There is no native functionality to move to a random next move. This comes exclusively from LiChess Tools. So you need some other shortcut than what Lichess uses natively.
If you set your study chapter to Interactive Lesson type, then the prc:N functionality is being recognized by it and you don't have to make any move. Of course, that means enabling Play all variations, that I still don't understand why you would have a problem with.
I could, maybe, implement a toggle that would switch the functionality of the right arrow to the one Ctril-right arrow has. Would that work?
The strange thing is that the study chapters are set to Interactive Lesson type, and yet prc:N is still recognized as a comment rather than a command and is displayed in the practice mode. The “Play all variations” option is enabled. I have the second newest version of LiChess Tools installed (v2.3.142).
I just want prc:N to be recognized and executed as a command, not displayed as a comment. There should be no display during the selection mechanism; it should simply choose a move based on the specified selection probability at the branching move.
Have I done something wrong? Why is it not working? I used prc:33 as a comment at a branching point, and it is only displayed as a comment in practice mode.
Aah. It's the display that bothers you. Yeah, I can fix that.
Wait. What do you mean Practice Mode? Practice is playing with the computer. It has no relevance to the existing PGN.
Explorer Practice also ignores it. It takes the info from the Lichess db.
I mean the interactive mode, which I have enabled for every chapter in the entire study. While it does bother me a bit that prc:33 is displayed as a comment, what bothers me much more is that it is not executed as a command! If something is displayed as a comment, it can't be executed as a command, right? Either something is displayed as a comment, or it is executed as a command.
I'm not exactly sure what you mean by the Opening Database Practice Mode. My study does only include opening positions, but it's just a study and doesn’t directly involve the opening book. I only use the opening book data to integrate into my study, for example, to simulate how often players in a specific rating category (I selected 2200-2500 in blitz games) choose a certain move at a branching point.
The key for me is functionality. How can I ensure that my prc:33 is recognized and executed as a command by Lichess Tools?
So the "command" is just something that LT reads from a comment. I have no other way to store stuff related to moves and chapters on the server (that are acceptable to Liichess). If I hide it or not that's a decision (and an effort) that I make.
prc:33 is recognized and used by:
- Ctrl-right-arrow - random next move
- Interactive mode - play all variations
- the mobile button to play a random move from the PGN
There is no way to "ensure" it works :D because I haven't tested it in a long while. but it SHOULD be working.
Yes, I understand that you can only store commands in comments on the server. So, am I wrong in thinking that something can either be executed as a command or just displayed, but not both? If I understood you correctly, in the interactive mode (with the 'Play all variations' option enabled), prc:33 should be displayed but also executed as the selection probability. I’ll test that right away.
However, what would still be annoying is that you have to click every time to dismiss the comment and then continue playing from there. So, at every branching point, you would have to click. Isn’t there a way to execute this command in the comment without it being displayed as a comment?
As an alternative, the only thing that comes to mind is setting the selection probability based on the Lichess opening book for a predefined cohort or skill level, but that would require the mechanism to look up the selection probabilities from the Lichess database each time and then imitate them... which probably wouldn’t work smoothly either.
So, the simplest solution might be if a comment could be executed as a command without being displayed. Maybe there’s some sort of 'secret language' in the Lichess API to implement commands that wouldn’t be visible later? I’m not a programmer, I’m an analyst trying to solve problems logically. I’m aware that this is a complex issue.
Command withdrawn, it works! :D :D
The error regarding having to click when the comment popped up was on my end! I accidentally entered the prc:33 in the half-move before the branching point, instead of in the exact move of the branch. The fact that the comment is briefly visible doesn’t bother me at all, it doesn’t interfere with the training flow, as you can simply continue playing, and the comment disappears. Plus, it provides interesting information even during training!
I’ve now extensively tested the functionality of the selection probability prc:N. I assigned prc:5 to eight variant moves at a branching point and prc:50 to a ninth variant move, and then tested how often one of the eight was selected versus the ninth move. Mathematically, you’d expect that since the eight moves combine for 40% and the ninth move has 50% (and the mechanism scales up to 100%), the ninth move would be selected in about 4/9 cases (roughly 44.44%). In my test of 27 runs, the ninth move was selected 10 times, or about 37% of the time, which is quite close to the expected 44%. So the mechanism works! :D
Thank you so much for this fantastic mechanism, I’m really happy and excited to fully implement it in my study! Thanks for helping me! Your prc:N mechanism works!
As they say, sometimes the problem with a computer error is the person sitting in front of the computer ;).
"Sometimes"?! Phah!
Anyway, have you tried the Fast Interaction setting? It might make your experience even better if efficiency is what you are looking for.
I admit it: Most of the time, when there's a computer error, the problem is sitting in front of the computer ^^. It's already very efficient, I haven't tried Fast Interaction yet, but this is already fast enough for me, I'm totally happy with the mechanism! :D
@Endspiellehrer said in #2:
> Additionally, is it possible to set selection probabilities for chapter selection in a study, similar to how it works for moves in a variation?
yes, in v2.3.144: https://siderite.dev/blog/lichess-tools---user-manual/#chapterNavigation
This topic has been archived and can no longer be replied to.