This 'n' that

Bill Spight · **#501**

moha wrote:

Bill Spight wrote:

When I calculated the posterior odds of the existence of a non-black raven, I was quite surprised to find that they were the same, regardless of whether the new observation was of a non-black non-raven, a black non-raven, or a black raven.

OC, that cannot happen if you are asking about the association between ravenhood and blackness. You have to ignore the margin totals to avoid that.

And if you ignore the margin totals, what point is a 2x2 table? You might as well have a 1x2 table, with one cell for non-black ravens and the other cell for everything else.

If the problem is narrowed down as a finite sequence of independent samples of four types A-B-C-D, then it's not surprising that the probability of the existence of a D sample depends only on the number of seen/unseen samples. But I'm not sure if this is the same as the original raven problem.

I found Hempel's original paper online some years ago. As I recall, Hempel did not make a statistical argument, but he did say that seeing a black non-raven was confirmatory.

RobertJasiek · **#502**

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

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

In the Miai Values List, Sensei's Library writes that White takes a stone in sente:

Click Here To Show Diagram Code: [go]$$W Sente sequence? $$ --------------------------------------- $$ | . . . . . X 1 X O . . . . . . . . . . | $$ | . X . X X 2 . O . . . . . . . . . . . | $$ | . . . X O O O O O O . . . . . . . . . | $$ | . . . X X . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . |[/go]

However, why is this sente? The answer depends on whether the following is a traversal sequence with CGT-reversal.

Click Here To Show Diagram Code: [go]$$W Traversal? $$ --------------------------------------- $$ | . . . . . X 1 X O . . . . . . . . . . | $$ | . X . X X 2 3 O . . . . . . . . . . . | $$ | . . . X O O O O O O . . . . . . . . . | $$ | . . . X X . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . |[/go]

To clarify whether this is a traversal sequence, the following difference game can be considered, whose initial prisoner difference is 1 because the colour-inverse copy after move 2 is constructed by creating a white prisoner stone.

Click Here To Show Diagram Code: [go]$$B Difference game, initial prisoner difference 1 $$ ------------------------------------------- $$ | . . . . . X . X O . . . X . X O . . . . . | $$ | . X . X X . . O . . . . . X . O O O . O . | $$ | . . . X O O O O O O . X X X X X X O . . . | $$ | . . . X X . . . . , . . . . . . O O . . . | $$ | . . . . . . . . . . . . . . . . . . . . . |[/go]

To assess White's tentative traversal sequence by playing the difference game, Black starts and White must achieve at most the count 0.

Click Here To Show Diagram Code: [go]$$B Black starts I, initial prisoner difference 1, White's mistake, count 1 $$ ------------------------------------------- $$ | . . C C . X 1 X O . . . X C X O C C C . . | $$ | . X . X X 2 . O . . . . . X 3 O O O . O . | $$ | . . . X O O O O O O . X X X X X X O . . . | $$ | . . . X X . . . . , . . . . . . O O . . . | $$ | . . . . . . . . . . . . . . . . . . . . . |[/go]

Black 1 is the interesting start. White avoids the mistake 2.

Click Here To Show Diagram Code: [go]$$B Black starts II, initial prisoner difference 1, Black's mistake, count 0 $$ ------------------------------------------- $$ | . . C C . X 1 X O . . . X 3 X O C C C . . | $$ | . X . X X 4 . O . . . . . X 2 O O O . O . | $$ | . . . X O O O O O O . X X X X X X O . . . | $$ | . . . X X . . . . , . . . . . . O O . . . | $$ | . . . . . . . . . . . . . . . . . . . . . |[/go]

The mistake Black 3 gives the wrong impression that White can win the difference game.

Click Here To Show Diagram Code: [go]$$B Black starts III, initial prisoner difference 1, ko $$ ------------------------------------------- $$ | . . . . . X 1 X O . . . X 4 X O . . . . . | $$ | . X . X X 3 . O . . . . . X 2 O O O . O . | $$ | . . . X O O O O O O . X X X X X X O . . . | $$ | . . . X X . . . . , . . . . . . O O . . . | $$ | . . . . . . . . . . . . . . . . . . . . . |[/go]

If White 4 passes and lets Black 5 connect the ko, the count is 1 because of the initial prisoner difference and White loses the difference game. Therefore, White must fight the ko to possibly win the difference game. The resulting count depends on the ko and ko threats.

However, the method of playing the difference game is inapplicable if necessarily its winner depends on ko threats.

