But the user might still check the point of the first week when the timeframe is “All time”. And “First year” also doesn’t mean the elapsed days since the first learning reach one year.
How about this: if the total elapsed time is >=365 days AND the currently selected timeframe is either “Last year” or “All time”, don’t display time, only dates?
I will make it adaptive to the width of the window.
Hopefully it will help this user as well: Anki 24.10 Beta - #4 by cjdduarte
I’m fixing these problem in:
Feedback and suggestion is welcome.
By the way, it’s time to go to bed for me now.
Bug: “double optimization”
Double Optimization - GIF on Imgur
It shouldn’t be possible to run optimization twice without getting “FSRS parameters currently appear to be optimal”.
Bug: optimization gets to 100% and then I get this error:
Anki 24.10 (201f1394) (ao)
Python 3.9.18 Qt 6.6.2 PyQt 6.6.1
Platform: Windows-10-10.0.19045
Traceback (most recent call last):
File "aqt.deckbrowser", line 314, in <lambda>
File "aqt.deckbrowser", line 340, in _options
File "aqt.deckoptions", line 150, in display_options_for_deck_id
File "anki.decks", line 247, in get
File "anki.decks", line 156, in get_legacy
File "anki._backend_generated", line 446, in get_deck_legacy
File "anki._backend", line 161, in _run_command
pyo3_runtime.PanicException: called `Result::unwrap()` on an `Err` value: PoisonError { .. }
Bug: last 2 parameters are always the same after optimization, for any preset. They are set to their default values.
There seem to be some problems with type-in-the-answer fields
You currently see this behavior in these two cases, right?:
Answer in your screenshot).(IOW, not when you input it wrongly.)
Both are separate code paths, allowing this to be resolved in two ways:
The old behavior (“strip tags out of type:: entirely”)…
The second case would effectively mean that type:: becomes an offer: If you skip it, Anki would now treat the field as if included ordinarily.
I’m partial to the second option, but it would introduce breakage.
Got the following error after optimizing FSRS parameters:
Anki 24.10 (201f1394) (ao)
Python 3.9.18 Qt 6.6.2 PyQt 6.6.1
Platform: Windows-10-10.0.19045
Traceback (most recent call last):
File "aqt.deckbrowser", line 314, in <lambda>
File "aqt.deckbrowser", line 340, in _options
File "aqt.deckoptions", line 150, in display_options_for_deck_id
File "anki.decks", line 247, in get
File "anki.decks", line 156, in get_legacy
File "anki._backend_generated", line 446, in get_deck_legacy
File "anki._backend", line 161, in _run_command
pyo3_runtime.PanicException: called `Result::unwrap()` on an `Err` value: PoisonError { .. }
Oops. I forget to update the filter in Anki, which removed the same-day reviews.
It will be fixed in:
I believe one thing that we should try to do here is making Anki more resilient to future changes in scheduling code.
If years later, we ever add additional parameters, upon syncing we will get this in the old client (or probably some other error):
Can we make it such that addition of extra parameters to the end is ignored by current FSRS? To be clear, I’m proposing future-proofing now so that we don’t need @expertium 's time machine (The Only 100% Foolproof Strategy Is Inventing A Time Machine) or need to do substantial changes in syncing.
Perhaps, there is a way?
Hi,
Just a little inconvenience, when clicking on a tag at the bottom of the add cards window it disappears.
When doing that it should switch to editing and I should be able to modify the tag.
To make the tag reappear I have to click to the right of where the tag is supposed to be, and to then modify the tag I have to have the cursor at the right of the tag and press backspace.
Right! If I type an incorrect answer, Anki behaves as usual.
I think I would be OK with both options. But I wonder if the second behavior might be a bit confusing, as it would make the type field behave in an inconsistent way based on the presence or absence of user input (e.g., some users might ask “why are my images not showing when I type an answer?”).
Maybe it would be possible to accomodate both use cases by having the second behavior as default, but allow for the tags to be skipped even when there is no user input by adding the text prefix to the type field, for example: {{type:text:FieldName}}.
We can make that in the current version, just not in the past versions. So we can add some future-proofing in case FSRS-6 ever becomes a thing, but we can’t help people who are currently using FSRS-4.5.
Anki Version: 24.10 beta-1
Hi,
The Reveal time is overlapping the due cards at the bottom. Seemed like a bug, so reporting. It could just be an add-on complatibility issue.
Yes, I was suggesting futute proofing
now. But I wish we had the time machine.
Well, Dae didn’t really do anything when I made that post, so you’d have to use all of your persuasion to make him implement it
@dae Is the iOS AnkiMobile 24.10 beta on testflight able to have FSRS-5 parameters optimized? Or are we still waiting for a newer build on that feature?
24.10 build 1:
If you have 19 parameters instead of 17 it has FSRS-5, otherwise probably no.
19 parameters after optimization, since the optimizer doesn’t automatically insert two new parameters into the field.
Ah, you said there’d be two empty params (0).
Well, turns out the beta doesn’t work that way, idk why.
