Life In 19x19

Bantari, there is not just one "my method", but I and others use various methods of different degrees of methodology or formality from "none" via, e.g., "single sorting", "double sorting" (e.g. [7]), "classification" (e.g., [8]), "semi-formal" (e.g., Ko definition paper), "formal" to "mathematical proofs" (e.g., Cycle Law). There are aspects (such as principles) not mentioned in this sample list of degrees.

RobertJasiek wrote:Bantari, there is not just one "my method", but I and others use various methods of different degrees of methodology or formality from "none" via, e.g., "single sorting", "double sorting" (e.g. [7]), "classification" (e.g., [8]), "semi-formal" (e.g., Ko definition paper), "formal" to "mathematical proofs" (e.g., Cycle Law). There are aspects (such as principles) not mentioned in this sample list of degrees.

This is neither here nor there, not sure why you bring this up.

The fact remain that we have two principal ways of to accumulate knowledge (teach/learn/'research'/whatever):
- more formal definition/theorem and escalation leading to (conscious and categorized) understanding with eventual possibility of application
- examples + problems + experiences leading to application and then (possibly subconscious) understanding

I understand that you have argued against the latter as being insufficient in the past, proposing more structured formalized approach - which i describe as the former. Because of this, I do make a distinction between where you are and where the traditional methods are. Regardless of the fact that both methods can be further divided into trends, degrees, methodologies, or levels of formality.

If you want to make such a two-types-only distinction, let it be

1) combination of explicit, generalising knowledge and implicit, exemplifying knowledge

versus

2) implicit, exemplifying knowledge only.

****************************************************************

You keep mentioning "the traditional methods" in plural. Which are they besides "teaching by examples"? How do you divide the method of teaching by examples into submethods of different degree of formality, without combining it with methods of explicit, generalising knowledge? If you want to combine it, then also you argue that the latter is insufficient. E.g., sorting examples already combines it with the method of expressing explicit, general knowledge in the form of the used sorting and its grouping.

I just want to add that the advance of computer go (as well as computer chess before) can mostly be attributed to new algorithmic approaches to the problem, that is computer science. Formal go research had nothing to do with it.

Actually this can even serve as a counterexample to Robert since the old traditional (weak) programs followed coded rules of play derived from human reasoning about go (aka principles). Modern programs using Monte-Carlo-evaluation find strong moves by running lots of simulations, collecting statistics and selecting the best move based on that (examples).

But ultimately I believe the whole discussion is moot. As Robert states, his books have as many diagrams as other ones. Principles have to be taught by example, there is no way around that. The problem is to select (or create) such examples that important principles can be derived easily which is the task of the teacher/author.

And in the end there is an exception to every rule and principle, every game is completely different (that's why we love it!), even one stone somewhere on the board can change the local situation completely. That is why the only real source of go strength is reading power, which is true and known both in east and west, and - alas! - has to be trained.

There is no quick and easy way to strength.

golem7 wrote:I just want to add that the advance of computer go (as well as computer chess before) can mostly be attributed to new algorithmic approaches to the problem, that is computer science. Formal go research had nothing to do with it.

Go programs succeed because of a combination of a) new applied computer science applicable to different things and games, b) go research (such as formulating the goal to distinguish winning from losing final scoring positions, c) a few go-specific knowledge tricks (such as using a few selected patterns) and d) brute force applied to (a) to (c).

Actually this can even serve as a counterexample to Robert since the old traditional (weak) programs followed coded rules of play derived from human reasoning about go (aka principles).

No. It does not serve as a counter-example. Instead, the situation is a bit complicated.

Current go programs also use go knowledge, and research was needed to find which is useful for MC programs. Surprisingly, very little go knowledge is needed for them. Research was needed to come to this conclusion, although little "formal" research. Some call it "engeneering".

Earlier weak programs tried various methods, and often combinations of them. E.g., combinations of heuristics and pattern (example) databases. Therefore, one cannot easily conclude that, in the past, usage of principles was improper; it could as well have been usage of example databases that hurt the usage of principles.

Expert system go programs of the past used, IMO, hopelessly weak or insufficient heuristics, and without a clear system and good hierarchy. Therefore, the past experience with weak expert systems is not a counter-example for designing expert system go programs at all. Rather it proves that badly chosen / codified expert knowlegde leads to relatively weak results. In the future, expert system can become much stronger than today, but currently programmers have no incentive to follow this path, ALA monte-carlo is a low effort path with (judging winning rates only) sufficiently good results.

Expert system go programs are no counter-example to human go knowledge use, because humans do not work like (Turing) computers.

Modern programs using Monte-Carlo-evaluation find strong moves by running lots of simulations, collecting statistics and selecting the best move based on that (examples).

No. The MC programs use a combination of (a) to (d). They do not use only (d).

