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

Robert Jasiek's Go Theory Research
http://www.lifein19x19.com/viewtopic.php?f=48&t=7791		 Page 4 of 9
Author:  Bonobo [ Mon Oct 14, 2013 2:47 pm ]
Post subject:  Re: Robert Jasiek's Go Theory Research
Bantari wrote:
[..] a hint: if people are telling you something en masse, the there usually is something to it, and might be worth your while to take it seriously instead of insisting that you are right and the world is wrong. Even when you *are* right. Think like a teacher - a communicator - it is up to *you* to bring your message across so that people listen, not to the world to adjust. At least - not immediately. Instead, what you do is often present your message in such way that people's instinct is to argue rather than to think and possibly agree. And this goes especially when you yourself start arguing.

I agree, Bantari.

But slippery territory here, and dangerous, millions of people have been burnt and still are b/c they dissent(ed) with the majority (or the majority with them). Also, many good inventions wouldn’t have been made if the inventors hadn’t keept on keepin’ on, no matter how their peers scoffed.

What I can understand is that some descriptions only work within a “closed” system which depends on certain axioms. To understand such a description, one must accept these axioms.

So perhaps, Robert, would it possible to offer something like a “beginner’s course” on your axioms? Slowly, one by one? And only move on to the next when the first has been digested? You may complain about too little time since you must do research etc., etc., but I’m sure once ppl find taste in this, i.e. they devour/understand the first few axioms and definitions, the rest can be swallowed easier.

And let the pudding be that people buy your books. I for one believe the two I’m sporadically reading in are worth reading (but then again, “who am I?”).


Greetings, Tom
Author:  Bantari [ Mon Oct 14, 2013 3:02 pm ]
Post subject:  Re: Robert Jasiek's Go Theory Research
Bonobo wrote:
Bantari wrote:
[..] a hint: if people are telling you something en masse, the there usually is something to it, and might be worth your while to take it seriously instead of insisting that you are right and the world is wrong. Even when you *are* right. Think like a teacher - a communicator - it is up to *you* to bring your message across so that people listen, not to the world to adjust. At least - not immediately. Instead, what you do is often present your message in such way that people's instinct is to argue rather than to think and possibly agree. And this goes especially when you yourself start arguing.

I agree, Bantari.

But slippery territory here, and dangerous, millions of people have been burnt and still are b/c they dissent(ed) with the majority (or the majority with them). Also, many good inventions wouldn’t have been made if the inventors hadn’t keept on keepin’ on, no matter how their peers scoffed.


You are, of course, right.

But I am not talking here about either blindly following the majority, blindly rejecting it, or not standing up for what you believe in. I am not asking Robert to change his views, invalidate his research, or abandon his methodology. What I am asking for is a little more of an open mind and better approach to communicating and realizing there are other viewpoints - every bit as 'valid' as his, even if contrary. I don't think this ever hurt anybody.
Author:  cyclops [ Mon Oct 14, 2013 4:42 pm ]
Post subject:  Re: Robert Jasiek's Go Theory Research
I think Bantari is helped a great deal if we install a robot in L19x19 that inserts in all Robert's posts the sentence: "My view on <subject title> is as follows. I allow anyone else to have a different view". My proposal then would be to have the robot insert this sentence in the post of all users.
That is my view on the subject and I don't allow ...
Author:  RobertJasiek [ Mon Oct 14, 2013 11:20 pm ]
Post subject:  Re: Robert Jasiek's Go Theory Research
Bantari, professional players are praised for their attention to detail. You seem to criticise me for my attention to detail (when speaking of me causing problems by being more rigid than most others (most amateur players?)), why? Because I dare to do what professional players are praised for?

Bonobo, "beginner courses" on my "axioms" are available at various places. Look around and read. If you mean a series of threads teaching the basics of go knowledge taught be me, I lack time for that, which I need for teaching work.
Author:  Cassandra [ Tue Oct 15, 2013 1:17 am ]
Post subject:  Re: Robert Jasiek's Go Theory Research
RobertJasiek wrote:
Bantari, professional players are praised for their attention to detail. You seem to criticise me for my attention to detail (when speaking of me causing problems by being more rigid than most others (most amateur players?)), why? Because I dare to do what professional players are praised for?

Robert, this statement is typical for your handling of the discussion's flow.
It seems to me to be a desperate attempt to bring the discussion back into "your world".

