Reflections on five years of conceptual Anki
It’s been about 5 years now since I’ve been using Anki in a pretty determined, conceptual way (not the kind of casual, vocab-centric approach I was doing back in like 2012–2014). What has happened in those years? I’d summarize my timeline as something like:
- mid 2018 – mid 2019: focus was on getting back in the swing of using Anki at all, and using it to learn math. Learning math was what I was doing pretty much all day during this period, and I was determined to not forget the math that I had learned. I wrote some rambles here of the stuff i was thinking during this period. Michael Nielsen’s essay was a big influence at the start of this period.
- late 2019 – mid 2021: by the beginning of this period, I was getting quite curious about the internal workings of anki and published my python implementation as well as this gist about how i messed up the ease factor on my problem cards. Then came spaced proof review, and spaced inbox development. Started using SuperMemo in April/May 2021, to get a taste of how other SRS work. Most of the spaced repetition articles on my wiki were written during this period. I think this was a very productive period where I was theorizing a lot and coming up with ideas that would become core parts of my “memory practice”.
- late 2021 – mid-2022: hiatus. was busy with travel, health stuff. wasn’t really even doing my anki reviews, though i still kept up on quantum country and orbit.
- mid-2022 – March 2023: continuation of the above, but i was also getting more disillusioned with spaced repetition/memory practice in general, instead getting more interested in my very old interest of math exposition (math explanations), and wrote two posts on LessWrong as well as progress on memory systems is bottlenecked on explanation science. There was a small burst of further spaced inbox development in early 2023 as i was trying to debug why i had become so averse to using it.
- March 2023 – mid 2023: focus has been getting back to doing my daily reviews, and also more deliberately reviewing my cards (and others’ cards). Trying to solve the problem of “chill anki” (how can one incorporate memory practice into one’s life given the reality of chronic illness?). wrote things like long-run Anki review load as part of this thought process (the page was trying to answer the question of “how chill can anki be?”).
In general, i think my first couple years were the most productive, and by the end of my third year (so by mid 2021), i think i had generated most of the core insights that i use in my spaced repetition practice. i’ve been just plateauing in terms of how good i am at making cards for the last 2 years or so. so anyway, given all of that, what do i want to accomplish in the coming years?
- i think i want to continue to explore the idea of explanation science. this post (also linked above) gives a roadmap. so in other words, i expect to further move away from explicit memory practice.
- the idea of chill anki is still interesting to me: i put a lot of effort into making my cards, and i want to make sure i keep reviewing them somehow. but i expect most of the thinking here will be about habits in general – how do i maintain a long-running habit under the stress of chronic illness? – than about memory stuff specifically. some of it will be specific to anki though; for example, i’ve been moving away from math proof cards/spaced proof review because that is just very intense and not really a sustainable habit for me in my present state of illness. is there some way to take the good parts of this practice and “chill-ify” it somehow? that’s something i’d like to figure out.
- as for actual memory practice, i think the main thing i want to focus on is to be more methodical. I’ve come up with techniques and internalized them sort of haphazardly over the years. i don’t have a checklist or anything that i can run through when writing or reviewing prompts. i was reading this post some time ago and quite liked the idea of explicitly carving out some time to do reviews of anki cards. what will “being more methodical” consist of? i’m not sure yet.