Hello. What this means is to have a go board, that has a solution or branches or tree ready, and the player is trying to find the right move. This is good for a guidebook in starting go (For ogs), making a perfect go bot software, and tsumegos. Thank you.
This is for the site and not the forums, but it is good though if we can use it in forums.
Sounds interesting. Do you mean something with a playing AI attached so it can counter oneās test moves?
IIRC @matburt or @anoek hinted at that they had something interactive/tutorial-like up the pipes ā¦
Cheers, Tom
Something like solving Tsumego is actually planned for OGS. We will announce further information when itās closer in development. 
Will need stats on the life-and-death problems. Let registered users tag problems from a predetermined group of tags. Perhaps auto complete. Maybe even as tsumego score. Also, let the community throw all of their problems into the system. The community can organize it. Make sure there is an āunsolvableā tag that would refer problems to a member of the team for removal.I am sure you guys already have this in the works, just wanted to mention it.
The AGA wants to do something like this but I think there are going to have significant problems with implementation.
Go problems has no oversight and generally sucks.
*@#$! goproblems.
If they were a solution, the AGA would not attempt to do it right and nobody here would ask for a life-and-death practice system.
Look at it this way: why was OGS developed when we already had KGS?
Go ahead, answer.
Sorry if I offended you. I was simply mentioning it as an example to demonstrate the concept, which seemed a bit vaguely described in the posts so far. I doubt that everyone knows about goproblems, so I wanted to make people aware of it, too. I wasnāt even sure if you had heard it.
Actually, Iām not a huge fan of how goproblems is implemented either, and dislike that they claim copyright over their community-generated trove of go problems rather than making things more open.
I think the real challenge probably lies not in the technical aspect of creating an interactive tsumego solver, but rather in collecting/producing, as well as curating/organizing, an effective collection of problems. As youāve alluded to, this would be huge undertaking that would probably require the collective efforts of a community. Designing the features to enable and best harness the efforts of a community would be the real challenge, and the exact details certainly arenāt obvious.
It will probably take a lot creativity and a strong community spirit to build a better tsumego collection and interface, but immature hostility wonāt be helpful in finding that solution.
Youāre right, I should apologize: Iām sorry.
I am frustrated by the way life and death exercises play out on the Internet.
Gochild and go problems should be models but we need to be far less restrictive than either of them.
Go problems gives you the Soup Nazi treatment when you get a wrong answer. Go child does nothing when you get the wrong answer. You only see a stone when you click in the right place. Too many clicks and the problem starts over. And then there is go grinder which is like being ātrainedā by Pa Mei in Kill Bill.
Iād like to see the freedom of the demo board here on OGS with, maybe, a quick green glow fade when you play in the correct spot. Like the demo board system that OGS already has, explanations can be attached to specific moves.
I think the life-and-death exercise system should have the move tree available but closed by default. This way, the life-and-death exercises operate the way they would from a book. You play on the board and, if after too many variations it becomes frustrating, look at the answer.
Classification is important but I donāt want to give people the opportunity to put any tag. They can select from a predetermined list. I think go child has a pretty good classification system and strong players here may be able to add to it. After implementation, only a few players should be able to add classifications or remove problems.
As to the rating system of these problems, it could go two ways:
Tag a problem as āunsolvableā or ācandidate for removal.ā When this happens, it goes to a submission page and requires an explanation. If youāre smart enough to see a problem like this, you are smart enough to tell the team why. This way, OGS does not get trolls tagging the entire system as unsolvable.
Thumbs up or thumbs down. Thumbs down refers the problem to the team immediately. Needless to say, this is troll candy. A five-star rating wonāt work because it does not communicate anything and is easily misinterpreted.
So, in short, I want to see the system built from the demo boards concept.
Do any of you know www.321go.org? IĖm almost through there. (And yes I can understand a bit Nederlands if necessary, but usually it isnāt ;-))
Other than there, Iāve mostly only done Tsumego on iOS and like those very much, and quite some on Ootakamoku, and very few on GoChild, which I found a little bit too infantile on one hand and complicated to handle on the other, IIRC, and Goproblems.com which I canāt even really remember.
Greetings, Tā»m