How do Japanese rules handle this?

[Gérard TAILLE]

I agree completly with your text but the wording you proposed does not look perfect.

$$
$$ -----------------------------
$$ | . X X O . O X . . . X . Q |
$$ | O O O O O O X . . . X X . |
$$ | X X X X X X X . . . . X X |
$$ | . . . . . . . . . . . X . |

Click Here To Show Diagram Code

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

What is the status of the white marked stone? No doubt it is dead but what about your wording, strictly speaking ?

Yes, this is exactly why I suggested adding the two B stones to your earlier position, to ensure B will have the choice even with W starting. B will find it hard to prove that W could have played either spot, for equal consequence, just only on B's choice (probably not impossible to prove, but still not nice).

IMO the real issue here is that J89 chose to formulate the rule around new STONES. It would be probably more robust if the rule itself would have been formulated differently:

• A string is dead if the opponent can take control (occupy or surround by uncapturable stones) of all its intersections, without allowing the player to control another intersection in exchange, that he couldn't have controlled otherwise.

This would probably be similar to your proposal - and interestingly, this would flip the ruling in lightvector's position:

$$
$$ | . . . . . .
$$ | X X X X X .
$$ | X O O O . .
$$ | W X X O . .
$$ | . X O O . .
$$ | X X O . . .
$$ | O X O . . .
$$ | . O O . . .
$$ -----------

Click Here To Show Diagram Code

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

Oops, I did not expect you go that far because now it looks like my proposal!

When you say:
"without allowing the player to control another intersection in exchange"
it seems that if the player can TURN a group of stones to be uncapturable then this player was allowed to CONTROL another intersection doesn't it?

Anyway, even if you do not go so far, I agree with your approach which is quite similar if not identical to mine. Do not forget that if the differences concerned only unfinished positions then the final result maybe always the same.

BTW, I agree that an empty intersection, surrounded by uncapturable STONES, is a controlled intersection, but it is not true that a STRING of empty intersections, surrounded by uncapturable STONES, are controlled intersections. Here again the wording has to be chosen carefully.

[Cassandra]

...but it is not true that a STRING of empty intersections, surrounded by uncapturable STONES, are controlled intersections.

As long as the "string" is not large enough to contain a permanent group of the opponent, it is of course controlled by the player.

[kvasir]

$$
$$ -------------
$$ | . X X O . .
$$ | O X X O . .
$$ | X O O O . .
$$ | X X X . . .
$$ | . . . . . .

Click Here To Show Diagram Code

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

For me it does not matter if the result is seki or +3 for black because in any case white will continue the game. See on the subject the post viewtopic.php?p=267480#p267480
BTW I prefer the result +3 for black but a seki doesn't harm.

I replied that ZERO is a special value in such positions but CGT/Shuwa (3 points for white) requires that we compute a value for every such position, it is attracted as an idea but impractical. When you say it is +3 for black that is something else, you are claiming the white stone is dead. Let's not muddy the water by pretending that is any different than claiming is +9 for white.

Claiming that this example is not ZERO, that the rules text is wrong and so on is bound to raise the question what this has to do with J89.

Positions where a teire move is necessary in one interpretation and not in an other interpretation are more problematic because the results of the game are really different.

'teire' is a serious matter for sure but this thread has degenerated.

[...]the written rule clearly require some kind of causal connection between the new stones and the capture (or the captured string at least).

Requiring a causal connection is a very weak requirement. For example you said my explanation might be described as (always) "involving" new stones but that would still be a causal connection; this could also be called 'creating new stones as a side effect'. I don't think it is really the causal connection that you want to require, as far as I understand I think you want to require the 'enabled' stones to be played were the opposing color could prevent them from being played.

Robert Jaseki has offered positions were it is rather nonsensical because the 'new stones' need to be played in such a way as to fill an eye. It also appears to contradict multiple examples in J89.

My opinion is simply that there is no reason to read this requirement into J89, when it fails spectacularly(?) we should realize that this is not the meaning of J89. Just to be clear, I think the J89 authors left the 'enable' business mostly undefined. Trying to be very strict about it is just not going to work.

