Life In 19x19
http://www.lifein19x19.com/

Tewari analysis
http://www.lifein19x19.com/viewtopic.php?f=15&t=11863		 Page 1 of 3
Author:  Pippen [ Fri May 22, 2015 6:25 pm ]
Post subject:  Tewari analysis
I wanna know how one would analyze this Fuseki with Tewari. Can one give a step-by-step description? I still dunno how exactly it works.

Click Here To Show Diagram Code
[go]$$B
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . 3 . . 1 . . . 2 6 . . . . |
$$ | . . . O . . . . . , . . . 5 . , O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . 4 . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . X . . . . . , . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ +---------------------------------------+[/go]
Author:  Bill Spight [ Fri May 22, 2015 9:40 pm ]
Post subject:  Re: Tewari analysis
In this case I would just permute the moves:

Click Here To Show Diagram Code
[go]$$B
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . 3 . . 5 . . . 6 2 . . . . |
$$ | . . . O . . . . . , . . . 1 . , O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . 4 . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . X . . . . . , . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ +---------------------------------------+[/go]


Then you can compare it with other sequences, such as

Click Here To Show Diagram Code
[go]$$B
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . 5 . . . . . . . . . . . . . . . |
$$ | . . 6 . . 3 . . 7 . . . . 8 2 . . . . |
$$ | . . . O . . . . . , . . . 1 . , O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . 4 . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . X . . . . . , . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ +---------------------------------------+[/go]


In this case am not sure that that gains much of anything at the amatuer dan level, though.
Author:  John Fairbairn [ Sat May 23, 2015 12:39 am ]
Post subject:  Re: Tewari analysis
Quote:
I wanna know how one would analyze this Fuseki with Tewari. Can one give a step-by-step description? I still dunno how exactly it works.


That's probably because tewari (dissection of moves, aka ishiwari) is not really intended for fuseki. It's mainly a joseki tool.

There are two separate modes of analysis: (1) try changing the move order to see whether you would have ended up making a different choice if you have had that freedom in actual play - if so, that hints at inefficiency in your actual choice; (2) remove an equal number of surplus stones for each side and see whether, after that, one side has any "silly" stones still there.

This is the Japanese version. There are snake-oil vendors of other versions.
Author:  RobertJasiek [ Sat May 23, 2015 5:05 am ]
Post subject:  Re: Tewari analysis
John Fairbairn wrote:
snake-oil vendors of other versions.


Also other forms of tewari are meaningful and useful (although hardly for the OP position) and therefore no "snake-oil".
Author:  Cassandra [ Sat May 23, 2015 6:54 am ]
Post subject:  Re: Tewari analysis
RobertJasiek wrote:
Also other forms of tewari are meaningful and useful (although hardly for the OP position) and therefore no "snake-oil".

Better use another technical term then ?

As John pointed out, 手割り (tewari) is for usage in 定石 (jôseki) only.
Author:  RobertJasiek [ Sat May 23, 2015 10:08 am ]
Post subject:  Re: Tewari analysis
Cassandra wrote:
(tewari) is for usage in 定石 (jôseki) only.


There is no need to restrict it to joseki.
Author:  Cassandra [ Sat May 23, 2015 10:53 am ]
Post subject:  Re: Tewari analysis
RobertJasiek wrote:
Cassandra wrote:
(tewari) is for usage in 定石 (jôseki) only.

There is no need to restrict it to joseki.

This might be true for the underlying technical concept of "reverse engineering".

However, in every Japanese Dictionary of Go terms that I own, the term "手割り" (TEWARI) is strictly restricted to JÔSEKI. Other
used cases are NOT mentioned at all.

I would like to suggest an application of what King Philip of Macedonia once told his heir:

"My son, ask for thyself another Kingdom, for what which I leave is too small for thee."
Author:  RobertJasiek [ Sat May 23, 2015 11:04 am ]
Post subject:  Re: Tewari analysis
If the dictionaries dictate unnecessarily restricted application of go theory under the name of tewari, it is time to abandon the Japanese name, speak of "methods for stone, move and shape analysis" and apply the theory without artificial restriction.
Author:  SoDesuNe [ Sat May 23, 2015 11:57 am ]
Post subject:  Re: Tewari analysis
RobertJasiek wrote:
If the dictionaries dictate unnecessarily restricted application of go theory under the name of tewari, it is time to abandon the Japanese name, speak of "methods for stone, move and shape analysis" and apply the theory without artificial restriction.


That's exactly what has been asked: Don't use the term Tewari for non-Joseki-sequences : D
Author:  topazg [ Sat May 23, 2015 12:03 pm ]
Post subject:  Re: Tewari analysis
RobertJasiek wrote:
... speak of "methods for stone, move and shape analysis" and ....


