Josekle #26 (7 moves)
This hurt my soul
This will make blind mode basically impossible 
maybe one-color mode is just a better game mode anyway
Fusekle #8 (8 moves, blind start)
Eh it wasn’t that bad 
Josekle #26 (17 moves, hard)
this one was rough on the other hand 
Is there a reason to check against Waltheri specifically? … Is it a better option than GoGoD?
Since GoGoD is paywalled, and because Fairbairn is notoriously (over-)protective of his copyright in even minor cases, I think there’s a definite possibility that problems could arise from habitually using GoGoD as a source for OJE, which isn’t a paywalled or Fairbairn-associated project.
Sure, GoGoD has more games, but ultimately it doesn’t seem right to base joseki inclusion on frequency data that the user can’t verify without purchasing a product. That doesn’t seem to gel with the position of OJE as a free resource.
Does a sequence have to be popular to be called joseki?
Does a joseki have to be popular to be selected as a default mode Josekle solution? I’d say yes.
Does a joseki have to be popular to be selected as a hard mode Josekle solution? I’d say not so much.
I’m really happy with how Fusekle has turned out 
I’m playing along most days, but since I have an unfair advantage (being familiar with the solution set), it’s a little hard for me to judge the difficulty. I think/hope that the combo of “one-color” mode and randomized orientation gives a good solving experience where there’s not much need for blind guessing, while also being challenging enough.
Thanks for an awesome game @yebellz!
Fusekle #8 (8 moves)
There’s a long discussion about this from a year or so ago.
But it’s named something innocuous like “Questions about the coverage of two 5-4 joseki in OJE”, iirc.
TLDR the current policy (if I understand correctly) is that sequences in OJE must either have some level of professional play (I think the number of five, ten, or twenty games was bandied around but never settled on) or be marked as not having it.
Sure.
A note saying Only played twice professionally, by Hashimoto Utaro in the 1920s wouldn’t go amiss.
That’s fair.
But how could you possibly be this specific, given such a small database? That comes across to me as very likely inaccurate. Maybe it would be better to say “only listed in two professional games on Waltheri” or something?
When it comes to Josekle, a comment wouldn’t really help, instead a tag would be more helpful, so it can automatically be filtered.
Have any go pros been spotted playing Josekle?
such a small database
Waltheri’s not that small. It has about 80k games. The total number of professional kifu (from career games) in circulation is, what, about 130k? Less? Somewhere between 110k and 140k is a number I think I’ve heard around.
Maybe it would be better to say “only listed in two professional games on Waltheri” or something?
Sure, that seems good. But if GoGoD has a third game, there’s no reason not to mention that.
Go4Go or any other database could also be cited, if it’s not too difficult for the user to get at the information.
I guess my suggestion was under the assumption things were being automated exclusively using Waltheri. But if you’re talking about doing it manually, then that makes sense, though if it’s already being done by hand, it might not hurt to give the extra data of how many games from each database that the commenter used are being referenced with which number
Edit: that could have the added benefit of raising awareness of the other specific tools and databases that exist, and clarity about these things always helps avoid confusion
Also, I’m no copyright expert but can’t up to 10% of a “work” be extracted legally without permission?
Since that 10% will be at at least 10k, that means we’d be fine to host as many cited GoGoD kifu as we like in OJE-associated library uploads. Kifu themselves have no copyright, * only commentaries and the compilation of kifu into a database.
Yeah, I was definitely discussing manual operation.
(* and I fail to see why Fairbairn would hold that copyright rather than the players or event organisers, anyway)
Perhaps the conversation from here on should be moved into its own thread (I’m OK with having my Josekle results deleted).
A thread about copyright in Go would be really useful
Covered~
That would be really cool.
Maybe I’ve already secretly hoped that Go Pro Yeonwoo might show it in one of her videos, and even thought about sending her a link, but felt a bit self-conscious and perfectionist in worrying that the game and interface could still be polished a little bit more.
On an “unrelated” note, what further UI / UX design and development do you think should be prioritized to really enhance the game experience??
I’m thinking about the following ideas:
I just noticed that the AI on one of my 9x9 games from last night recommended that we play into a position matching today’s Fusekle!
Our moves are in a totally different order and half of the colors don’t even match. But maybe it would be optimal enough to show up as a puzzle some day.
@yebellz could you add a Fusekle link to your initial post? It’s kind of hard to find.
Also, apparently:
Josekle #27 (7 moves)
I was using OJE today to figure out the hard solution, but I can’t find the last move with OJE, or I’m doing something wrong.
