Quote:
I am so turned on by this idea, that I feel the need to clarify once again: It is not about guessing the move the professional made, it's about thinking about and improving your own reasoning. One is not trying to guess correctly, but rather to evaluate one's own decisions by comparing them to the decision a professional made in exactly the same situation. Because of this, there is no reason to use a program such as goscorer. These sort of programs are great for guessing how a pro plays - but for the purpose of this exercise, any old sgf viewer will do. The goal is to become aware of the shortcomings of one's own reasoning, so instead of guessing until you find the move the pro decided on, it's better to just choose one move yourself and then compare it to the move the pro chose. I assure you, practically every move can offer you an insight about your game.
Although this is brimming with good sense, I'm not sure that it quite hits the mark. My own observations of how pros learn suggests that there are two separate processes: absorption and adsorption.
First, for the purpose of exposition, assume simply that we have a subconscious part of the brain and a conscious one. Go uses both. The subconscious part is good at absorbing information in large quantities and sorting it, creating useful links and networks while you sleep. To feed this beast, you simply expose it to lots and lots of game records. The ideal state for doing this appears to be NOT to study a single game in the way you suggest, but simply to play it over, fairly briskly but not rushing, in a state of "No Mind" (i.e. not letting the conscious brain have a look-in). The end result is that when it comes to playing a game yourself, this subconscious beast, or intuition, will say things like, "I've seen this before - this is what happens next" or "I've seen something similar - try X, Y and Z." The more games you absorb in this way, the better the advice your intuition will give you." Most pros seem to pick a favourite collection (e.g. games of Yasui Chitoku) and play these over repeatedly, otherwise they just absorb new game after new game. As far as I can make out, though, the repetition of a favourite set is less about creating new networks in the brain and more about strengthening certain links so as to create a stylistic bias, so that when the beast throws up its suggestions, these will be biased towards a style you admire, prefer or aspire to.
The conscious part of the brain seems to be developed in a different way, which I call adsorption. Here, you are not packing your brain with data, but you simply allow data to attach to your pre-processor, where you mull it over, test it, and maybe reject it. The end result here is like a bee turning nectar into honey. This "essence" is then packed into the deeper part of the brain, more as a programming module than as pure data of the absorbed type. Data is swallowed wholesale; essence is digested first.
Again with the caveat that this is only what I have gleaned from talking to pros or reading accounts of their formative years, the adsorption process seems to be divided into three approaches. One (the main one, I think) is to attend a study group with students of similar strength and to toss ideas back and forwards, typically by group-studying a single game. I don't recall a single instance of tossing ideas back and forwards like this with a teacher.
A second approach is to study by oneself but in this case it seems relatively rare to study a whole single game. Instead, the student will study one part of one game (e.g. trying to read out a major fight), or will study a set of games by looking at a related theme (e.g. building moyos) in each. In the latter case, the more advanced students (meaning also established high dans) seem to focus on assessing other players' styles, and so act in a kind of GoScorer mode, but with the important difference that they guess only moves at critical points (i.e. those where a true choice - style - exists). Again the emphasis is on distilling an essence which can later be sent to the deeper brain.
The third approach is to have a teaching game. In other words, you get some external help with your moonshine distilling skills!
It is my strong impression that the most difficult part of this whole process is not the adsorption part, where (for adults, at least) reasoning and experience from other fields can be applied, but the absorption part. To some extent this can be said to be difficult for busy adults because the time required to play over a large number of games can seem immense, but I don't think that's the real problem. The real problem is achieving the state of No Mind, i.e. abandoning attempts to reason or to actively memorise. Indeed, if this state of No Mind can be achieved, the number of games to play over can be made rather small - usually 1,000 for 1-dan is quoted but I've seen 10,000 for 5-dan.
If my analysis is right, daal's approach seems to mix absorption and adsorption (not a good idea, as a working assumption). It also explains why blitz games are not a great idea either for players who are still at the stage where data absorption is needed. - the actual playing of the game uses the front part of the brain, is distracting and prevents No Mind (which is different from the mindlessness usually associated with blitz, of course). It may also explain to some degree why young children have an advantage - perhaps they are better at achieving No Mind.
If you are not familiar with No Mind, it goes by lots of other names and phrases in the Far East, especially on fans signed by go pros e.g. Stone Mind, Bare Mind, No Self, No Gates, Cleansing the Mind, Child-like Mind, Pure and Empty, Innocent and Unworldly, Stiffened Bones and Empty Mind, Interference-free Thoughts, Clearing the Mind, Calming Oneself, Observing Calmly is Beneficial to Oneself.... I am quoting mainly from a pile of fans behind me and could add quite a few more, but I imagine you've already got the message that No Mind is a very important concept to a pro.
There are even Japanese pros who - shock, horror - think that amateurs take no notice of this advice. From memory, I think it was Miyamoto Naoki who came up with a nice joke about this on one fan for amateurs. Instead of writing the usual characters for
Musou to mean No Thoughts, he wrote characters that meant Dream Thoughts, which is what most amateur attempts end up as. In similar vein, the shogi pro Aono Teruichi once told me that pros regard amateur thinking time as dreaming time - meaning (once he'd stopped wetting himself with laughing) that instead of thinking they should spend more time just absorbing data.