Life In 19x19

Click Here To Show Diagram Code

[go]$$Bc Black to Play.
$$ . . . . , . . . |
$$ . . . . . . . . |
$$ . . . . . X . . |
$$ . . . X X . X . |
$$ . . . X O O O . |
$$ . . . . X . . . |
$$ . . . . X O . . |
$$ . . . . X O . . |
$$ . . . . X X O . |
$$ . . . . . . O . |
$$ ----------------+[/go]

I disagree with the solution that is proposed but here it is: However, it is quite possible for black to achieve an unambiguous result: Personally, I consider the seki to be the better result for black. What say you?

Agree that your variation is unambiguous seki, but I think ko is generally a better result than seki (of course, it depends on the board).

In Tsume-Go, the order to evaluate "kill" is

-- kill unconditionally
-- kill conditionally (i.e. by means of Ko)

Achieving a Seki usually belongs to the evaluation of "live".

The whole board does not matter in Tsume-Go, unless it is a whole-board-problem.

If you don't know the whole-board situation, surely you must assume that the loser of the Ko will get some points elsewhere and an unconditional result is preferable?

Cassandra's post already says it all. For further reading, check out http://senseis.xmp.net/?TsumegoConventions

Charlie wrote:If you don't know the whole-board situation, surely you must assume that the loser of the Ko will get some points elsewhere and an unconditional result is preferable?

So .. why can't black "lose" the ko and gain some points elsewhere? A ko threat for 10 points gote should already be a net gain over your sente-seki variation.
And don't forget, if white loses the ko, black will gain about 30 points locally, so white needs much bigger threats than black.

Charlie wrote:If you don't know the whole-board situation, surely you must assume that the loser of the Ko will get some points elsewhere and an unconditional result is preferable?

No, as I said, the rest of the board does not matter with Tsume-Go.

But let's think of the whole board.

If White wins the Ko, she gets about 6 points of territory.
If Black wins the Ko, he gets about 24 points of territory.
If Black gains a Seki, no side has any points of territory.

White needs much larger Ko-threats than Black, so the burden of the Ko lies on White's side.

This implies that we can surely assume that Black will be able to get an advantage of more than 6 points (the differece between losing the Ko and Seki) elsewhere on the board.

Only in the unlikely case that Black has (let's say) no Ko-threats available, but White still has some of a size that Black must answer, Black will choose the Seki-version.

This is a slight variation on position 20 in Cho Chikun's life and death dictionary.
Black makes a mistake in timing in your 'better' result. Correct is to simply cut at 3. Black keeps the hane in reserve until White tries to live from the other side with 4. White can not connect after 7. White's liberty count is the same in the OP so the internal result is as given. The difference in the surrounding position does leave events on the outside unclear in the OP (e.g. if White captures the hane stone 5), but that's the nasty real world for you.

Click Here To Show Diagram Code

[go]$$Bc Cho Chikun version (colors reversed)
$$ . . . . , . . . |
$$ . . . . . . X . |
$$ . . . . . . . . |
$$ . . . . X X X . |
$$ . . . . . O O 5 |
$$ . . . . X 7 2 6 |
$$ . . . . X W . 1 |
$$ . . . . X W 3 . |
$$ . . . . X X O 4 |
$$ . . . . . . O . |
$$ ----------------+[/go]