What does this mean for the tentative traversal sequence? Does an inapplicable difference game mean that there is no traversal or do we need to clarify whether we have a traversal sequence by other methods? Which other methods let us decide this? How?

Maybe White 1 - Black 2 is a sente sequence. If so, just claiming this is insufficient. We need to prove that it is either a sente sequence or a gote traversal sequence. But how?

Clarification enables determination of the correct count and move value of the local endgame in the initial position. By only pretending to have a sente sequence or else to have a traversal sequence, we do not know whether the values are correct and whether a sente sequence or long gote sequence should be played.

If the local endgame is White's local sente, the move value is 5/6. If the local endgame is White's long gote, the move value is 3/4. The counts also differ.

Bill Spight · **#503**

Click Here To Show Diagram Code: [go]$$Bc Locale $$ --------------------------------------- $$ | . . C C C B C B O . . . . . . . . . . | $$ | . X . X X C C O . . . . . . . . . . . | $$ | . . . X O O O O O O . . . . . . . . . | $$ | . . . X X . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . |[/go]

Click Here To Show Diagram Code: [go]$$Wc Sente? $$ --------------------------------------- $$ | . . C C C X 1 B O . . . . . . . . . . | $$ | . X . X X 2 . O . . . . . . . . . . . | $$ | . . . X O O O O O O . . . . . . . . . | $$ | . . . X X . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . |[/go]

After

-

the local count is 1⅔ and the temperature is ⅔.

Click Here To Show Diagram Code: [go]$$Wc Reverse? $$ --------------------------------------- $$ | . . C C C X 1 B O . . . . . . . . . . | $$ | . X . X X 2 3 O . . . . . . . . . . . | $$ | . . . X O O O O O O . . . . . . . . . | $$ | . . . X X . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . |[/go]

If

-

reverses to this position, the local score is 1.

Click Here To Show Diagram Code: [go]$$Bc Black first $$ --------------------------------------- $$ | . . C C C X 1 B O . . . . . . . . . . | $$ | . X . X X . . O . . . . . . . . . . . | $$ | . . . X O O O O O O . . . . . . . . . | $$ | . . . X X . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . |[/go]

After

the local count is 2½ and the temperature is ½.

If the previous diagram shows White correct play, so that the original position is gote, then the count of the original position is 1¾ and its temperature is ¾. Since ¾ > ⅔ in the previous diagrams where White plays first, the temperature has dropped after :b2:

, so White does not continue with :w3:

. There is no reverse.

----

Black komonster (territory scoring)

If Black is komonster under territory scoring, then. . . .

Click Here To Show Diagram Code: [go]$$Wc W3 elsewhere $$ --------------------------------------- $$ | . . C C C X 1 B O . . . . . . . . . . | $$ | . X . X X 2 4 O . . . . . . . . . . . | $$ | . . . X O O O O O O . . . . . . . . . | $$ | . . . X X . . . . , . . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . |[/go]

After

the local count is 2½ and the temperature is ½, which is the same as after Black plays first. In that case we may consider :w1:

-

to reverse. And then the original count is 1¾ with a temperature of ¾, and the position is ambiguous.

----
White komonster (territory scoring)

If White is komonster at territory scoring, then the local count after :w1:

-

is 1½ and the temperature is ½. Since :w1:

is sente, the count of the original position is also 1½. With very rare exceptions somebody is komonster, so we can regard the original count as 1¾ or 1½ depending upon who it is. If we don't know who is komonster, we can regard it as 1⅔.

----

Note: Difference games are not guaranteed to work with kos, as kos destroy the independence of the positions.

RobertJasiek · **#504**

Suppose we have a local endgame with a player's alternating 3-move sequence with the tentative gote traversal move value M (calculated from his traversal follower and the opponent's gote sequence follower) and the follow-up move value F in the position created by move 2. So are you saying that necessarily M > F determines 'no traversal (CGT reversal)', M = F determines 'possible traversal' and M < F determines 'traversal'?

I think difference games work well with kos if the winner of a difference game does not fight the ko but avoids all ko threats elsewhere or locally affecting the kotype by letting the opponent win and dissolve the ko. Pass if neccessary to let the opponent win the ko. It is good enough for the winner to win the difference game nevertheless.

Bill Spight · **#505**

RobertJasiek wrote:

Suppose we have a local endgame with a player's alternating 3-move sequence with the tentative gote traversal move value M (calculated from his traversal follower and the opponent's gote sequence follower) and the follow-up move value F in the position created by move 2. So are you saying that necessarily M > F determines 'no traversal (CGT reversal)', M = F determines 'possible traversal' and M < F determines 'traversal'?