As far as I can see, Bantari has not picked "detail" out as a central theme of his postings. I doubt that he even used this word. In the contrary, Bantari has written that your methodical approach (wherein "detail" is one of the leading factors) is OK for him -- but only with a view severly limited to "your world".

The professionals' attention to detail is nothing that happens inside "your world". Therefore, it is not comparable at all to what you might understand by "your attention to detail". And -- most important -- "rigidity" has nothing to do with "detail", as you claimed above.

It is your inability to accept another point of view than yours from "inside-your-world" that Bantari misses. And Bantari is not alone with his assessment.

In principle, Bantari has written several texts in "prose" in conformity to what I had written before in a more "mathematical" style. None of both display formats really had any effect on you.

+ + + + + + + + + +

Let me try to give you an example from another "world", the three amateurs' research on Igo Hatsuyoron 120.

Our insights into the problem have grown over time. The same is true for the variations' tree, which has become very large, and very deep. Our books draft has grown from (approximately) 100 pages to 700 pages now.

From time to time it happens that we realize a "mistake" in our sequences, mostly on accident, not on purpose, in one of the several sub-sub-variations. These "mistakes" are not "really disturbing" in the sense that they would put our solution to the problem in question, but incorrectnesses in the correct order of moves "only".

This means that a distinct move should have been played at another moment, to give (with a now "correct" sequence) the same result as we got before (with our previous, now known "incorrect", sequence).

The "researcher" in me tries to avoid the (in priciple now necessary) re-structuring of the book, with changing (let's say) 65 pages with 200 diagrams. Especially, when the "researcher" currently has a high working load to manage. Instead, a comment is inserted that makes the incorrect order of moves evident, asks for kind understanding that the editing has not been done, and points to the now correct sequence, which is hidden is sub-sub-variation XYZ.

As far as the core information is concerned, the presentation is still "correct". From the standpoint of the "researcher" everything is fine again.

However, the "editor" in me knows that this kind of presentation does not fit the expectations of the "usual" reader. The "usual" reader expects the "correct" sequence to be in the main line, with all alternate moves, mistakes, etc. explained in sub-(sub-)variations.

Approximately one time a year, the "editor" wins the "fight" with the "researcher" (a good moment to battle is when the "researcher" has just finished a larger task), and then several weeks will be filled with re-structuring work, which -- in principle -- does not add anything to the contents of the book.

It is this change of standpoint that Bantari and others think that you are unable to do.

Instead, you behave as the "researcher" in the above example, "proving" until the very end that everything was OK. The reader simply missed some comments on pages A, B, C, and D.
Author:  RobertJasiek [ Tue Oct 15, 2013 2:26 am ]
Post subject:  Re: Robert Jasiek's Go Theory Research
Cassandra, the editor in me frequently beats the researcher in me.
Author:  tchan001 [ Tue Oct 15, 2013 3:25 am ]
Post subject:  Re: Robert Jasiek's Go Theory Research
Personally, I don't mind people sharing their dislikes and criticism on RJ's thought and works on other threads he may have posted, but when it comes to a study journal, I think it ought to be a place where a person can get some helpful ideas about how to improve his game. Notice the section "study journals" falls under? >>> "Improve Your Game".

Please think about being more relaxed and giving more leeway when posting on someone else's personal journal. It's his thread, it's his world. If you don't enjoy it, there are plenty of other study journals to explore ;)
Author:  Bantari [ Tue Oct 15, 2013 10:36 am ]
Post subject:  Re: Robert Jasiek's Go Theory Research
RobertJasiek wrote:
Bantari, professional players are praised for their attention to detail. You seem to criticise me for my attention to detail (when speaking of me causing problems by being more rigid than most others (most amateur players?)), why? Because I dare to do what professional players are praised for?

Think of it like this:
- "Precision" the professional players profess helps them win their professional games, so its good.
- "Precision" which you exercise puts people off and causes a rift in the forum, so its bad.
Now where is the disconnect here?

Notice I put "precision" in quotes. This is because this is not what we are talking about, and I suspect we understand this word slightly differently in this context anyways. You need to understand that nobody has anything to say against precision - when you make your arguments, be as precise as you want, the more the better. Precision is not the problem.