Catchy ...
Author:  Pippen [ Sat May 23, 2015 2:25 pm ]
Post subject:  Re: Tewari analysis
John Fairbairn wrote:
(1) try changing the move order to see whether you would have ended up making a different choice if you have had that freedom in actual play - if so, that hints at inefficiency in your actual choice;


Let me try a Tewari-Algorithm: out of what you suggested:

1. Original position of stones
2. Re-shuffle move order of 1.
3. If a move in 2. looks misplaced (despite good plays from opponent) then this move in 1. was not a good one.

Example:

1. Original position, interesting stone: 7.

Click Here To Show Diagram Code
[go]$$B
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . 6 . . 7 . . . |
$$ | . . . 4 . . . . . 8 . . . . . 1 . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . 5 . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . 2 . . . . . , . . . . . 3 . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ +---------------------------------------+[/go]


2. We re-shuffle the move order and give 7 a new number:

Click Here To Show Diagram Code
[go]$$B
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . 8 . . 5 . . . |
$$ | . . . 4 . . . . . 6 . . . . . 1 . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . 7 . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . 2 . . . . . , . . . . . 3 . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ +---------------------------------------+[/go]


3. In the re-shuffle 5 is not good because too small, White plays all moves alright (no double bad plays which equalize) therefore the original stone 7 was a mistake.

What do you think? I'd like to have a theoretical algorithm of tewari where every dummy just follow the rules and comes to a conclusion. Because then we could really check if Tewari is for real. Maybe someone can already try to refute my algorithm-attempt with an example?
Author:  Bantari [ Sat May 23, 2015 2:53 pm ]
Post subject:  Re: Tewari analysis
SoDesuNe wrote:
RobertJasiek wrote:
If the dictionaries dictate unnecessarily restricted application of go theory under the name of tewari, it is time to abandon the Japanese name, speak of "methods for stone, move and shape analysis" and apply the theory without artificial restriction.


That's exactly what has been asked: Don't use the term Tewari for non-Joseki-sequences : D

So what is it called? Or do they not analyse anything else than joseki in Japan this way?
I mean - if you perform exactly the same analysis as tewari for non-joeski sequence, with exactly the same methodology, what is it called? Or is such analysis invalid?

It seems to me that if we *can* analyse non-joseki positions like that, it makes more sense to loosen the definition and still call it "tewari" than to come up with yet another foreign-sounding (or convoluted) term. Not to metion that I have seen analyses of non-joseki sequences even in this ng which were still called "tewari" and nobody objected.

Or we can just loosen up the definition of "joseki". ;)
Author:  John Fairbairn [ Sat May 23, 2015 2:58 pm ]
Post subject:  Re: Tewari analysis
Quote:
What do you think? I'd like to have a theoretical algorithm of tewari where every dummy just follow the rules and comes to a conclusion. Because then we could really check if Tewari is for real.


Here's an algorithm you might like to try first:

1. Read what was written.

2. Did it say it referred mainly to joseki?

- If NO, exit

- If YES

-- Is the test position a joseki?

-- If NO, exit

-- If YES, try tewari.

I translated a series by Takagawa on tewari in Go World if you want something in English, but there are many articles in the oriental languages, and also in-line comments in very many commentaries. It's not a magic bullet. You apply it as a verification when you suspect something is wrong, or are just plain unsure. It is probably also a matter of taste of how you apply it. My impression is that Also, of the two main methods that exist, the reshuffle one is perhaps the more popular one but the bark-stripping method is considered more reliable (it is used especially with thickness versus territory josekis)..
Author:  Pippen [ Sat May 23, 2015 3:16 pm ]
Post subject:  Re: Tewari analysis
I still think tewari can be applied for joseki, fuseki and even in middle game. Why not? Again: I invite Go-players to give an example based on my algorithm where my algorithm comes to the cocnlusion the move was bad/good when indeed it was good/bad and hence proves it unreliable.
Author:  SoDesuNe [ Sat May 23, 2015 4:13 pm ]
Post subject:  Re: Tewari analysis
Bantari wrote:
SoDesuNe wrote:
RobertJasiek wrote:
If the dictionaries dictate unnecessarily restricted application of go theory under the name of tewari, it is time to abandon the Japanese name, speak of "methods for stone, move and shape analysis" and apply the theory without artificial restriction.


That's exactly what has been asked: Don't use the term Tewari for non-Joseki-sequences : D

So what is it called? Or do they not analyse anything else than joseki in Japan this way?
I mean - if you perform exactly the same analysis as tewari for non-joeski sequence, with exactly the same methodology, what is it called? Or is such analysis invalid?

It seems to me that if we *can* analyse non-joseki positions like that, it makes more sense to loosen the definition and still call it "tewari" than to come up with yet another foreign-sounding (or convoluted) term. Not to metion that I have seen analyses of non-joseki sequences even in this ng which were still called "tewari" and nobody objected.

