Meanwhile I’ve done this almost 1.800 times … it tells me that I’m probably around 10k I know well not to take this serious. But I’d really like to know whether strong players would recommend to use this as training … or perhaps better: up to which strength could this be useful? I have the impression that at my rank (13k EGF) it might help me train some “reflexes”.

What I’d really, really, REALLY wish for would be that it explains to me why some of my moves are wrong and what the thoughts behind the recommended moves are, even if very short, like “because of diagonally opposing corner” or “must survive”, since I sometimes just don’t get it.

Anyway, I enjoy it, and I still have the subjective feeling that I learn something, no matter how nebulous, thank you.

<edit>

What would also be nice, I think, would be to see the opponent’s next two or three “best” answers when hovering over a “recommended” stone.

</edit>

Greetings, Tom

_________________
“The only difference between me and a madman is that I’m not mad.” — Salvador Dali ★ Play a slooooow correspondence game with me on OGS?

My impression is that many 'wrong' answers are not actually wrong, or not very wrong, but just have not been played in the past couple years, since the database is just based off of recent pro games. As a result, even though I would love to have some sort of explanation as well, I don't think it would be easy to create a correct one. The FAQ lists adding proverbs to the problems at some point, but many of the problems I think are beyond proverbs to explain in detail.

I have noticed that many of the problems can boil down to 'do you know the joseki being played here'. Similarly, you can play in the right area, but be one point off and be completely wrong on a problem, regardless of how much poorer that move would turn out. I think this is due to the way the problem info is collected, as mentioned above, but I don't have a better suggestion offhand and do enjoy the site.

One website can't do everything Perhaps you should do some reading, for example Yilun Yang's Fundamental Principles of Go or Opening Theory Made Easy by Otake Hideo and then see how well the "correct" answers jibe with what you've learned.

_________________
Patience, grasshopper.

The idea of presenting test questions whose difficulty matches the ability of the test taker is sound, both in terms of pedagogy and of accurate assessment of the ability of the test taker. However, when you start with questions of unknown difficulty and test takers of unknown ability, that is not an easy thing to do. The difficulty is further compounded by not knowing the answers to the questions with certainty.

There are two major properties of test questions, the difficulty of the question and how well it discriminates between test takers of different ability. The test takers have similar properties. For instance, a person may do better on hard questions than on easy ones. Neither the ability of a person nor the difficulty of a question is one dimensional. They can be reduced to single numbers only approximately or on average. (You see the same thing with go ratings or ranks.)



This example is based on one by Go Seigen in his 21st Century Go series, vol. 4. Go to . You will see six options for . All are pincers. Let us assume that White should pincer. Even though Go Seigen may believe that one pincer is better than another, he did not address that question. Rather, he focused on the practical matter that gave White many options. (I'll come back to that later. )

To me this seems like an easy question. To pincer or not to pincer. (As Hamlet may have put it. ) My guess is also that it differentiates between SDKs and DDks pretty well, even though a lot of DDKs will pincer and some SDKs will not. So maybe it has a difficulty of around 12 kyu and a discrimination score of around 0.7 (on a scale of 0 to 1). Now, it may be, if we look closely at the specific answers, that it could discriminate between players who play on the second line and those who do not. That would mean rating the answers, and not just dividing them into pincers vs. non-pincers.

But if we present this question to people around 12 kyu (in terms of fuseki ability), we are not interested in sorting out those whose non-pincers are worse than the non-pincers of others. We want to present easier questions to them. Nor are we interested in sorting out those who make different pincers. Go Seigen included in his example the three space high pincer. I do not like it, myself, and I suspect that few pros would play it, if any. But for a 12 kyu problem it should be one of the correct answers. Perhaps some 1 kyu or stronger players would choose it. If we consider it to be incorrect, then the discrimination score of this question may drop. It may become more difficult and less discriminating at the same time. We want to present harder questions to the 1 kyus.

