Spaced repetition is super valuable, but popular spaced repetition systems today fall short for many people unnecessarily.
Here are 6 ways we can improve spaced repetition for the next generation:
9
10
61
Conversation
1 Card creation: Zero friction card creation
2 Card creation: Good cards by default
3 Card maintenance: Eject bad cards quickly
4 Schedule flexibility: Allow people to study more
5 Schedule flexibility: Allow people to study less
6 Review friction: Bring the reviews to the users
2
1
21
Details here:
davidbieber.com/snippets/2021-
but I'll summarize in this short thread to save you some time:
1
1
7
* (1,2): Creating good cards should be as easy as a Google search.
I made a prototype with Browserflow that I've fallen in love with: davidbieber.com/snippets/2021-, and we can push this even further with ML
2
6
* (3) The UX of SR systems should lead users to drop bad cards early and often; let's make keeping bad cards hard
2
5
* (4,5): SR systems should encourage studying when you're most motivated; they shouldn't rely on you studying even on a vacation, or discourage extra studying when you complete your day's review. This requires changes to both algorithm and UX.
4
1
4
* (6) An extra app like Anki is extra friction; brings SR to your email inbox, but you still need to click through to review. Great progress, but we can do even better.
1
3
I want to share these ideas with folks thinking about spaced repetition. So, here's some unsolicited @-ing; I hope you don't mind:
@hermannebbinghaus
3
9
Broadly agreed! Though alas, I believe #1 and #2 are deep research projects: my experiments with cloze-driven extracts like the ones you describe have been pretty unsuccessful in the long-term.
4
1
I think there's a tradeoff between amount of understanding produced per card and time taken to formulate each card. You might get less understanding per cloze but they are more efficient to produce - as a networked mass they might produce more understanding for less cost.
1
Yes, that's right, and that's been Wozniak's argument historically. With low-cost iterative editing, it might be true! It hasn't been my experience personally, though.
Perhaps the card creation problem is really a problem of the implicit “knowledge map” those cards are trying to represent. Without making that knowledge map explicit it’s hard to know what card structures are efficient translations, and it might depend on the subject / user.
1
If this is true, then the spaced repetition problem might be broken down into three problems.
(1) creating high quality knowledge graphs
(2) translating those knowledge graphs into cards
(3) building quality interfaces for users
1
Show replies