68.15.102.215: Author's Intent & Errata Page (2010-08-16 09:39) [#7885]

Regarding Problem 124

Bill: The book solution is correct.

Logan?: I'm not seeing how it is correct. Additional global rewards come second to a better local solution in locally constructed problems. The only exception is when the author specifically states otherwise--which is not the case in this book.

Bill: You are accusing the professional author of the book of making a mistake because he does not make the same assumptions that you do. Besides which, go is played on a full board. If White played for seki and took gote to gain a few points locally, that would almost always be the wrong thing to do.

Logan?: I am not "accusing the professional author of the book of making a mistake because he does not make the same assumptions that you do." Your action of addressing me personally with the intent and making assumptions about my beliefs and intentions is inappropriate and does not help to explain the points which I raised in the previous statement. If you would like to talk about the author's intent with this book and use that as a platform to support an argument for why you believe that the answer given in the book is correct, then that is a great place to begin.

The argument which appears to be most prevalent in your reasoning is: All problems could be correct or incorrect in a given global situation. However, this is a platitude which could apply to any problem. The problem with this line of reasoning as a premise is that it means that no correct solution could ever be determined, as any global situation could exist in which a local problem is correct--or incorrect. As the use of this line of reasoning entirely negates a possibility of there being 'one' correct solution, then it must be set aside as a ceteris paribus. Once set aside, only one main factor exists in helping to determine which possible answer is the most correct in a book?--author's intent.

From doing every single problem in this book, not once does the author choose an incomplete variation as correct for it's sente purposes. Nor does the author stop at any point in a problem and declare 'neither players should play here anymore, as the whole board position is likely more important' or to a similar effect. The only instances in which two variations exist and the author chooses one over the other is when one 'settled position' contains more ko threats than other. Therefore, two factors seem consistent in all 800-problems:

1) The author never suggests that players should tenuki in an unsettled life & death position.

2) The author only chooses one solution over another when the alternate solution contains more ko threats or more points.

With these in mind a couple of things seem clear to me (regarding which solutions can be considered most correct relative to the internal rules of the problems in this book, as practiced by the author). First, one cannot declare a life & death position complete until it's status has been fully settled. Second, no alternate solution will have sente/gote as being more important the more ko threats and/or points. I still need to determine which the author holds more important, more points or more ko threats. I believe that it is more points, because something of actual worth versus something of potential worth seems more consistent with the definite attitude of the author.

Bill: Logan, you stated your main assumption quite clearly: "Additional global rewards come second to a better local solution in locally constructed problems." I do not think that I am assuming anything different about your assumptions. :)

Now, if this is a life and death problem, then both diagrams are correct, in the sense that Black lives. They are simply variations. If the author left one of them out, then there could be any number of reasons for that. It is unwarranted to say that the book variation is a mistake. And you will note that I did not claim that the "supposed correction" was a mistake, only that it is not a correction, and that, in a real game, it would almost always be worse for White than the book variation. (If I were presenting this as a life and death problem, I would certainly include both variations. OTOH, if I were forced to leave one out, it would be the one with seki, as it is less realistic.)

Oh, I see I was wrong about the author's being a pro. He is a strong amateur.

Logan?: Your reply about assumptions isn't an address to the concerns that I've stated above. Therefore, I won't be addressing that portion of your reply.

I would like to address a couple of idea, beginning first with what I believe is the lesser critical of the two.

1. I believe that the idea "it is unwarranted to say that the book variation is a mistake" is incorrect. Instead, I believe it is warranted. As outlined in my previous entry, the basis for the belief that it is warranted is that the solution is not consistent with the other solutions provided by the author throughout the text.

2. "Now, if this is a life and death problem, then both diagrams are correct, in the sense that Black lives. They are simply variations. If the author left one of them out, then there could be any number of reasons for that." First, the conclusion that one should not correct (make an errata) on a work for the single reason that there "could" be any number of reasons is unsatisfactory. This line is reasoning is problematic for a few reasons.

a. The conclusion would eliminate any errata ever made by any person in any text throughout history, except possibly for the author. Even if one were to accept this conclusion, then several other problems await and are argued quite thoroughly in the literary, philosophical, cognitive, et. al fields of:

I. Hermeneutics II. Critical theory III. Philosophy of the mind & language, primarily in the discussion of self-knowledge and mental transparency IV. Too many more to list

However, most of these fields do not believe in the inability of others to make persuasive, correct, or insightful comments about the works of others, esp. with regards to determining errata.

b. As mentioned in my previous entry, the conclusion doesn't help to determine whether any problem is incorrect as well as correct! If one cannot determine the incorrectness of a solution by virtue of not knowing the intent of the author, then it must follow that one cannot determine the correctness as well. Because both processes must depend on this reason in order to be validated. However, this type of situation (where a person cannot make head either way) is categorized as a 'ceteris paribus' by intellectuals so that a person can make head.

With these in place, an adequate conclusion can be reached, and a few other things are true. First, this is an errata page. The purpose of such pages are to determine the correctness or incorrectness of items, or to help re-stylize text for specific audiences. Since the book is a problem book that boasts only one correct solution to each problem, then only one correct solution is allowed on the errata page too. I have already outlined how this procedure is and can be done, and have provided several head-starts to support the problem solution that I think is most correct.

I may soon move all of this discussion onto...The "Discussion" section. Then, likely eliminate, or re-locate, the other variations (as they do not fit the purpose of an errata page) until they can be shown to be more consistent than the current errata and with the style of solutions given throughout the rest of the book.

Bill:

Problem 124: White sente  

@ 1.

Problem 124: White sente (cont.)  

The supposed correction is also White sente.

Problem 124: White sente 2  

After Black is alive, so the local temperature has dropped. Black can capture the White stones or White can save them. Plainly this position is worse for White, on average, than the result in the book solution.

Logan?: Black is not yet alive. W @ a would still keep Black's life undecided. Since the position is a local problem, then White must continue the sequence. There are no global assumptions in this book, thus no points given for incomplete answers. Therefore, local positions must be fully played. Also, who is the position worse for? To me, it seems worse for Black.

Bill: Black is alive because White to play cannot prevent Black from living. That is normal usage in go. True, Black might die if White takes the single stone as a ko threat, but go players still say that Black is alive. The position after is worse for White because Black can capture the White stones for a huge corner.