . O X O . O X X X O . .
$$ | O X . X O X X O O O . .
$$ | X X X X X X O O . . . .
$$ | O O O O O O O . . . . .
$$ | . . . . . . . . . . . .
$$ | . . . . . . . . . . . .[/go]

In the position above with three ko between two groups we have only to choose two ko as being miai ko and the third one will be automatically handled by the pass-for-ko rule.

This does not seem obvious. Above you have three overlapping double ko pairs. As you now clarified the player(s) have to choose (randomly) one of the three pairs and apply your rule for that, and at the same time they must not apply the rule for the other two (potential) pairs. In reality this is a moonshine life where the double ko pair is shifting continuously among potential pairs.


As you understood, when more that two ko exist between a black group and a white group my idea is to choose one pair of ko and this choice will never change providing OC the two ko remain unresolved. To avoid making a randomly choice (which looks strange indeed) my proposal is that the attacker (I mean the player allowed to start the confirmation play) must make this choice as soon as this multiple ko situation appear. Here again as soon as this choice has been made it could not be changed providing the two ko remain unresolved.

In the above position, in my previous post I chose the two ko on the left, and the expected japanese result is reached.

Top
 Profile  
 
Offline
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #48 Posted: Wed Sep 29, 2021 6:36 am 
Lives in sente

Posts: 821
Liked others: 14
Was liked: 37
Rank: 1er dan
jann wrote:
kvasir wrote:
This was a response to positions like this one were one side seems to get multiple moves in a row.

If your target are those reinforcement problems, then you may want to look at their original (from Davies I think):
Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | . . X X X X O X . .
$$ | X X O X O X O X . .
$$ | X O . O . O O X . .
$$ | O . O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]


What is the expected result? It looks obvious: in any area scoring rule and in NORMAL play of any japanese rule black is captured even if it is black to play => black is dead and white has not to add a reinforcement move.
In J89 it is not the case but with "no ban in double ko" rule it is the case OC.

Top
 Profile  
 
Offline
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #49 Posted: Wed Sep 29, 2021 7:23 am 
Lives in sente

Posts: 821
Liked others: 14
Was liked: 37
Rank: 1er dan
kvasir wrote:
At any rate, this is consistent with j89 intention to not allow one side to defend this kind of a ko shape. To explain what I mean, take this shape when both are alive and we ask if white can leave the kos unfilled in case white can recycle ko threats elsewhere.

Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | X . . . . X X X
$$ | . X X X X X O X . .
$$ | X X O X O X O X . .
$$ | X O . O . O O X . .
$$ | O . O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]



====
I am interested how these positions are identified as "strong double ko", "weak double ko" and so on and if the no-ban rule would apply.


I do not see what white expects by passing in this quite unsettled position.
It seems to me that white should continue in normal play by:
Click Here To Show Diagram Code
[go]$$W
$$ ---------------------
$$ | X . . . . X X X
$$ | . X X X X X O X . .
$$ | X X O X O X O X . .
$$ | X O 1 O . O O X . .
$$ | O . O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]
and then white will connect the other ko by using her other ko threats => white score 3 points of territory.
Do white expect more by passing? If white passes black in normal play will be always able to capture and connect one ko => white loses one point.

OC if white has an infinite number of ko threats it is another point because you would introduce a NO RESULT possibility.

Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | X . . . . X X X
$$ | . X X X X X O X . .
$$ | X X O X O X O X . .
$$ | X O 1 O . O O X . .
$$ | O 2 O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]
BTW after the exchange :b1: :w2: the remmaining two ko form a "strong ko" but I guess you will conclude that the position is in anycase unsettled => one of the player should continue the game.

Top
 Profile  
 
Online
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #50 Posted: Wed Sep 29, 2021 7:44 am 
Lives with ko

Posts: 201
Liked others: 17
Was liked: 20
Rank: panda 4 dan
IGS: kvasir
Gérard TAILLE wrote:
I do not see what white expects by passing in this quite unsettled position.


This one was meant as an example of how pass-ko does work correctly to prevent white from ending the game at this point. The need to reinforce in the last ko was disputed in Japan for a time before written rules. My point was that since this is the same shape as in the reinforcement problem, and we seem to use the same pass-ko argument, that we can't just brush away the need to reinforce in the first problem.

Top
 Profile  
 
Offline
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #51 Posted: Wed Sep 29, 2021 7:58 am 
Lives in sente

Posts: 821
Liked others: 14
Was liked: 37
Rank: 1er dan
kvasir wrote:
Gérard TAILLE wrote:
I do not see what white expects by passing in this quite unsettled position.


This one was meant as an example of how pass-ko does work correctly to prevent white from ending the game at this point. The need to reinforce in the last ko was disputed in Japan for a time before written rules. My point was that since this is the same shape as in the reinforcement problem, and we seem to use the same pass-ko argument, that we can't just brush away the need to reinforce in the first problem.


Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | . . X X X X O X . .
$$ | X X O X O X O X . .
$$ | X O . O . O O X . .
$$ | O . O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]

Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | X . . . . X X X
$$ | . X X X X X O X . .
$$ | X X O X O X O X . .
$$ | X O . O . O O X . .
$$ | O . O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]


I do not see really the similarity.
Assume your are still in NORMAL play.
In the first position black is dead even if black plays first.
In the second position white loses points if black plays first.
=> reinforcement is necessary in the second position but not in the first one.

Top
 Profile  
 
Online
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #52 Posted: Wed Sep 29, 2021 8:32 am 
Lives with ko

Posts: 201
Liked others: 17
Was liked: 20
Rank: panda 4 dan
IGS: kvasir
This comment was about pass-ko in j89 status confirmation. Both black and white are dead in j89 status confirmation in the first position (I gave the line for capturing white already), it would seem that white is dead in exactly the same way in the second one. Brushing aside the need for a reinforcement move in the first positions doesn't seem to me to be valid in j89 status confirmation. I was curious if your no-ban rule would change how black can capture white during status confirmation, or if this issue was outside of scope.

Top
 Profile  
 
Online
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #53 Posted: Wed Sep 29, 2021 8:56 am 
Lives in sente
User avatar

Posts: 1198
Liked others: 11
Was liked: 136
Rank: German 1 Kyu
kvasir wrote:
Brushing aside the need for a reinforcement move in the first positions doesn't seem to me to be valid in j89 status confirmation.

In J89's status confirmation, capturing into a single ko-shape enables the capturer to play TWO moves in a row (locally).

If such a ko capture is sente (for what reason ever), the capturer is enabled to play THREE moves in a row (if White's answer is considered to be enforced, and therefore is not counted).
And so on for any additional of such sente ko captures.

This makes clear that it is Japanese understanding that the defender in these cases has to add a protective move during actual play, in order to get rid of such "open" ko-shapes.

_________________
The really most difficult Go problem ever: http://igohatsuyoron120.de/index.htm
Igo Hatsuyoron #120 (still unresolved by professionals, maybe solved by four amateurs)

Top
 Profile  
 
Offline
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #54 Posted: Wed Sep 29, 2021 11:16 am 
Lives in sente

Posts: 821
Liked others: 14
Was liked: 37
Rank: 1er dan
kvasir wrote:
This comment was about pass-ko in j89 status confirmation. Both black and white are dead in j89 status confirmation in the first position (I gave the line for capturing white already), it would seem that white is dead in exactly the same way in the second one. Brushing aside the need for a reinforcement move in the first positions doesn't seem to me to be valid in j89 status confirmation. I was curious if your no-ban rule would change how black can capture white during status confirmation, or if this issue was outside of scope.

Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | . . X X X X O X . .
$$ | X X O X O X O X . .
$$ | X O . O . O O X . .
$$ | O . O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]


Maybe I am wrong but when in a position like the one above, black is dead in all area scoring rules, and is dead in NORMAL play in any japanese rule, even if it black to move, then I consider the intention of japanese rule should be that white does not need a reinforcement move.
In J89 it is not the case => it is an unexpected side effect of pass-for-ko rule (as is the closed cycle in the double ko).

How it works with "no ban in double ko" rule:

Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | . . X X X X O X . .
$$ | X X O X O X O X . .
$$ | X O 1 O 3 O O X . .
$$ | O 2 O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]


Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | . . X X X X O X . .
$$ | X X M X M X O X . .
$$ | X O X O X O O X . .
$$ | O O O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]
the two ko are "strong double ko" and the position is critical (black has taken the two ko) => white can ignore the pass-for-ko request.

Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | . . X X X X O X . .
$$ | X X 4 X M X O X . .
$$ | X O X O X O O X . .
$$ | O O O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]
the move :w4: is allowed by the "no ban in double ko" rule (white cannot take the other ko due to the regular ko ban).

Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | . . X X X X O X . .
$$ | X X O X M X O X . .
$$ | X O M O X O O X . .
$$ | O O O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]
After white :w4: the pass-for-ko request remain active because the position is no more critical => white while be able to connect her ko making two eyes and black group will be dead => white, in the initial position, do not need to add a move.


Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | X . . . . X X X
$$ | . X X X X X O X . .
$$ | X X W X W X O X . .
$$ | X O . O . O O X . .
$$ | O . O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]
This position is quite different because black is already alive. The "no ban in double ko" rule avoid to declare all white stones dead.

The result is exactly the same as the following position
Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | X . . . . X X X
$$ | . X X X X X O X . .
$$ | X X . X O X O X . .
$$ | X O X O . O O X . .
$$ | O O O O O O X X . .
$$ | O . O . O X X . . .
$$ | O O O O X . . . .
$$ | X X X X X . . . . .[/go]

I suspect that J89 conclude to a large seki due to the unresolved ko => that is the result with "no ban in double ko" rule.

Top
 Profile  
 
Offline
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #55 Posted: Wed Sep 29, 2021 3:35 pm 
Lives in gote

Posts: 380
Liked others: 0
Was liked: 32
kvasir wrote:
The need to reinforce in the last ko was disputed in Japan for a time before written rules.

