We are finally working on putting together a new interactive tutorial system that will hopefully do a much better job at introducing beginners to the game.
We’re taking a page from lichess in the format (you will notice many similarities to https://lichess.org/learn/) as we think this is format a bit more approachable for players new to the game. This we hope will be a good improvement over our current tutorial format which forces you through a very linear path and requires too much reading by todays standards.
However, while I can write code, I’m pretty horrible at coming up with interesting puzzles that help teach people, so I would love some help and guidance for the actual content. I’d like to put somewhere between 4-7 “puzzles” under each section that focus on the particular section topic.
So for anyone interested in helping come up with the little mini puzzles, your help would be welcome!
Here’s the current need (I’ll update this as we go along):
4-7 puzzles teaching how to defend and create eyes.
4-7 puzzles on ladders 4-7 puzzles on snapback
4-7 puzzles on identifying false eyes
4-7 puzzles on cutting stones
4-7 puzzles on jumping stones
4-7 puzzles on semai 4-7 puzzles on identifying seki 4-7 puzzles on ko battles
Any other puzzles on other fundamental concepts are also very welcome, I’d be happy to add new categories.
Thanks in advance for everyone who contributes!
P.S.
To contribute, just make a demo board of a position, upload an SGF, or create a puzzle - then paste a link to the board/sgf/puzzle here along with a quick one line description about what category you think it belongs to.
I think this is a great idea! I love the idea of community-driven “intro to go” content.
I just have one suggestion: ask that all of the contributors agree to releasing their content under a permissive CC (Creative Commons) license, and make the collection of content available under that CC license on GitHub.
After all, if it is going to be a community effort, it would be nice for the content to ultimately belong to the open-source community.
Make the chime audio optional, remove it some places (like during navigation), or remove it altogether. A chime might offer a nice reward signal for when the user gets something right, but it quickly loses its meaning and gets a bit annoying if it plays during every navigation action.
I’m not so sure about the organization of the content:
Building and invading territory are quite complex topics, but they only seem to be represented on the beta page as “territory”, which is listed under fundamentals. I think that merely counting territory/score in a finished game position is a fundamental topic, but problems dealing with how to best build or invade territory are quite advanced.
Counting liberties is listed under advanced. I think this should be put under the fundamentals category, assuming that it will be very basic exercises for familiarizing the user with the concept of liberties. I guess problems in this area could just be answering how many liberties does a particular chain have or identifying which chain has the least/most/fewer/more liberties, etc. It does not have to be too complex, since semeai is another topic for applying this knowledge.
Since this is aimed at novices, I would recommend placing a rough English translation adjacent to the Japanese go terms (wherever they appear), like this: “Capturing Races (Semeai)” or “Semeai: Capturing Races” or “Seki - Mutual Life”.
Hi, as an intermediate beginner, I think invading and building groups that will live is REALLY IMPORTANT! It’s the one area that I am consistently aware of where I really need more understanding. It is rarely taught in beginner lessons, and is so important. It would be refreshing to have it included. Don’t dumb it down! Hold the beginners as being able.
An interesting point. I wanted to find something like this for the Android app that I am building… it would be nice if I could use this set that the comunity is building.
On another note I do agree that we need translations (I am 10k and had no idea what semeai meant).
As usual after a long pause anoek comes out with a surprise!
So naturally I think this is all going to be handled by dans, right?
Ladders should include nets, or we should have separate category for nets. E.g. “White wants to capture a stone, but the ladder doesn’t work. How can white capture the stone?”
I imagine there’re whole books about invading, or at least giant chapters of books, naturally there’s going to be some simplification of it.
This is fairly in line with a beginner’s series I started a few weeks ago… Sadly I don’t have much time right now to work on that so I’ll just throw in my 2 cents.
I’ll have to largely agree with yebellz.
Eyes are a step up from liberties (think about it this way: if you place a single stone, you can already point out 4 liberties, but to make something resembling an eye we need at least 2 stones)
Territory relates to the game of Go as, say, money relates to vocation. Territory is a corollary and shouldn’t be thought of as the goal. Territory is how we compare two players’ overall efficiency when pitted against each other. It should be taught last. Instead, teach liberties and connections first.