What about ? Go Seigen does not like it, nor do I, nor, I expect, do most pros. My guess is that the one space high approach is popular. But Go Seigen does not fault on theoretical grounds. He does not say that it is objectively worse than other approaches, and give a variation or two to show what is wrong with it. Rather, he makes the practical point that it gives White too many options. He likes the large knight's approach, which reduces White's options and reduces the severity of a pincer. So if the question were where to play , I think that the small knight's approach should be among the correct answers.

Another play that has fallen out of favor is the large knight's response to the small knight's approach to the 4-4 stone. That was popular for many centuries. But in the 20th century Go Seigen argued that the small knight's response was OK. (If you start with one player on the 3-6 and the other on the 6-3, an obviously equal position in the corner, the 4-4 is a good play. So it must be OK to respond on the 3-6 if you have a stone on the 4-4 and the opponent plays on the 6-3.) That does not mean that responding on the 3-7 is bad, but it gives your opponent a whole lot of options! Again, without a theoretical argument against the large knight's response, it should be included among the correct answers in many positions, even if it does not show up in any recent databases.

There are not only differences in style in go, but differences in fashion. The Mini-Chinese was in the textbooks at the start of the 19th century, but went out of fashion, only to stage a comeback in the late 20th century. Just because a play is not currently popular among pros does not mean that it is inferior.

Now I do not know how much statistical sophistication Ootakamoku's web site might need, but I do think that it would be advisable to try to weed out positions that do not discriminate well at their level of difficulty. I also think that if a pro makes a play in the database, it should be considered correct, even though some mistakes may be counted as correct. The site is not presenting any positions to pros, in an attempt to rate their abilities. Pro errors are better than amateur shodan errors, as a rule. Including them among the correct answers in positions at the amateur shodan level may well improve the discrimination scores of those positions. I also think that it would be advisable to extend the database back to the early 20th century, at least. Yes, we have made progress in the fuseki since then, but a large database is good for the discrimination scores of the questions. For dan level positions, it might be better to go back only to the mid-20th century, however.

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

If there is a problem like which pincer to choose, and it would only have say.. 2 correct pincers out of the 6 possible you show. And now we worry that this simple position has too harsh standards for ddks, who should be content with merely finding any pincer. But if those criteria are too harsh, doesn't that imply that the ddks get this position wrong too often? And that in turn makes system reclassifies the problem as a harder problem and presents it to users who can see the difference between the pincers and have a good chance of picking the correct pincer.

So it should correct itself in time. Most problems presented to ddks are, answer the peep, save the stones in atari, answer approach on hoshi stone, etc.

Ootakamoku,

Really nice work! I notice that you now have active links to pro games. This has already exceed my expectation. To improve it, first you don't have to show the go4go URLs on the page - just make the game information clickable so that you can save space when there is a long list of games matching your pattern. It would also help if pro games open in new windows. Finally, I am sure you are working on this - you can append another parameter to the go4go URL to jump to the particular move that match your pattern.

Simply reclassifying a problem as harder does not necessarily improve the discrimination. Suppose that only 2 pincers are considered correct, based upon the database in use, and that that makes it, say, a 7 kyu problem. The trouble is that at the same time you are increasing the number of players who are stronger than 7 kyu who get supposedly wrong answers, because they played one of the other pincers. So the problem is not as good for discriminating between better fuseki ability and worse ability. But that is what you want it to do.

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

Increasing the rank for it mitigates the problem, hopefully to the point that it becomes irrelevant on the larger scale. In any case, ideal situation would be to know exact point loss from each move, compared to the best possible move in the given situation. And use that to judge the players fuseki skill. However having no such data available, I make do with what is almost binary data, but it seems to still yield accurate result with increased sample size. Just like in go, we can play better or worse and anywhere in between, yet the end result is almost binary, you either win or you lose. Here we are working backwards, we know the move was almost perfect, or it was not perfect, yet this distinction multiplied by many others from different situations adds up to quite accurate prediction of eventual skill.

Increasing the difficulty level (rank) of a problem by counting some acceptable answers as unacceptable does not help. In fact, it makes things worse. You do not have to take my word for it. Take a look at the testing literature.

Edit: And when your examples are biased, such as by drawing them all from recent games, that biases the concept of skill.