Still, I yet to see any other interpretation that would be simply and logically evaluatable in all examples, without being clearly abusable in other examples. If you know such, please do share.

In my opinion, going from 'enable' that is mostly undefined to 'enable' that is usable in any position is likely not going to be identifiable as J89. Take J2003 for example, a side from many of the most critical sentences being undecipherable, the whole framework is really something completely different from J89. I am more interested in workable understanding than something that doesn't fail in wild situations.

I gave an example of how many of the intersection in the original position (this thread started when lightvector asked a question about a specific position) would be placed differently in relation to the boundary between living and dead groups depending on the outcome of status confirmation, then I argued that at least some of these points should count as valid for 'new stones' basically because black captures something but loses control over the boundary. I think I used the phrase "new ground". You called this a 'locality hack' that 'involves' new stones instead of actually enabling them, I'd maybe call it something else (boundary hack?), and it is sure to fail on many positions (like everything else). While this might not be easy to apply to every imaginable situation, I think it can be a workable explanation for many normal situations. I'd give it more thought but I don't have time for the kind of wild positions this sub-topic generates every time someone states something that is directly applicable to life and death confirmation and can actually be tested.

Also why engage in detailed analysis of explanations and interpretations in every thread in the Go Rules forum?

[kvasir]

And it's nice to talk about things that are not the rules, but what does your dictionary say about 確認? Because my dictionary defines it using 争い and 疑い.

Common! It is 'confirming' same as in 'confirming a hotel booking', 'acknowledge' as in 'acknowledge a receipt' and so forth.

Even these situations involve something unknown -- whether the reservation was actually booked or not. And even then, it's not like the hotel can't become overbooked. Some confirmation that is.

You are basing a rather complicated argument on the sole foundation that the word 確認/确认 allows you to EXCLUDE anything that is already known from consideration. Lot of people know what it means, this is not some advanced vocabulary with rich and complicated nuances. The two players are just pointing out groups that need to be removed from the board and identifying seki. There is absolutely nothing implied about the method or process, except maybe that it doesn't need many words.

[CDavis7M]

You are basing a rather complicated argument on the sole foundation that the word 確認/确认 allows you to EXCLUDE anything that is already known from consideration. Lot of people know what it means, this is not some advanced vocabulary with rich and complicated nuances. The two players are just pointing out groups that need to be removed from the board and identifying seki. There is absolutely nothing implied about the method or process, except maybe that it doesn't need many words.

It's not that there are nuances. It's that some interpretations ignore the meaning of the word. And I'm definitely not excluding known things. The known status is in the consideration, it just isn't the confirmation itself.

For example, in the position being discussed, when White captures the ko back and then captures another Black stone, these newly placed White stones can have an uncapturable shape along with the White group with 2 eyes. So that group with 2 eyes is being considered. But it is the fact that White can capture Black to form an uncapturable shape that confirms White's living status. Directly adding stones to the group with 2 eyes to create uncapturable stones confirms nothing about L&D. Capturing stones to create the uncapturable shape is what confirms living status.

The fact that the 1 White stone can be captured and then can go back and capture the Black stones to create an uncapturable shape, all without the 5 White stones in the corner being captured, confirms that the 5 stones are dead -- any of the various forms of uncapturable stones that could be formed can be formed from the capture of the 1 stone without the 5 stones being captured.

[Gérard TAILLE]

$$
$$ -------------
$$ | . X X O . .
$$ | O X X O . .
$$ | X O O O . .
$$ | X X X . . .
$$ | . . . . . .

Click Here To Show Diagram Code

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

For me it does not matter if the result is seki or +3 for black because in any case white will continue the game. See on the subject the post viewtopic.php?p=267480#p267480
BTW I prefer the result +3 for black but a seki doesn't harm.

I replied that ZERO is a special value in such positions but CGT/Shuwa (3 points for white) requires that we compute a value for every such position, it is attracted as an idea but impractical. When you say it is +3 for black that is something else, you are claiming the white stone is dead. Let's not muddy the water by pretending that is any different than claiming is +9 for white.

Claiming that this example is not ZERO, that the rules text is wrong and so on is bound to raise the question what this has to do with J89.