If I understand you correctly, yes. F is the local temperature and M is the original temperature, assuming 'traversal'. If M > F then why should the first player continue to play? That is inconsistent with traversal, right?

Quote:

I think difference games work well with kos if the winner of a difference game does not fight the ko but avoids all ko threats elsewhere or locally affecting the kotype by letting the opponent win and dissolve the ko. Pass if neccessary to let the opponent win the ko. It is good enough for the winner to win the difference game nevertheless.

Difference games work with komonster by area scoring, and maybe with territory scoring.

Click Here To Show Diagram Code: [go]$$B Black komonster, Black has one prisoner $$ ------------------------------------------- $$ | . . . . . X 1 X O . . . X 5 X O . . . . . | $$ | . X . X X 3 4 O . . . . . X 2 O O O . O . | $$ | . . . X O O O O O O . X X X X X X O . . . | $$ | . . . X X . . . . , . . . . . . O O . . . | $$ | . . . . . . . . . . . . . . . . . . . . . |[/go]

Because Black is komonster, White cannot take the ko, but must play :w4:

. Black wins the difference game by either form of scoring.

Click Here To Show Diagram Code: [go]$$B White komonster, Black has one prisoner $$ ------------------------------------------- $$ | . . . . 5 X 1 X O . . . X 3 X O . . . . . | $$ | . X . X X 4 6 O . . . . . X 2 O O O . O . | $$ | . . . X O O O O O O . X X X X X X O . . . | $$ | . . . X X . . . . , . . . . . . O O . . . | $$ | . . . . . . . . . . . . . . . . . . . . . |[/go]

If White is komonster, :b3:

is at least as big as :w4:

, since Black cannot take the ko back. The result is jigo, which is a second player (White) win.

Click Here To Show Diagram Code: [go]$$B White komonster, Black has one prisoner $$ ------------------------------------------- $$ | . . . . . X 1 X O . . . X 4 B O . . . . . | $$ | . X . X X 3 5 O . . . . . X 2 O O O . O . | $$ | . . . X O O O O O O . X X X X X X O . . . | $$ | . . . X X . . . . , . . . . . . O O . . . | $$ | . . . . . . . . . . . . . . . . . . . . . |[/go]

fills. White wins jigo.

Bill Spight · **#506**

On combining the study of tsumego with the study of the endgame

Here, I think, is a good example. It is a modification of a problem from a Chinese book about semeai. (See viewtopic.php?p=231227#p231227 )

Click Here To Show Diagram Code: [go]$$Bc $$ | . . . . . . . . . . $$ | . . O . . . . . . . $$ | . . . O . . . . . . $$ | . . . . X . X . . . $$ | . O O O X . . . . , $$ | . O X X O X X . X . $$ | . . . X O O O X . . $$ | . . O X . . . . . . $$ +--------------------[/go]

The original problem is for Black to play and win the semeai, despite having only two dame to three. I have added a few stones on the outside to make it more well defined as a yose problem. (One defect of many yose problems is poor definition.)

Click Here To Show Diagram Code: [go]$$Bc Black wins the semeai $$ | . . . . . . . . . . $$ | . . O . . . . . . . $$ | . . . O . . . . . . $$ | . . . . X . X . . . $$ | . O O O X . . . . , $$ | C O X X W X X . X . $$ | 4 1 3 X W W W X . . $$ | 5 2 O X . . 7 C C . $$ +--------------------[/go]

@ 2

is fairly obvious, making this in a way an easy problem. However, it is easy for the solver to miss :w2:

, which is White's best reply.

I don't know if this is the main line in the book, but it should be. Note that :b3:

at 7 also wins the semeai, but :b3:

is the better play. That fact should be taken into account, combining tsumego with yose.

In the region indicated by the marked points and stones, the local territory after :b7:

is Black +11.

Since we are now studying this as a yose problem, let's look at the play when White plays first. We'll return to the play when Black plays first later.

Click Here To Show Diagram Code: [go]$$Wc White first $$ | . . . . . . . . . . $$ | . . O . . . . . . . $$ | . . . O . . . . . . $$ | . . . . X . X . . . $$ | . O O O X . . . . , $$ | C O B B O X X . X . $$ | C C 1 B O O O X . . $$ | C C O B . . 2 . C . $$ +-------------------[/go]

is the obvious play. Then after :b2:

the local territory is White +12.

OC, :w1:

is hardly a White sente, so let's look at the position after a White follow-up.

