**Goproblems conventions**

Post Reply
Emile Wester

Post by Emile Wester »

Wouldn't it be a good idea to put some conventions in place that we all agree on and which can be a guideline for problem creators in how they present their problems?

There was a thread about tsumego conventions, but I think that other types of problems would also benefit from some agreed upon conventions, to prevent unnecessary irritations due to misunderstandings and disagreements.

Mostly this resolves to how much you should state as extra information in the problem comment and how much you may assume should be evident from the problem genre and stated goal of the problem. What I mean is: what should be marked 'wrong' and what shouldn't?

Connection problems:
Aside from tsumego, we have e.g. connect problems, problems with 'connect your stones' in the description. When should a variation be marked 'wrong' without comment, with comment, marked 'solved' without comment and with comment?

I think:
'wrong without comment': nothing good about your move, no connection
'wrong with comment': you didn't connect, but you tried a common connection move/tesuji which doesn't work in this particular case. It was a reasonable try.
'solved without comment': you connected and in the best way
'solved with comment': you connected, but there is a better or simpler way to connect.

It is also possible to simply say that you should *always* look for the best solution, even when there are more possible answers and it's not a 'best move' problem. Everything worse than the best would be marked 'wrong'.

I personally believe that people don't spend enough time on finding all the ways to accomplish a certain goal and that most people wouldn't even know that there were other solutions than the one they found, even though they found the best one first (maybe by coincidence). This is not how professionals consider a problem. As I have read somewhere (I think on sensei's), professionals consider all the possibilities in a problem, then pick the one that yields the best result. I think you learn the most from a problem if you consider all possibilities, not just the simple nakade that probably kills the group ("it's a go-problem, so it's probably that eye-stealing tesuji over there, that's sitting there so conspicuously..;)), but also e.g. seki's, possible ko variations, etc..
Ofcourse, if you've found a way to kill/live unconditionally, ko's become less important, but still, knowing that there is a possible ko gives you more possibilities to win the game. E.g. you have to kill two groups to win, but if you play in the first one, you can't kill the second anymore, but if you play in the second, you can still make ko out of the first.

I'm gonna stick to connection problems now. Tsumego 'how many ways' problems has already been covered. I think that 'best move' problems are always under debate (not everyone agrees with what's best ;-). Joseki and fuseki problems pretty much the same.
Endgame problems are often pretty straightforward, though there is sometimes a discussion about sente/gote and the temperature of the rest of the board if that isn't mentioned.

Please tell me what you think. I hope everyone thinks that some agreement on conventions would benefit the clearness of the problems and would bring the problems' goals more forward, instead of being source of debate on their interpretation (which I hate personally).
Anyway, I don't say that what I think should become the standard or anything, just that some conventions on which you can rely to interpret the problems would be handy to have.


{Posted by Emile Wester}
Post Reply