From where I stand - and let me reiterate - the problem is twofold:

  • You fracture the threads by splitting the argument about a specific issue into a bunch of smaller arguments about each specific sentence/statement your detractors said. At some point one needs to realize that this turns into bickering and ads nothing to the main subject. From where you stand, I guess you call it "precision", and it might be precise, but its not helpful and it contributes very little. You might think correcting every single little thing you disagree with is "precision". Technically, it might be, but it is also very distracting and not always constructive. A much better "precision" is to carefully pick *what* you want to comment on, and try to stick to that with as much precision as you can. Of course, discussions are fluid, so its not always a cut-and-dry proposition... And sometimes there are two or three main points, so its not always easy.

  • You often miss the point when further arguing is pointless, when the bickering stops being productive. I call it your 'bulldog behavior.' Bulldogs are known for holding on to something once they grab it, so this is why. This is what I was alluding to when I said that sometimes it is enough to just say your piece, make your point or argument, and leave it at that.

Each of the two above behaviors can be annoying - but combined they create the problem we have now.

In short, what you might call "precision" - and think of it as such - can be a major distraction and thread derailment. I think it is actually much more precise to make a single well-thought-out argument for or against the main topic of the thread - than to insistently bicker about every word and sentence you disagree with separately.

In Go terms:
  • Think of it as a discussion 'honte'. Instead of making wild extensions into all kinds of unimportant places, or trying desperately to save minor and unimportant stones (just because you can), you should make one proper move, 'honte', safe and sound (or whatever) and remove your aji. This will change the flow of the game/discussion into your favor instead of you chasing your tail because of unimportant (or less important) stuff.

I have explained it all in detail in my previous posts, or at least tried to.
Still, I really want to help here, so keep coming at me with all that, I'll keep explaining as best I can, in public or in private. I see you making the effort, and I see you are getting much better. So its a YAY!!! ;)

PS>
Robert - I think the difficulty you might have here is that the issues/behaviors we talking about here is not conductive to be formally codified. I cannot think of a 'set of clear and precise rules' for posting I can give you and say: here, do this and you will be fine. Its all about such fuzzy and imprecise concepts as 'common sense' and 'social acceptance'. And what makes things worse - it all varies from person to person, so what I see as a problem, John mike like, and his problem might be something I don't mind at all. I have noticed in the past that you often need/want precise formalism, or at least feel much more comfortable surrounded by such. I am not sure if anything like that can be defined to solve this problem. Maybe this is why it is so hard...
Author:  Boidhre [ Tue Oct 15, 2013 8:34 pm ]
Post subject:  Re: Robert Jasiek's Go Theory Research
I think just from reading one thing that gets peoples' backs up is when you come up with a definition for a common go concept. Your definition is usually (to me anyway) interesting to think about but if it jibes with people feel for that concept the result is a thread of nit picking over definitions which generally misses what your definition was trying to capture and mostly consists of people trying to show how what you defined is not the concept in its popular understanding.



This is considered a very serious problem in Psychology of Mind, if I try and define what a feeling is the chances are it will be somewhat different to what you think a feeling is and we end up arguing about this difference rather than the implications of my definition. One way some philosophers got around this was to avoid defining common words and create their own jargon to remove the argument about the terms meaning vs popular meaning and just get to debate the idea they're trying to put forward instead. If we're talking about QThickness (to come up with a trite term) you can very easily deflect any digs about it not being the same [insert player's personal definition of thickness here] by pointing out it's not supposed to be the same and is supposed to be trying to reveal something about the game in a specific manner.


I don't know if the above will be of any use but I hope that helps. :)
Author:  RobertJasiek [ Tue Oct 15, 2013 11:08 pm ]
Post subject:  Re: Robert Jasiek's Go Theory Research
It is always good to discuss definitions and see whether they agree to prior understanding, feeling, tradition, alternative definitions etc. or whether the definitions or other understanding is better for the purpose of what we or some think shall be expressed. When pretty much the same is expressed, then the same words for terms should be used (e.g., thickness; if somebody really wants to express only the outside, he can still say what has been said before for this specialised case: "outside thickness"). When something significantly different shall be expressed, then different terms make sense (e.g., n-connected). If somebody's (e.g., my) definition is not good enough yet, it can be improved (e.g., solid connection).