Invasions are a strategic concept and don’t belong to fundamentals, which should largely focus on tactics.
As I see it, this would be a logically coherent list of topics and subtopics:
Liberties and Capture
– Single stone capture (atari, self-atari)
– Multi-stone capture (atari, self-atari)
– Ladders (working, broken, fake ladder)
– Indirect capture (e.g. loose ladder, net)
Legal and illegal moves (surrounded intersections)
– Recapture
– Snapback
– Ko
– Eyes
– False eyes
– (Unconditional) Life
– Mutual life (Seki)
Connections and Cuts (Shape)
– Nobi
– Kosumi and hane
– 1-space jump
– 2-space jump
– Keima, Ogeima
Not sure if its good idea to instruct beginners to watch youtube videos straight away, but it might be nice idea to give them a post note, like " congratulations, you finished the tutorial, now feel free to play lot’s of games on ogs and you can also check out OGS Link Collection for more studying material" at the end of the intro?
I am always afraid of teaching too many things in a single time and get the newcomer bored.
Some list here looking a bit too much already even if all of these are necessary.
Like to show different connections between stones.
Maybe a way would be to split in successive lectures so the reader can come back after a few games?
I agree about the chime for every time navigation buttons are pressed being a bit much; I don’t imagine this being a feature of the final iteration of this project.
There’re all these difficult questions of how the material should be organized and how difficult it should be. Probably we need to have experienced smart people figure it out.
But! It wouldn’t hurt to try for us kyus to compose some simple problems. I assume the point is not to have difficult problems but straightforward ones showing the technique. So I picked a topic I feel I sort of understand - snapback, conversed with @chartreuse about what it truly means to be captured in a snapback, and came up with these problems. Naturally since they’re from my head, I don’t know how correct or appropriate they’re.
By the way, if it’s a “call” wouldn’t it make sense to put a banner on OGS or maybe post on reddit?
I appreciate your effort, but considering we’re talking about 25-30k here, these problems seem much too difficult.
In 1 there are too many things going on, it’s not just a snapback, it’s also a semeai. I think we should focus on a single, simple aspect per problem. The more stones are involved, the less likely beginners are to see anything. Remember they don’t have Go / group-vision yet.
Try to keep the number of stones on the board to a minimum, as well as the number of steps to solve the problems. Problem 4 requires reading 7 moves ahead (if white resists) and though it may look simple to you, I doubt it’s the same for 25k. Same for problem 5. If you know the correct intended responses, you can get there in 5 moves (still too many), but if black resists, it takes 9 to capture (e.g.: w E6, D5, D6, C5, B5, C6, C7, D2, E5) and it’s not a snapback.
Please compare these to Cho’s tsumego. I suggest making it 1-2 steps except for ladders/nets.
We should probably invite some fresh 25k beginners to proofread these problems.
Most of the time they have not enough variation. It’s a bit tiring to put All but I find very useful for players who have a low level are reading.
You remember that: “but if I play here, why they even didn’t consider it?!”
Well, it’s a work in progress. Dumbing down is easier than smarting it up. As I said, I don’t really know how difficult we should aim. If it’s 25k-30k, then maybe snapback is too difficult in general and we should do problems to teach not to do this (real beginners game, though more like 40k):
If we completely go with lichess model, then look at what they do. lichess.org/learn teaches rules only, how pieces move, how the game ends, what’s stalemate (and uses no full board positions). Most advanced topic is check in two consecutive moves. So no techniques are taught, only rules.
Should we adopt that?
On the other hand https://lichess.org/practice uses full board positions to teach you techniques of chess - pins, skewers, forks and all that, and positions can be pretty difficult, and certainly not for complete beginners.
You can introduce a few “strategy” if it is easy to explain/understand, if the moves are almost obvious.
For example to not always follow your opponent (4 quarter territories not finished on the edge, if black start yose hane/connection and white always answer sure to lose…)
Another easy concept is center Vs edge lay down 3d line territory Vs 4 line center and ask what is bigger (btw showing the important line of building, not so hard to count territory and useful )