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

I really love this initiative, and it's a great tool already. Thanks.

I had an idea as I was browsing through the discussion about some 'wrong' moves that might actually be not so bad. You could create a system where 'highly contested' positions will be saved somewhere for your viewing. Like, let's say in a particular position a sufficiently high number of players will play the same 'wrong' move. Then you get a message about it and you can view that position. Then you (or someone else/a few others) could either a) let the program know that this is actually a good move and change it from 'wrong' to 'right' or b) make a note as to why this is not a good move.

Of course, for this you may need some help from stronger players. But maybe you could try it with a decently high threshold for what counts as a 'highly contested' position. If the positions that the system feeds you are not too many and/or too frequent, at least I wouldn't mind having a look at them and passing them on to pro players if I can't decide whether it's a good move (a), or why it's a bad move (b).

I think in the long run this could improve the teaching potential of the program a lot, especially when players can get advice on why their choice was a mistake.

Cheerio, happy new year!

Biggest lesson for me, up to now, is this:

about 70% of the “WRONG!” messages I get are because of me focusing on one corner while the fire’s in another corner

—> I need to look everywhere first before I re/act too quickly, #wholeboard


Then again, this lesson is not new to me. It’s only that I seem to have to repeat classes


Grtz, Tom

_________________
“The only difference between me and a madman is that I’m not mad.” — Salvador Dali ★ Play a slooooow correspondence game with me on OGS?

I get this too, but I think part of it may come from not being part of the flow of the game, since nothing is numbered.

For me its very rarely about playing in the wrong area of the board. Well, maybe occasionally if its about "where to spend sente", or which approach to play. But as for defensives moves the order of importance of different moves usually seems rather obvious.

It already does this. Its the reason why it states the strength of those who the correct answers are based on. It could well be for example 7.5dan+ for some problem if there are lots of answers, but very few strong answers, then it cant afford to be picky and accepts slightly weaker answers as "correct" since so many weaker players will benefit from the increased diversity of answers in that problem.

Looks like a very useful tool for grinding opening positions - i like it!

And here's what I think would make it even better:

1) After a right answer, don't jump to next problem right away. Instead, have a big "next" button on the side and show all the options like with a wrong answer. Only move on after I decide to click on the button, or perhaps just somewhere on the board. (I think this is pretty big, the following are not as important)

2) After I signed up, the text on the right side doesn't reset between problems and keeps scrolling (did it do that in the first place?) and is maybe a bit too much of a wall of text. Have it clear the area after each problem, so that it shows the standard "your ranke is" etc. only for the current problem.
3) On my computer (newest Firefox) the board position is very slightly off. There's a small part of black visible on the bottom (this looks normal), but on the top my bookmark toolbar covers a few pixels of the board's edge. I can also keep scrolling up (because of the wall of text on the right) and as the board is at the bottom of the page, it will disappear.
4) Not sure if feasible to implement, but browsing different paths that occurred in pro games directly on your website would be really cool. (i.e. after I answer, I can keep clicking on the next move that took place in pro game X)
5) The link to pro games, while it opens up in go4go.net, doesn't seem to actually have the position I just saw, and it opens up somewhere in end game each time.

In any case, great work and regardless of the above, I'll use it for some fuseki study from now on.

+1

+1

I also noticed this change … did it happen today? And I ditto what walpurgis says about the wall of text.

confirmed with Chrome/OS X

+1


Same here. I only hope I don’t learn wrong moves for my strength … not sure about how to answer follow-up moves.


Greetings, Tom

_________________
“The only difference between me and a madman is that I’m not mad.” — Salvador Dali ★ Play a slooooow correspondence game with me on OGS?

Not sure about this. I rather ensure a fast experience to those who want it. Especially now that there is the "possible answers" link where you can go look at the answers if it really interests you for that position.


Working on this...


Fixed.


Unlikely to happen, sry.


Fixed.

It now contains 15000 pro games, all from past 5 years.

I have something working now to address this, let me know what you think.

This is comfortably the most addictive go site ever.