Principles have to be taught by example, there is no way around that.

Completely wrong.

1) Principles can be taught explicitly, e.g., by text, and then illustrated by examples.

2) Alternatively (NOT: exclusively!), examples can be shown, and their consumer is expected to invent and derive suitably fitting principles on his own. This does not prevent learners from them stating the derived principles explicitly, so that then other learners have the choice for learning principles by (1) or (2).

The problem is to select (or create) such examples that important principles can be derived easily which is the task of the teacher/author.

This is a problem only for (2). For (1), the learner can profit also from, but need not, particularly well chosen examples.

And in the end there is an exception to every rule and principle, every game is completely different (that's why we love it!), even one stone somewhere on the board can change the local situation completely.

In particular, there are also principles without exception, regardless of every game being different and one stone affecting the local situation completely.

That is why the only real source of go strength is reading power,

This is completely wrong, because it is saying that brute force reading almost alone could generate much go strength. This is completely wrong, because the game tree explodes very quickly.

It is always a combination of - conscious or subconscious - go theory knowledge for how to prune reading well and correctly and the reading power itself.

has to be trained.

Reading power has to be trained so that the combination of go theory knowledge and reading power generates go strength.

There is no quick and easy way to strength.

In this generality, it is wrong. E.g., there are very quickly improving kyu players.

It is also very misleading, because there are ways to ease improvement and there are ways to greatly accelerate improvement. Except that such ways do not succeed for all players; so you can say "There is no quick and easy way to strength working (equally) well for all players.".

RobertJasiek wrote:

golem7 wrote:

There is no quick and easy way to strength.

In this generality, it is wrong. E.g., there are very quickly improving kyu players.

It is also very misleading, because there are ways to ease improvement and there are ways to greatly accelerate improvement. Except that such ways do not succeed for all players; so you can say "There is no quick and easy way to strength working (equally) well for all players.".

Quickly improving kyu players =/= quick and easy way to strength. There is never a quick way to strength, even if some people learn faster than others. Even if there was someone was crazy enough to learn all of your formal definitions of nakade and "almost fill" and what have you - to actually become stronger at go - they would need practice, practice, tsumego and repetition. Theory doesn't make strong players. At most (if even then), it might help refine some tiny area of the game for someone who's getting stronger by "traditional means".

RobertJasiek wrote:If you want to make such a two-types-only distinction, let it be

1) combination of explicit, generalising knowledge and implicit, exemplifying knowledge

versus

2) implicit, exemplifying knowledge only.

Why do you assume that #2 does not include "generalizing knowledge" as well? To me, it simply must, or we would never have had any strong players. Pros did not get that strong by simply regurgitating memorized examples - they have to understand the position on a very high level - so highly generalized knowledge must be part of the equation. The difference might be in the way by which this knowledge is attained and possibly by the means by which it is stored.

RobertJasiek wrote:****************************************************************

You keep mentioning "the traditional methods" in plural. Which are they besides "teaching by examples"? How do you divide the method of teaching by examples into submethods of different degree of formality, without combining it with methods of explicit, generalising knowledge? If you want to combine it, then also you argue that the latter is insufficient. E.g., sorting examples already combines it with the method of expressing explicit, general knowledge in the form of the used sorting and its grouping.

This is all just semantics.
Whatever word or words I chose, right or wrong, the underlying idea is that there are two ways of teaching/learning which I am trying to differentiate. Do you really want to side-line this conversation into the issue of me having used a singular or plural? Please....

The bottom of for what I was trying to say before you sidelined it into a discussion about semantics is that I disagree with you that there are more and more books out there which use non-traditional method(s). It might be some are more methodical than others, or that they give some more explanation that others, but they still all basically teach by example/problem not by definition/theorem approach.

If you disagree, we can discuss that. A discussion about me using plural or singular, or about using your words of differentiating between methods or mine - this is really waste of my time and not what I signed up for when I entered this conversation.

walpurgis wrote:There is never a quick way to strength

What then do you call, e.g., my increment from 9k to 1k in about half a year? It was so quick, because I studied much go theory (and only a few problems) during that period. The major obstacle during that period was a necessity to learn only from examples about such topics, for which then go theory was hardly available at all.

all of your formal definitions of nakade and "almost fill"

1) To learn from go theory, it is not necessary to learn from only one author.

2) To learn from go theory, one should not learn only from formal definitions. Concepts, principles, methods, values often are more important.

3) It is a bad joke to sugggest that nakade and almost-fill suffice to become stronger a lot. It is an even worse joke to paint a picture as if learning from one specific person would equate learning only these two terms. Above 30 kyu, there is a lot of go theory that is much more important.

and what have you - to actually become stronger at go - they would need practice, practice, tsumego and repetition.