Click Here To Show Diagram Code: [go]$$Wc White first, White follow-up $$ | . . . . . . . . . . $$ | . . O . . . . . . . $$ | . . . O . . . . . . $$ | . . . . X . X . . . $$ | . O O O X . . . . , $$ | C O B B O X X . X . $$ | C C O B O O O X . . $$ | C C O B C C 3 5 6 . $$ +-------------------[/go]

The descent, :w3:

, is best. Then later :w5:

is a 1 pt. sente.

The local territory is White +15.

And that means:

Click Here To Show Diagram Code: [go]$$Wc White first $$ | . . . . . . . . . . $$ | . . O . . . . . . . $$ | . . . O . . . . . . $$ | . . . . X . X . . . $$ | . O O O X . . . . , $$ | C O B B O X X . X . $$ | C C 1 B O O O X . . $$ | C C O B C C C C C . $$ +-------------------[/go]

After

the territorial count in the marked region is White +13½.

This is a gote position. Each play gains 1½ pts.

Now let's go back and look at the play after Black plays first.

Click Here To Show Diagram Code: [go]$$Bc Black first, Black follow-up $$ | . . . . . . . . . . $$ | . . O . . . . . . . $$ | . . . O . . . . . . $$ | . . . . X . X . . . $$ | . O O O X . . . . , $$ | 4 O X X W X X . X . $$ | 3 1 C X W W W X . . $$ | C C W X C C 5 C C . $$ +-------------------[/go]

makes territory in the corner. :b5:

could also make two eyes. The result is a local score of Black +17. :b3:

-

gains 6 pts.

Backing up:

Click Here To Show Diagram Code: [go]$$Bc Black first $$ | . . . . . . . . . . $$ | . . O . . . . . . . $$ | . . . O . . . . . . $$ | . . . . X . X . . . $$ | . O O O X . . . . , $$ | C O X X W X X . X . $$ | C 1 C X W W W X . . $$ | C C W X C C C C C . $$ +-------------------[/go]

After

the territorial count is Black +11. This position is a White sente, following the main line above. The reverse sente gains 6 pts.

(No, I have not shown that it is a White sente. You are welcome to do so.

)

(BTW, you can tell that after :b1:

the local temperature has dropped because :b1:

prevents the atari at C-02 after White B-01.

)

Click Here To Show Diagram Code: [go]$$Wc Original position $$ | . . . . . . . . . . $$ | . . O . . . . . . . $$ | . . . O . . . . . . $$ | . . . . X . X . . . $$ | . O O O X . . . . , $$ | C O B B W X X . X . $$ | C C C B W W W X . . $$ | C C W B C C C C C . $$ +-------------------[/go]

In the marked region the territorial count is White +1¼. This is a gote position. Each play gains 12¼ pts.

Bill Spight · **#507**

Today we are holding a Celebration of the Life of my late wife, Winona Adkins.

At the same cemetery where we got married.

EdLee · **#508**

The wedding wasn't on Halloween, was it ? :shock:

Bill Spight · **#509**

EdLee wrote:

The wedding wasn't on Halloween, was it ? :shock:

No such luck.

It was May 15.

goTony · **#510**

Bill Spight wrote:

DrStraw wrote:

Once you say contrived (oops! AGA) rules I lose interest.

OK. Chinese rules.

Well played indeed!

EdLee · **#511**

Was it an exceptional case or was it a thing to get married at a cemetery...? :study:

Bill Spight · **#512**

Winona wanted to get married outdoors, and I was in charge of finding the place. I assumed that it would not be difficult to find room in a local park. And if we had just done it, that probably would have worked. But we wanted a reception afterwards, and that presented more of a problem.

Finally, with only a few weeks to go, I was getting a bit desperate. Then I found a web site by a couple who had a great deal of information about outdoor weddings in the East Bay Area. They mentioned the possibility of cemeteries. The Mountain View Cemetery turned out to be a very good choice. They host three to four weddings a year. In addition, their event coordinator took very good care of us.

EdLee · **#513**

Hi Bill,

Quote:

They host three to four weddings a year. In addition, their event coordinator took very good care of us.

Good for you. ( And it was a regular thing for them.

)

dfan · **#514**

Bill Spight wrote:

Click Here To Show Diagram Code: [go]$$Bc Black wins the semeai $$ | . . . . . . . . . . $$ | . . O . . . . . . . $$ | . . . O . . . . . . $$ | . . . . X . X . . . $$ | . O O O X . . . . , $$ | C O X X W X X . X . $$ | 4 1 3 X W W W X . . $$ | 5 2 O X . . 7 C C . $$ +--------------------[/go]