When we try to apply the article 7.1 of J89 to the diagram above (example 1 of the rule) we all agree that the four black stones are dead but it it is not clear what is the status of the one white stone. Some of us undertand it is dead and others understand it is alive. In addition, even with the same result, different reasonning are applied that prove that the wording of J89 is ambiguous.
In this context I am happy to see the explanation given by all of us in order to have a better understanding of the problem.
BTW I understood you consider the white stone being alive (I know it is what is said explicitly in the example 1 of the rule but you know also that some of us consider it is not really consistant with article 7.1). OC a new stone with no relation with the group of stone under consideration has not to be taken into account. How do you define when a new stone has some relation with the group under consideration? This question is essential to be able to give the status of a slightly different position with is not mentionned through the examples of the rule.

[kvasir]

I just don't understand the argument. You said it had to do with the meaning of 'confirming', 'uncertainty' and other words.

The fact that the 1 White stone can be captured and then can go back and capture the Black stones to create an uncapturable shape, all without the 5 White stones in the corner being captured, confirms that the 5 stones are dead -- any of the various forms of uncapturable stones that could be formed can be formed from the capture of the 1 stone without the 5 stones being captured.

If I were to say that this does instead confirm that the black stones are dead. Would that mean the 1 stone was dead because there is nothing to 'confirm' about the 1 stone? Everything about the dead black stones is confirmed, therefore nothing left to confirm about the 1 stone.

I suppose you would not agree that this is a similar argument.

[kvasir]

When we try to apply the article 7.1 of J89 to the diagram above (example 1 of the rule) we all agree that the four black stones are dead but it it is not clear what is the status of the one white stone.

I don't remember agreeing that the black stones are dead, just that it is interesting that we get the same result of the game if we consider black and white alive as if we consider them dead. It is not at all clear why you are claiming that black is dead but maybe not white. Why would it not be the same? Almost the same thing happens when you capture either black or white.

[John Fairbairn]

Would someone be willing to try to explain to me what this and similar threads are actually about?

As far as I can tell, almost everyone agrees that the Japanese rules are, in a theoretical sense, a mess. But, at the same time, almost everyone outside China agrees that Japanese rules in a practical sense are just fine.

The argument seems to be about the confirmation phase. But I have never (I think) come across a single case where a confirmation phase has ever been needed. But if such a case were to arise in Japanese pro play, I find it very hard to imagine that anything said here (or by Robert Jasiek - I'm puzzled by his absence here) would influence the decision made by a Japanese referee. I find it equally hard to imagine the rules being changed because of anything said here.

I understand that "beasts" are interesting to those of a mathematical bent, but I don't see that we are talking about beasts here. It seems like going to the zoo to see the animals and ending up arguing about the quality of the hamburger in the cafe.

On the (safe, I think) assumption that it's not a case of trolling, and the hopeful assumption that it's not a case of an egos ladder competition, what is it all about. My tentative conclusion is that, to continue the zoological analogy, it's just a case of passing the time by pulling the legs off spiders.

[jann]

When we try to apply the article 7.1 of J89 to the diagram above (example 1 of the rule) we all agree that the four black stones are dead

Is this a typo? IMO B is alive, since it is partially reestablishable.

Claiming that this example is not ZERO, that the rules text is wrong and so on is bound to raise the question what this has to do with J89.
...
My opinion is simply that there is no reason to read this requirement into J89, when it fails spectacularly(?)

I think the traditional interpretation does NOT fail. It actually works well, and provides a robust L/D theory - no small feat - with clear reasoning of the "whys". And nobody said "the rules text is wrong". I (we) merely notice that the commentary on example 1 contradicts the actual rules text (7.1), especially in the English version (which is debatable OC but still). And for the above reason the rules text make more sense than the commentary here.

For example you said my explanation might be described as (always) "involving" new stones but that would still be a causal connection; this could also be called 'creating new stones as a side effect'.

The problem is "involve" does not make a clear, logically verifiable claim. If a player says that a capture "involves" a new stone on one-sided dame, how do you refute it? The capture cannot be performed without it if the defender is allowed to play freely - and no stone is ever enabled if he isn't.