Practice alone can easily be insufficient. E.g., I was DDK too long because of practicing only. I left DDK and then quickly improved when I started to study go theory, while continuing practice (playing games).

From 14.5 kyu to 1 kyu, I did very little problem solving and my reading skill remained almost constant (I was lucky that it was pretty good already as a 14.5 kyu).

Tsumego was 1/3 responsible for improving from 1 kyu to 3d and 1/4 for 3d to 5d.

Repetition was not useless, but only a minor factor. It is suitable for updating rare taisha josekis one has not needed to apply for 10 years:)

All this experience points to a COMBINATION of go theory, practice etc. being the most useful. Especially go theory.

How fast and to which level have you improved with your own "practice, practice, tsumego and repetition" advice?

Theory doesn't make strong players.

"Strong" is relative. Amateur 1d: theory does it. Amateur high dan: autodidactic theory does it and has to do it, because there is still not enough research on go theory. Professional: similar, but it is even harder to access the necessary "remaining" theory.

Note that theory ALONE does not make strong players. It needs a COMBINATION of ... (I have said that before). Quite like practice / examples ALONE do not make strong players. It also needs the aforementioned combination.

At most (if even then), it might help refine some tiny area of the game for someone who's getting stronger by "traditional means".

My example of improving quickly from 9k to 3d mostly by studying go theory proves that your "at most" statement is wrong.

Earlier research shows that it is much more than "some tiny area of the game".

Bantari wrote:Why do you assume that #2 does not include "generalizing knowledge" as well? To me, it simply must,

Thank you for the clarification of your view. (But now, can please explain again what you call "my method"?)

It might be some are more methodical than others, or that they give some more explanation that others,

Texts can be more methodical, have more explanation, or profit from both aspects more than other texts do.

but they still all basically teach by example/problem not by definition/theorem approach.

Eh. Some of those you call "methodical" basically teach by example. Some of those I call "methodical" basically teach by principle with greater emphasis than by examples. (If you want this to become a books-specific thread, we need to move to the Books forum.)

Sigh...

RobertJasiek wrote:

Bantari wrote:Why do you assume that #2 does not include "generalizing knowledge" as well? To me, it simply must,

Thank you for the clarification of your view. (But now, can please explain again what you call "my method"?)

By 'your method' I understand a hard-research-based approach, in which the general principles are derived from each other, presented in explicit and methodical fashion, and used to convey content and facilitate understanding - while examples are used more to illustrate and emphasize already attained knowledge. The whole process is mainly on the conscious level, and any knowledge and understanding is attained on that level as well. I call it definition/theorem approach - and even if you disagree, from what I see this is pretty much what you do... or at least - what you present to me/us in general.

To clarify - by 'you' I do not necessarily mean you personally, Robert Jasiek, but a group of people following this path. I single you out because I am talking to you, not to them.

By contrast, what I call 'traditional method' emphasizes studying examples, solving problems, and experience rather than any formal definition/theorem-based approach. The underlying idea is that by seeing enough examples and solving enough problems, the student will be able to derive the general principles by himself, consciously or subconsciously, without everything being explicitly spelled out.

What's more, it is my opinion that, while your method *might* (or might not) have advantages at certain level or for certain very limited goals - if you set your sights at truly mastering the game, traditional method is much superior. Your method might not even work at all, other than in purely supplementary fashion. I do not really have any proof of that, its just my opinion, and I am open-minded enough to admit that I am not sure that I am right. Thus - we have a discussion. If you wish we can go into what I base my opinion on, although I have already mentioned most of it somewhere in this thread.

What then do you call, e.g., my increment from 9k to 1k in about half a year? It was so quick, because I studied much go theory (and only a few problems) during that period. The major obstacle during that period was a necessity to learn only from examples about such topics, for which then go theory was hardly available at all.

I went from learning to European 2-dan in six months. It was so quick because I studied traditional Japanese books. The major obstacle from that period was having no teachers except peers of mathematical bent whose approach bored me. I then effectively gave up studying and then playing, and now that there is so much talk of rules and go theory, sometimes feel like giving up even observing the game.

I think one problem in the debate between RJ and Bantari (et al.) is that "research" is being used to refer to much of what we usually call just study. Making lists and cataloguing definitions is not really research. True research can, but does not need to, include such clerical work. Doing it well is of course meritorious and often useful, but is still not the main criterion for calling it research. True research must also include testing and peer review, which I think RJ certainly strives for, but trying out someone else's study method does not qualify for research air miles. I think true research should also include consideration of all the prior art, something too often lacking here, as I no doubt tediously keep pointing out. I don't think application is a good criterion, either - there's nothing wrong with research for the sake of it, but even if an application results, that does not make what went before true research. Giving the work a lofty aim (e.g. solving go) does not cut the mustard either.