Or we can just loosen up the definition of "joseki". ;)


I'm not really the person to ask since I neither speak (or read sufficiently) any oriental language nor did I really use Tewari, but I guess my answer would be to look up oriental (in this example most likely japanese) literature about it - the source so to speak.

By the way, I assume it is highly possible that Tewari was/is used out of the original japanese context and nobody objects. But this bears the question: Who can actually object?
It's a japanese concept. Unless you read it up at the source, you can just take the name and do whatever with it. If the result is any good, when some non-oriental amateurs do so, well...
Author:  Kirby [ Sat May 23, 2015 5:58 pm ]
Post subject:  Re: Tewari analysis
Pippen, I think tewari is kind of like a heuristic anyway, so coming up with a counterexample doesn't prove it invalid.

It's just a tool that can be useful for analysis.
Author:  illluck [ Sat May 23, 2015 6:08 pm ]
Post subject:  Re: Tewari analysis
Pippen, I honestly don't know what your beef with tewari is (from this and many, many other posts) XD

Tewari is useful in certain situations (when you are able to give reorder sequences reasonably for at least one side), not so useful in others, and can be very easy to misuse if you start giving both sides inferior moves.

I don't know what it would take to persuade you that tewari could be useful, and I guess to be honest it's not that important. Players before Dosaku reached very high levels without really using tewari.
Author:  Cassandra [ Sun May 24, 2015 12:53 am ]
Post subject:  Re: Tewari analysis
Pippen wrote:
Let me try a Tewari-Algorithm: ...:

1. Original position of stones
2. Re-shuffle move order of 1.
3. If a move in 2. looks misplaced (despite good plays from opponent) then this move in 1. was not a good one.

Dear Pippen,

This is NOT Tewari !!!

I assume that you refer to specification (1) of John's "definition".

This kind of Tewari is NOT simply changing the order of moves only, as you apparently understood.

It is always necessary that you have another position as reference from which you already know whether it is good (= already Jôseki in the original context) or bad (= better for one side). This reference position includes several of the stones placed as in your "new" sequence, but not all, as a matter of course.

After you have identified the "sur-plus" stones of your "new" sequence, you will play these (in an order as "realistic" as possible) to finally create the final position of your "new" sequence. For each exchange of moves you will eveluate whether it was better for one side, or "neutral". Combining the single results will give you your evaluation overall.

+ + + + + + + + + + + + + +

Therefore, you have to ask first, whether the sequence you want to evaluate, is really finished.

If the answer is "Yes", you have to ask yourself second, for your "reference sequence", whether the postion of Black 1, 3, and 5, vs. White 2, 4, and 6, is "Jôseki" (in the sense that neither side has the edge). As a matter of course, you could use the position until White 4 as a master, too, is your answer was "No".

Third step (provided that your answer was "Yes") is to evaluate the pros and cons of Black 7. I.e. asking whether you would like playing this stone elsewhere on the board.

Fourth step is to evaluate the pros and cons of White 8 in the same manner.

Last step is judging the combination of your single results.
Author:  Bill Spight [ Sun May 24, 2015 1:22 am ]
Post subject:  Re: Tewari analysis
Pippen wrote:
Let me try a Tewari-Algorithm: out of what you suggested:

1. Original position of stones
2. Re-shuffle move order of 1.
3. If a move in 2. looks misplaced (despite good plays from opponent) then this move in 1. was not a good one.

Example:

1. Original position, interesting stone: 7.

Click Here To Show Diagram Code
[go]$$B
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . 6 . . 7 . . . |
$$ | . . . 4 . . . . . 8 . . . . . 1 . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . 5 . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . 2 . . . . . , . . . . . 3 . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ +---------------------------------------+[/go]


2. We re-shuffle the move order and give 7 a new number:

Click Here To Show Diagram Code
[go]$$B
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . 8 . . 5 . . . |
$$ | . . . 4 . . . . . 6 . . . . . 1 . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . 7 . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . 2 . . . . . , . . . . . 3 . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ +---------------------------------------+[/go]


3. In the re-shuffle 5 is not good because too small, White plays all moves alright (no double bad plays which equalize) therefore the original stone 7 was a mistake.



In the reshuffle :w8: is bad.
Author:  John Fairbairn [ Sun May 24, 2015 4:59 am ]
Post subject:  Re: Tewari analysis
Quote:
In the reshuffle :w8: is bad.


But the even worse Black 5 suggests this is not about go but gomokunarabe. In fact the thread has just been a trolling exercise. We have been mercifully free of these recently and have perhaps let our guard down.
Page 1 of 3 All times are UTC - 8 hours [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/