However, the understanding / feeling of a player with a weak (and so different) understanding of a term does not beat the understanding of a player with a strong understanding of the term. Instead of falling into a bad feeling due to his weak understanding, a player can use the chance to acquire a better understanding (e.g., that there can also be inside thickness).
Author:  Boidhre [ Tue Oct 15, 2013 11:25 pm ]
Post subject:  Re: Robert Jasiek's Go Theory Research
RobertJasiek wrote:
It is always good to discuss definitions and see whether they agree to prior understanding, feeling, tradition, alternative definitions etc. or whether the definitions or other understanding is better for the purpose of what we or some think shall be expressed. When pretty much the same is expressed, then the same words for terms should be used


This is pretty much exactly what I'm talking about. You're deciding that your definition is suitably close to your understanding of a concept to share the same name. The problem here is that you're assuming the last bit is correct. This will bog down the discussion and you'll end up arguing about what thickness means with people rather than getting to talk about what you want to talk about.
Author:  Loons [ Wed Oct 16, 2013 12:00 am ]
Post subject:  Re: Robert Jasiek's Go Theory Research
On a different note, Robert could you give us some insight into your research process specifically regarding your evaluation of joseki?
Author:  RobertJasiek [ Wed Oct 16, 2013 12:01 am ]
Post subject:  Re: Robert Jasiek's Go Theory Research
There are terms for which
a) I strongly believe my definition to be correct (ko),
b) I have not seen one counter-example since I have written my definition (thickness, if understood to be generalised to include also inside thickness),
c) I try to be a bit better than a random go dictionary entry, but I am aware that more study and quite likely changes to the definition are needed (aji, if understood to be used for the bad possibility variety of the term).

For (a) and (b), counter-examples can be provided, if they exist at all. For (c), a pretty broad discussion is possible easily.
Author:  Boidhre [ Wed Oct 16, 2013 12:03 am ]
Post subject:  Re: Robert Jasiek's Go Theory Research
Sure Robert, I don't for a second doubt the sincerity of your belief. I'm also not saying that you're incorrect. I'm just saying you'll get grief on forums for doing it.
Author:  RobertJasiek [ Wed Oct 16, 2013 1:26 am ]
Post subject:  Re: Robert Jasiek's Go Theory Research
Loons, joseki evaluation:

1. I started with the ideas that the stone difference and the territory of a joseki must be known at all, i.e., determined and stated.

2. Since influence plays an important role (or I could say: the second most important role above the most basic aspects), I wanted to include it in the considerations. Many years ago, my first idea was to count additional territory in front of outside walls. This was too ambiguous, and any such model would have been too arbitrary. However, for several years, I did not have any better alternative, because I lacked a proper understanding of influence.

3. I needed good descriptions of the influence concept, so I searched for related models and found them in mobility difference and my formal influence definition. I considered whether either could be used to describe conveniently and usefully the influence of josekis. I found that, for that purpose, mobility difference was not good enough and, for that purpose, my formal influence definition was a conceptual, too complicated overkill.

4. For other joseki-related purposes (evaluating early corner moves), I continued research ideas by others to find a new, reasonable justification for the miai value 14 of an early corner move.

5. My thinking about the (3.) concepts let me discover the concept 'influence stone difference'. I found it to be about as accurate / imprecise and convenient as useful and meaningful for josekis, so I used it.

6. I had the idea that the influence stone difference of every joseki must be known at all, i.e., determined and stated.

7. I reflected more carefully how to do territorial positional judgement (conceptually, Cho Chikun in his book was by far not good enough), because accuracy there allows a more accurate value of a joseki's territory count.

8. So I determined and stated the stone difference, the territory count and the influence stone difference for ca. 400 more or less representative josekis (excluding purely tactical josekis immediately leading to very complicated middle game fights).

9. I did not expect anything more, but, during (8.), I strongly suspected that there appeared to be a systematic behaviour relating territory count and influence stone difference. Since the stone difference obviously affects any such behaviour, I studied firstly separately for stone differences 0, 1 or 2. For the 400 josekis, I listed the pairs of territory count and influence stone difference.

10. I reflected which function or operation should relate territory count and influence stone difference for the case of the stone difference 0. I concluded that the right operation is division, because it expresses how many excess points of territory one player makes per number of excess infuence stones of the opponent. To simplify, I took the absolute.

11. For those of the 400 josekis with stone difference 0, I listed territory count, influence stone difference and their ratio.

12. I classified the combinations of territory count, influence stone difference and their ratio.

13. I studied and found explanations for the value classes.

14. For the stone differences 1 and 2, I listed territory count and influence stone difference.

15. I looked for a relation between the value classes for stone difference 0 and (14.) and found such relations. They use also (4.); I provided an explanation for such use.

16. I double checked my developed theory by applying it again to the 400 josekis and a couple of non-josekis. It applied very well to 399 josekis, but not to 1 joseki. However, there is a perfectly good explanation for this exception: the joseki involves a global ko exchange, and my theory was not meant to be a middle game ko fight evaluation theory.