What I feel distinguishes the best research is that it includes a strong spark of creativity, or novelty. Do we have that here?

If we equate what is being done in the name of go theory here to a commercial venture, would the results translate into the granting of letters patent or a utility model? If so, we can deem that as proper research. If it simply translates into having a market edge over your commercial rivals, that would seem to me more like studying in go - we work because we just want to beat the guy who at present keeps beating us. Great fun and worthy - but "research"?

Give Robert a break. He likes making lists and categorizing things.

Is Robert's research actually research? Absolutely, in the same way that taxonomy is research. Just as we need taxonomy to make sure that the little labels in front of museum exhibits are correct (got to make sure the furry specimen has a name), so too do we need Robert's work to ensure that we can properly label a nakade shape.

On the other hand, just as knowledge of taxonomy is in no way necessary for advancement in molecular biology or drug development, a formal definition of a nakade is clearly not necessary for go playing strength.

RobertJasiek wrote:Eh. Some of those you call "methodical" basically teach by example. Some of those I call "methodical" basically teach by principle with greater emphasis than by examples. (If you want this to become a books-specific thread, we need to move to the Books forum.)

There is no 'some of those I mentioned' since I mentioned only one author by name, I think - Davies.

And I have seen no books yet which would teach in the way you seem to be presenting your knowledge and findings, even if some of them do stress general principles more than others. To me, they are all example-centric rather than definition/theorem-centric, which makes them traditional in my opinion.

A discussion of a specific book does not have to be moved to the books forum - since it will happen in the context and support of this very general discussion. Which, it seems to me, gets more and more diluted into sidelines and squabbles about words.

RobertJasiek wrote:What then do you call, e.g., my increment from 9k to 1k in about half a year? It was so quick, because I studied much go theory (and only a few problems) during that period. The major obstacle during that period was a necessity to learn only from examples about such topics, for which then go theory was hardly available at all.

1k is not strong. And you didn't get there with just theory, as you said. Besides, anecdotes don't count for much.

1) To learn from go theory, it is not necessary to learn from only one author.

2) To learn from go theory, one should not learn only from formal definitions. Concepts, principles, methods, values often are more important.

3) It is a bad joke to sugggest that nakade and almost-fill suffice to become stronger a lot. It is an even worse joke to paint a picture as if learning from one specific person would equate learning only these two terms. Above 30 kyu, there is a lot of go theory that is much more important.

1) & 2) Not relevant here. 3) If by more important go theory you mean playing games and getting familiar with common shapes and tsumego, I agree.

... words ...

All this experience points to a COMBINATION of go theory, practice etc. being the most useful. Especially go theory.

Correct. Except for the especially theory part. You just might be the unique gem who learns that way (or claims to), but you have been blinded by all your years of theorycrafting. Your general player has little use for the kind of theory you vouch for. And judging by how practically all the top Europeans as well as top pros have learned, I even seem to be right.

Why are you still "only" 4-5 dan and not competing for top 10 in EGC if go theory is so useful?

How fast and to which level have you improved with your own "practice, practice, tsumego and repetition" advice?

More anecdotes? Fine. I got to ~7k quickly and since then I have spent next to no time studying go. Never did I need the knowledge of formal nakade definition, perfect understanding of superko rules, or what "cannot necessarily permanently partition" means. (! Your theory just might kill people by overloading their brains)

"Strong" is relative. Amateur 1d: theory does it. Amateur high dan: autodidactic theory does it and has to do it, because there is still not enough research on go theory. Professional: similar, but it is even harder to access the necessary "remaining" theory.

Note that theory ALONE does not make strong players. It needs a COMBINATION of ... (I have said that before). Quite like practice / examples ALONE do not make strong players. It also needs the aforementioned combination.

The claim you make in your first paragraph is absolutely ridiculous. Say I'm an amateur 1 dan, I spend time studying go, but progress is slow as usual. Then BAM, I only study formal go theory and behold! I become strong by knowing formal principles and definitions of go terms, rules, theory and not playing.

.. So, in 2nd paragraph you agree that theory doesn't really make people stronger, aye? If you mean (in the combination part) that one needs to know how to count right in mid/endgame (ko's) then that I agree with. But theory the way you're advocating on these forums? You'll be hard pressed to find even one person who has directly become stronger by theory research (in this case, excluding you).

My example of improving quickly from 9k to 3d mostly by studying go theory proves that your "at most" statement is wrong.
Earlier research shows that it is much more than "some tiny area of the game".

You make bold claims, but don't show any proof to back it up.

RobertJasiek wrote:Current go programs also use go knowledge, and research was needed to find which is useful for MC programs. Surprisingly, very little go knowledge is needed for them. Research was needed to come to this conclusion

Remembers me to the sentence of star treck: "We needed hundreds of years to learn that it doesn't need hundreds of years to learn"