This has raised the question about the place of fuseki in OJE, again.
At the beginning of OJE the basic fuseki were entered, even though clearly they are not joseki and do not meet the policy guidelines.
I can’t clearly say why - it felt right at the time. Maybe we just needed some basic content. Maybe it was because Josekipedia had them.
Since then, the contributors have largely focussed on actual sourced joseki.
My feeling is it should stay that way - the goal of keeping OJE “lean, not bloated” remains. OJE is not about “joseki used in specific positions”, it’s about “joseki in the purer sense - sequences that make sense where ‘all other things are equal’”.
when you choose joseki in top left corner, for example, there is difference when in bottom left corner is your stone and when there is stone of opponent.
It probably risks a lot of repetition to start putting in fuseki and then all the relevant joseki on top of that.
My personal opinion was always that people who want to learn Go and improve will come up against openings and fuseki like the Kobayashi opening where moves like a two space approach to a 3-4 stones will come up often in older resources. Now it’s likely that a two space approach won’t be joseki if the one space approach (high or low) or attachments for example lead to better or equal results
so in theory they have no place in OJE and yet eg https://online-go.com/joseki/45853. I do think non-joseki sequences in the sense of getting an even corner result will still be useful to many people, but of course would need to be marked clearly if entered.
I think that if a sequence such as the one linked is not even, but still plays a role in certain fuseki, I don’t think it should be added in the way that was done in the OP. I think instead it should just use board markings, like letters/numbers x’s etc to indicate the context of the fuseki instead, so that way it doesn’t muddy up the location of the resources in the OJE move trees.
Even better would be the ability to add stones to the board like with the stone editor tool in the analysis/demo board. That is, if one got to the end of a joseki sequence and wanted to mention that it happens in such and such situation you could add a bunch of white and black stones to the end position in the same way one might add X’s and triangles, without it storing these as extra moves.
My personal opinion was always that people who want to learn Go and improve will come up against openings and fuseki like the Kobayashi opening where moves like a two space approach to a 3-4 stones will come up often in older resources. Now it’s likely that a two space approach won’t be joseki if the one space approach (high or low) or attachments for example lead to better or equal results
I’d actually quite appreciate a rundown of where Kobayashi theory is at right now.
Yesterday I was teaching a DDK about the Kobayashi and explaining the traditional two-space approach lines, since I don’t know the ideas involved in approaching more closely.
Some people have explained the new meta, at least in passing, on GoKibitz but I’d have no idea where to find those comments now. I think this is well with @gennan 's ballpark if he has the time, but I’m sure many other players have a good grasp on this as well.
I agree with you that including fuseki in the joseki explorer would blow it up too much. But there is a widespread misconception on joseki that I’ve tried to communicate to people throughout my whole go career, and I’m just itching to do so here, sorry.
What does it mean “all other things being equal”? I’ve also heard “joseki are optimal sequences that give both sides an equal result”. The way I see it, both of these sayings are unjustifiedly glorifying joseki. It’s not like some deity handed us a list of joseki that are “perfect sequences”.
Joseki are just variations that some players and / or computers came up with at some point. Not all variations are considered joseki obviously, so I’d say “josekis are local variations that are commonly played and largely recognised as good in many situations.”
On the topic of fusekis, as @shinuito already pointed out, there are variations that may be considered joseki only in the context of specific fusekis, like the kobayashi. If we include such josekis without mentioning in what fuseki they should be played in, to me that doesn’t seem didactically right
I’m not sure how I would approach counting that eg with waltheri, excluding cases where for example it’s played much later in the game than the first or second corner sequence.
I’m not even convinced fuseki fit in OJE really.
Some separate fuseki or opening explorer might be better, which could link to the OJE even for certain corner patterns and variantions.
Better to store joseki only, but there should be explanation when to choose which.
In my collection of bot joseki, I use circle to show white stones and X to show black stones. Stones of joseki itself are stones, and stones that were around when bot played it, are circles and X.
3 years ago I made a video (Learning from AlphaGo #6: the Kobayashi opening) about AI opinions on the Kobayashi opening. But at that time, LeelaZero was not as strong as AlphaGo and KataGo did not exist yet, so I had to go by the AlphaGo Teaching Tool, supplemented with what else was available at the time.
Most of that video may still get approval of KataGo, but I haven’t checked it.
