snorri wrote:I haven't read "First Fundamentals" but the Davies' Tesuji treatment is mostly just depth-first search of a finite graph. That's great when the problem is closed and simple enough for one's skill level, and no doubt it's a skill worth working on, but it doesn't characterize most game positions which are more open and evaluating common lines of play starts to be become useful.
I'd be interested in learning more about how reading relates to risk management, given the reality that our reading skills and available time in a game are both limited. For example, I have seen commentaries to the effect of "black might consider continuing here, but the position is complex and there is no way to know whether black can get out in sente, so black played over here instead."
Such a comment is really just a frank admission that the author's (and maybe also the players') reading is limited, but there is still a need to manage the position somehow.
Another example: when reading out a sequence in the mind's eye, there could be a path where at some point the opponent can cut---the shape makes it looks a bit unreasonable to do that, but due to limited skill and/or time it is hard to see what happens after that cut. Many players would just play it out and hope for the best.
For beginners, almost every position is like this and they have no choice. But once some reading skill is developed maybe some better options become available. Maybe there's a way to fuzzily consider: what's the worst that could happen if my opponent saves that cutting stone? and compare it to: what's the worst that could happen if reach that position and suddenly realize I have to defend that cut in gote? Or maybe: which extra moves could appear later that could make this sequence obviously good for me?.
In my experience, it's impossible to read any position to much depth if you don't engage in aggressive pruning of variations. The only way to prune effectively, however, seems to be to train your subconscious to make particular moves stand out to you for consideration. When you look at the board in some position, say an early one so that there are less stones to take in, with experience your eye tends to be drawn to certain areas or points on the board. I've heard people describe them as shining, but to me they just stand out more than the others. Often, and more often as I get better, they are good candidate moves for consideration, but I didn't perform any mental calculation to make them appear. Once they're there, I can read through what I might expect my opponent to do, but again, their options get limited by what stands out in my mind as reasonable. It's also frequent that common situations get broken into chunks rather than individual moves, which has benefits and drawbacks. It helps me read farther, but I need to be careful to consider other alternatives in the process as well.
I suppose what I'm saying is that the bulk of reading is done by my subconscious, followed up with conscious verification of limited sets of moves. I don't have the time to read large open-ended areas out exhaustively, and I doubt too many other people do either. There are documented examples of professional games where both players played out a certain area, where there was a variation that completely destroys one side or the other. Both players missed the variation, because it involves a bad-shape move that happens to work in that particular situation, and generally the stronger the player, the harder it is for them to consider automatically.
The questions you posed, and particularly the last one, are all good questions to be considering as you play as well. In Zone Press Park, one of the things that O Meien speaks of are finite moves and infinite moves. Finite ones he defines as moves where the variations can be calculated by the players to a certain conclusion. Infinite ones are those with too many variations or otherwise too much complexity for either player to be able to discern what will happen. He stresses that (my interpretation) it's important to consider the surrounding positions to get an idea of whether the result of an infinite move will lead to an overall favourable result for one side or the other, or if the position should favour a balanced result. As for your last question, this is a good thing to consider about any position, as it helps you generate a plan for the rest of the game that might get you stones in the right places for these sequences to start working. You can't have a plan that doesn't involve aims, after all.