This was at a time when pass was not yet a move, and games did not end or stop on two passes as today. So I doubt the analogy to these reinforcement problems would hold (different mechanisms).

kvasir wrote:
So, are we to conclude j89 is not "not bad" now? I don't know, but would you expect someone to protect in the first position as white?

I would say not bad, since these problems are relatively minor. About the same level as 3-torazu: an unfortunate but natural and maybe tolerable side effect of the chosen approach, for one point scoring error. I think the bent4 problems from the other topic are more serious, especially considering the inconsistency (three different and incorrect rulings given for two almost identical positions, when Korean rules are also included).

J89 simplifies ko play for L/D, using ko rules that changed heavily from normal ko. It is actually surprising this only causes such minor anomalies (but who knows what will be found later). These reinforcement problems are incorrect and are defects, but caused naturally by J89 ko. So if one doesn't like them the right way is not to invent further complicated hacks or patches, but move towards confirmation with normal ko.

Whatever difference you find compared to continued normal play, it almost always means your new rules are at fault (except moonshine life).

Top
 Profile  
 
Online
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #56 Posted: Wed Sep 29, 2021 4:39 pm 
Lives with ko

Posts: 201
Liked others: 17
Was liked: 20
Rank: panda 4 dan
IGS: kvasir
jann wrote:
kvasir wrote:
The need to reinforce in the last ko was disputed in Japan for a time before written rules.

This was at a time when pass was not yet a move, and games did not end or stop on two passes as today. So I doubt the analogy to these reinforcement problems would hold (different mechanisms).


In j89 you need to connect the last ko no matter if you have infinite ko threats or not. It was just an example to show that it is not possible to always say "it is an anomaly so we will ignore that".

jann wrote:
So if one doesn't like them the right way is not to invent further complicated hacks or patches, but move towards confirmation with normal ko.


It is not that I like or dislike them, it is just that the rules say one thing and people do a completely different thing. I don't see why it is a hack to try to describe what people actually do. In any case it is not like we are actually changing these rules.

Top
 Profile  
 
Offline
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #57 Posted: Wed Sep 29, 2021 5:03 pm 
Lives in gote

Posts: 380
Liked others: 0
Was liked: 32
By hacks I mean exceptional rules that do not have a theoretical rationale or go against one. J89 ko has its logic, which one either accepts or not. If not: fine there are other ways. Same as pass stones: if you accept them, one sided dame become points. It would be doubtful try to patch the rule to exclude that particular consequence only.

Top
 Profile  
 
Online
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #58 Posted: Wed Sep 29, 2021 5:54 pm 
Lives with ko

Posts: 201
Liked others: 17
Was liked: 20
Rank: panda 4 dan
IGS: kvasir
I think there is quite a bit of selective reading of j89 around. Passing once for a ko, stating that the number of captured stones matter during status confirmation, brushing aside the pass-ko cycle, refusing that dame means seki, and so on. I don't see why j89 can't both have a clear logic and be open to discussion of why something is needed, if something could be different, and just interpretations.

If it didn't have an attractive logic, no one would actually care whatsoever.

Top
 Profile  
 
Offline
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #59 Posted: Wed Sep 29, 2021 6:04 pm 
Lives in gote

Posts: 380
Liked others: 0
Was liked: 32
Sure, I have nothing against discussing anything, just gave my (prejudiced) opinion. I think most rule problems are born from such exceptional rule inventions (without theoretical ground), moonshine life is about the only problem inherent in normal go. OC J89 ko is also a bit doubtful, even theoretically, so I'd include it in that list (none of those bent4 examples are problems for normal ko).

Top
 Profile  
 
Offline
 Post subject: Re: The "no ban in double ko" rule to solve pass-for-ko issu
Post #60 Posted: Wed Sep 29, 2021 7:48 pm 
Lives in gote

Posts: 380
Liked others: 0
Was liked: 32
Btw - NOT as serious suggestion, just a random idea for theory consideration: reinforcement problems can be examined from J89 view on the basis of what "the same ko" means.

Click Here To Show Diagram Code
[go]$$B
$$ ---------------------
$$ | . . X X X X O X . .
$$ | X X O X O X O X . .
$$ | X O . O . O O X . .
$$ | O . O . O O X X . .
$$ | O . O O O X X . . .
$$ | O O O X X X . . . .
$$ | X X X X . . . . . .[/go]

Click Here To Show Diagram Code
[go]$$B
$$ -------------------
$$ | O O X . X X O . .
$$ | O O O X . X O . .
$$ | X O O O X O O . .
$$ | . X O . X X O . .
$$ | X . X X . X O . .
$$ | X X O X X X O . .
$$ | O O O O O O O . .
$$ | . . . . . . . . .[/go]

In both of the above cases the stones making the ko shape get changed initially (become a different string). One (not me) may argue this is not necessarily the "same ko" afterwards. Oc this wouldn't affect lightvector's example or its derivative.

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 100 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next

All times are UTC - 8 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group