I guess the "involve" approach would consist of playing out the capture, where both sides play for their original goal (maximize territory), and see what new stones were played in the vicinity. But playing out the capture already loses points, so it doesn't seem clear what logic would dictate the moves in this case. And I doubt this could compare to the clarity and robustness of the "made possible" approach.

[jann]

As far as I can tell, almost everyone agrees that the Japanese rules are, in a theoretical sense, a mess.
...
But I have never (I think) come across a single case where a confirmation phase has ever been needed.

I think J89 is much better than usually claimed, many problems (like with double kos) can be explained as simple misunderstandings. And it is oc true that after kyu levels there are almost no disputes, so just scoring the board on agreement is nearly 100% as a rule.

But Go has a strange kind of completeness. Once you make an actual rule with verifiable claims and consequences, it is usually possible to create an artificial counterexample. The rules game is similarly interesting as the real one. Why people do jigsaw puzzles?

[Gérard TAILLE]

When we try to apply the article 7.1 of J89 to the diagram above (example 1 of the rule) we all agree that the four black stones are dead but it it is not clear what is the status of the one white stone.

I don't remember agreeing that the black stones are dead, just that it is interesting that we get the same result of the game if we consider black and white alive as if we consider them dead. It is not at all clear why you are claiming that black is dead but maybe not white. Why would it not be the same? Almost the same thing happens when you capture either black or white.

Yes Kvasir OC a typing error. Sorry.
The four black stones are alive. The issue is only on white stone.

[kvasir]

Would someone be willing to try to explain to me what this and similar threads are actually about?

Usually it starts as a serious discussion about something, like this thread, but it usually degenerates into discussion about hypotheticals. After that it seems to really be anyone's guess what it is about. Someone was bound to bring up 'trolling'.

[Gérard TAILLE]

As far as I am concerned I consider J89 being a good rule which is perfect in practice in almost 100% of the games but that does not prevent me from liking discussion on beast positions. BTW seeing the difficulty to find a wording corresponding to my own interpretation I consider, as Jann said, that J89 is much better than usually claimed.

$$
$$ -----------------------
$$ | O . O . X . . O O . O
$$ | O O O X O X O O O O O
$$ | X X X O . O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .

Click Here To Show Diagram Code

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

Coming back to this interesting position it appears to me as a perfect (?) beast (thank you lightvector!) because this position adresses a number of difficulties with the wording of the rule I will try to summarize:

Let's take the main variation in order to try and find the status of the 5 white stones:

$$B
$$ -----------------------
$$ | O . O . X . . O O . O
$$ | O O O X O X O O O O O
$$ | X X X O 1 O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .

Click Here To Show Diagram Code

[go]$$B
$$ -----------------------
$$ | O . O . X . . O O . O
$$ | O O O X O X O O O O O
$$ | X X X O 1 O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .[/go]

$$B
$$ -----------------------
$$ | O . O . X . . O O . O
$$ | O O O X a X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .

Click Here To Show Diagram Code

[go]$$B
$$ -----------------------
$$ | O . O . X . . O O . O
$$ | O O O X a X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .[/go]

Seeing this position it is clear that black has taken a ko and white need a pass-for-ko before being allowed to retake the ko at "a". Now seeing some previous comment in this thread, we can also observe that, in order to try and capture the 5 white stone, black begins by taking another white stone for wich the status is unkown though black is not looking at the status of this white stone.

$$B
$$ -----------------------
$$ | O . O . X 2 . O O . O
$$ | O O O X . X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .

Click Here To Show Diagram Code

[go]$$B
$$ -----------------------
$$ | O . O . X 2 . O O . O
$$ | O O O X . X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .[/go]

$$B
$$ -----------------------
$$ | O . O . X O b O O . O
$$ | O O O X a X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .

Click Here To Show Diagram Code

[go]$$B
$$ -----------------------
$$ | O . O . X O b O O . O
$$ | O O O X a X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .[/go]

After the move the position has completly changed:
everybody can see that a ko has been created in "b" but we can also see that "a" is no more a ko because a white move here will take two black stones. Now a question arises : have now white to use a pass-for-ko if this ko reappears in the future?

$$B
$$ -----------------------
$$ | O . O . X O 3 O O . O
$$ | O O O X a X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .

