I see that a toggle is a good middle ground for this.
You either:
1-Keep learning with normal learning steps as before
2-Keep learning with normal learning steps and toggle on further automatic scheduling
3-Leave learning steps empty and only learn with automatic scheduling
Regarding the complexity of the UI, the only change would be the addition of a toggle like a toggle for FSRS. I see that if more advanced features were to come to Anki, it is only natural that more Anki panel features will be added eventually.
Inconvenience of <1d scheduling
The “Next Day starts at” setting is already there and it stops Reviews from being scheduled past a particular time.
Or as I suggested above, a toggle would be a good middle ground
You either:
1-Keep learning with normal learning steps as before
2-Keep learning with normal learning steps and toggle on further automatic scheduling
3-Leave learning steps empty and only learn with automatic scheduling
We want to make FSRS easier to use, not harder. When I say “complexity” I don’t just mean “the number of windows/options/toggles” (although that too), I mean “How much mental effort the user needs to put into setting FSRS (or Anki itself) up?”. Every new thingy increases the amount of mental effort required to set everything up, and FSRS is already in the “some turbo complex thing only for advanced users” category and a lot of people are afraid of using it because they don’t understand how.
And with either algorithm people are agonizing over the exact values of learning steps, and every single Anki guru is suggesting different steps. We need to decrease the amount of decisions that the user has to make.
To me it sounds pretty complex to mix intraday intervals from this setting with those from FSRS. For example: what to do if you set it to 15m 6h and FSRS adds a step of 8h? You’ll end up with 3 steps, unless FSRS uses some sort of fuzz factor and detects that the 6h interval is already close enough. Why not just use your own if you don’t trust FSRS, or use FSRS if you do?
I’m not opposed to a switch though, but in my opinion it should disable the input field entirely.
Yes, I agree that too many options would be agonizing. However, I have listed only 3 scenarios of user cases. I don’t think a total of 3 different user case scenarios make a big difference than to 2, especially since these 3 add to greater flexibility during learning whilst not being gratuitous as suboptimal deck sorting orders.
These 3 scenarios are not listed in a list like the sorting orders. I think I find it intuitive.
*you have option A
*or you have options A and B
*or you have option B
There is no C. We are dealing with the same options A and B.
In the grand scheme of things, I think flexibility>complexity.
Besides, the user will find themselves only rarely (at least I think so) tumbling around with the learning steps. In any case, with this feature, there will be a rise in complexity, regardless whether the toggle is added or not.
It is a new feature. A new tool. New instructions.
Yes, feasibility is a very different question to the topic of this discussion though. I don’t know if it is even possible to combine fixed Learning Steps with <1d automatic Learning Steps.
At this current stage, the <1d automatic Learning Step scheduler is not even done yet, let alone be compatible with fixed Learning Steps.
Why not just use your own if you don’t trust FSRS, or use FSRS if you do?
Well in my case for example, I would like to have my again interval be fixed at a certain amount of time whilst also keep using the automatic <1d Learning Step scheduler for the rest of the day.
What if you want to fix some steps, like the again interval
Well in my case for example, I would like to have my again interval be fixed at a certain amount of time whilst also keep using the automatic <1d Learning Step scheduler for the rest of the day.
I get it, but as I said, we should make it so that the user has to make as few decisions as possible to make FSRS easier to use. The whole point of removing learning steps and letting FSRS take over is not so much efficiency as decreasing the decision fatigue/cognitive burden/mental burden/whatever you wanna call it.
What are major decision fatigues (other than reviewing sorting orders) that Anki users have to deal with I see this as a net positive if it indeed turns out to work somehow for the reasons mentioned above, at least for users who would like to fix some steps whilst also wanting to benefit from FSRS scheduling.
Yes. When we have 0 steps, the card becomes a review for any response. There is nothing new here.
The review card is not capable of having intervals of less than one day and as far as I know there are no changes planned here.