EDIT: minor corrections.
Author:  leichtloeslich [ Wed Oct 16, 2013 1:41 am ]
Post subject:  Re: Robert Jasiek's Go Theory Research
RJ wrote:
I strongly believe my definition to be correct (ko)

Definitions cannot be correct. They may be more or less useful, but the word correct makes absolutely 0 sense in this context.

And you could use the most esoteric, abstract, counterintuitive and hard definitions imaginable, as long as they would produce useful results (read: actual non-trivial theorems), I believe many people would be inclined to learn them.

In the absence of such results, they're just definitions, formalizations of spoken language. And in my opinion (and I suspect I'm not alone here), lacking any theorems that can be derived from those definitions, they are basically useless.
Author:  Cassandra [ Wed Oct 16, 2013 1:49 am ]
Post subject:  Re: Robert Jasiek's Go Theory Research
RobertJasiek wrote:
There are terms for which
a) I strongly believe my definition to be correct (ko),
b) I have not seen one counter-example since I have written my definition (thickness, if understood to be generalised to include also inside thickness),
c) I try to be a bit better than a random go dictionary entry, but I am aware that more study and quite likely changes to the definition are needed (aji, if understood to be used for the bad possibility variety of the term).

For (a) and (b), counter-examples can be provided, if they exist at all. For (c), a pretty broad discussion is possible easily.

Definitions do never stand on their own, but persue a specific purpose.
In my understanding, definitions are used in general like

"For the following, J-Thickness shall be understood as ..."

with its most important element highlighted. Please note that this is NOT "J-Thickness", NEITHER what follows the "as" !!!

The "following" is a closed system (e.g. a book, or a theory), so a) follows from your internal attitude that the results of your work have to meet your very high-level requirements. This means that one can be sure that your definitions are well thought-out, and reliable. This also means that "correct" has to be understood as "consistent" inside the "following".
So, a) is NOT the problem !!!

b) follows immediately out of a). It is very unlikely -- and will happen only by accident -- that you will overlook a tiny detail within your work. This also is a result of your very high-level requirements, concerning the results of your work.
So, b) in NOT the problem !!!

The main problem is c) !!!
Your statement includes that you compare the result of your work with something OUTSIDE the "following". But what is the basis to do this ???

Have the "usual" readers of a "usual" Go dictionary complained en masse that the Go dictionary does not fulfil their needs at all ?
If so, how should another explanation help that is valid only INSIDE the "following" ?

In addition, there might be no need to have an extended, more "precise" definition of a very special subject.
Let me try to explain this using "thickness" as example, which seems to have lost parts of its meaning during its journey from Japanese to English. Please be aware that the majority of the English-speaking Go community has absolute no idea that the ship has lost some containers around Cape of Good Hope. So they do not miss anything.

An "unsettled" group that is surrounded by "already settled" ones, may be called "weak" or -- better fitting here -- "thin". It will be part of "common knowledge" that such a "thin" group is very likely to become much of a burden during the rest of the game, as well as an attractive aim for the opponent.

Let us assume that the opponent has just played the move, which settled the last of his surrounding groups.
Does it contribute much to the understanding of the position to call this -- somewhat special application of a -- move a "thick" one ?
If so, why should it contribute much to the general understanding of the game to try to develop a "generalization" of what we have just seen ? Just because it is a matter of course that -- if you wanted to have "settled" groups -- you would have to "settle" your groups.
Author:  RobertJasiek [ Wed Oct 16, 2013 1:54 am ]
Post subject:  Re: Robert Jasiek's Go Theory Research
leichtloeslich, when I say "my definition of ko is correct", I mean that it determines and distinguishes ko examples from non-ko examples. This is so for ALL known ko shapes and every non-ko shape I tested. There is (still) not even one counter-example to the definition of a ko that that the definition does not identify as a ko, or of a non-ko that the definition identifies as a ko.
Author:  daal [ Wed Oct 16, 2013 5:10 am ]
Post subject:  Re: Robert Jasiek's Go Theory Research
What does "influence stone difference" mean?
Author:  SoDesuNe [ Wed Oct 16, 2013 5:14 am ]
Post subject:  Re: Robert Jasiek's Go Theory Research
daal wrote:
What does "influence stone difference" mean?


Coincidently: viewtopic.php?p=150913#p150913 : D
Page 4 of 9 All times are UTC - 8 hours [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/