Click Here To Show Diagram Code

[go]$$B
$$ -----------------------
$$ | O . O . X O 3 O O . O
$$ | O O O X a X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .[/go]

$$B
$$ -----------------------
$$ | O . O . X . X O O . O
$$ | O O O X a X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .

Click Here To Show Diagram Code

[go]$$B
$$ -----------------------
$$ | O . O . X . X O O . O
$$ | O O O X a X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .[/go]

After "a" becomes again a ko. Has this ko to be considered the same ko or is it a new ko?
If it is a new ko then white can immediatly take at "a" because no pass-for-ko is needed for a new ko.
If it is the same ko the situation is still not obvious : we are in a double ko situation and we know that, strictly speaking, the rule does not handle correctly a double ko situation. Is it really the intention of the rule to prevent white to play one of the two ko in such situation?

$$B :w4: pass for ko at "a"
$$ -----------------------
$$ | O 7 O 5 X 8 X O O . O
$$ | O O O X 6 X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .

Click Here To Show Diagram Code

[go]$$B :w4: pass for ko at "a"
$$ -----------------------
$$ | O 7 O 5 X 8 X O O . O
$$ | O O O X 6 X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .[/go]

$$B :w4: pass for ko at "a"
$$ -----------------------
$$ | . X . X X 8 . O O . O
$$ | . . . X 6 . O O O O O
$$ | X X X O . O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .

Click Here To Show Diagram Code

[go]$$B :w4: pass for ko at "a"
$$ -----------------------
$$ | . X . X X 8 . O O . O
$$ | . . . X 6 . O O O O O
$$ | X X X O . O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .[/go]

In this final position can we consider the capture of the 5 white stones has enabled to play and ? In any case, I mean even if black is not trying to kill the 5 white stones, white is able to play the uncapturable stones at and

As you can see this alone position adresses number of issues and we can be sure that a common interpretation is quite impossible to reach.
Anyway any comment are welcome.

[CDavis7M]

As you can see this alone position adresses number of issues and we can be sure that a common interpretation is quite impossible to reach.

I think you have presented the issue and questions well. A few comments below.

----------

$$B
$$ -----------------------
$$ | O . O . X O b O O . O
$$ | O O O X a X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .

Click Here To Show Diagram Code

[go]$$B
$$ -----------------------
$$ | O . O . X O b O O . O
$$ | O O O X a X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .[/go]

After the move the position has completly changed:
everybody can see that a ko has been created in "b" but we can also see that "a" is no more a ko because a white move here will take two black stones. Now a question arises : have now white to use a pass-for-ko if this ko reappears in the future?

White would not have to pass. White must pass to retake a ko. But the Japanese Rules define a ko as a position where stones can continue to be captured and recaptured. A position where recapture would cause the opponent's stones to be taken off the board such that your stones become uncapturable does not require a pass. This is shown in Example 8 of the Japanese Rules.

黒4 パス, 白5 九子取り上げる. Black's stones are dead. Even though Black took a ko, does not require a pass before capturing because it is not a ko position since it would take Black's stones off the board. Of course there is an intervening move as is required for kos.


----------

$$B
$$ -----------------------
$$ | O . O . X . X O O . O
$$ | O O O X a X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .

Click Here To Show Diagram Code

[go]$$B
$$ -----------------------
$$ | O . O . X . X O O . O
$$ | O O O X a X O O O O O
$$ | X X X O X O O . O . .
$$ | X . X O O O O O . . .
$$ | . X X . . . . . . . .
$$ | X X . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .[/go]

After "a" becomes again a ko. Has this ko to be considered the same ko or is it a new ko?

It is the same ko because it is the same shape even though it was briefly a shape where White could take Black's stones off the board if White had 2 moves in a row.

we are in a double ko situation and we know that, strictly speaking, the rule does not handle correctly a double ko situation.

[/quote]They don't? I missed that discussion. Would you point me to it?