Cross-posted from timot.cool

Spaced repetition (SR) is still an early field of collective experimentation. People have been coming up with many ideas on what to use SR for: trivia like the capitals of the world, foreign vocabulary, their domain of expertise... What I almost never see discussed is the use of SR for content like blog posts and non-fiction books. We're reading them to induce long-term change in our behavior or thinking capabilities, yet these sources of knowledge seemingly don't trigger the SR reflex as much or at all.

Why? Because blog posts and books are mostly not about raw facts, which are the easiest way to get started with SR. Yet I've personally found a lot of value in using SR for the ideas, arguments and concepts I find in my readings. A few (overlapping) subcategories with examples from my collection:

1. Concepts (examples). Some writings give a crystal-clear name to some idea that was complex or vague, and the idea now becomes a tool in a cognitive toolkit. These concepts are essential for thinking. You'll think better if you have chunked vague ideas into things with names, and these names are even more essential if you want to discuss things with others, or share these vague ideas. In section 9 of his Nonfiction Writing Advice, Scott Alexander writes that some of the more important things a blog can do is to put names on such vague ideas. Scott called these "concept handles", after previously calling them "crystallized patterns". He says:

If you figure out something interesting and very briefly cram it into somebody else’s head, don’t waste that! Give it a nice concept-handle so that they’ll remember it and be able to use it to solve other problems!

And I say: don't waste that! Put the concept-handle into your SR collection to avoid the risk of forgetting about it forever, and so you can easily share it with friends or refer back to the original post later on!

These concepts may come with examples (or you may make up your own), which are often also worth remembering.

2. Arguments, ideas (examples). After all, why should we forget them?

3. Summaries (examples). I like these a lot, and they're often appropriate for blog posts. This is the kind of SR card whose formulation (writing a personal summary) requires putting in some effort, but is very valuable.

A few thoughts on the above usage of SR:

• people vary wildly in their experience with SR. So experimentation, which is in this case high-upside and low-downside, is the thing to do.
• this post is about ideas rather than behavior: I've personally found SR much more useful to enhance thinking than to induce behavioral change. This seems quite natural. Things like decision-making checklists or note-taking still have their place.
• it seems particularly important to formulate these SR cards oneself. If you have a look at my cards, you'll probably find that you would have formulated them very differently (and these aren't my most personal). This is linked to rules 13 and 14 of Wozniak's 20 rules for formulating knowledge: personal cards are generally better, and these kinds of cards are more customizable than what's the capital city of Guatemala, Guatemala City. In other words, these cards carry hidden context only accessible to their author.
• I think there's still value in reading other people's cards. Precisely because our approaches to SR vary wildly, we may find inspiration in how others tackle the problem of remembering what they read.
• perhaps another reason why blog posts and books don't trigger the SR reflex is that they superficially appear to contradict the minimum information principle (rule 4 of the 20 rules), in that the cards are longer than usual "raw fact" cards, and it's harder to tell if you got them right when reviewing. I think the resolution of this is mostly a matter of personal taste. When reviewing, I personally verify that I've got the answer of the card in mind, rather than aiming for precision: I don't always verbalize words, and when I do, I formulate something that's usually simpler than the actual card. So even though the formulation of my cards doesn't always conform to the minimum information principle, my reviewing process does.
• this kind of practice is a virtuoso skill (to use Michael Nielsen's words). It gets better with experience.
• of course you know what a prisoner's dilemma is at this point. But not all concepts are reviewed naturally so often. This is the whole point of SR. It's sometimes unnecessary, but in these cases the formulation of the card was probably valuable when it was performed, and the reviews can be very fast. I do have cards that have turned out to be very easy. But when reviewing, I see the question, think something like "I know the answer" and immediately hit "easy" on Anki which catapults the card exponentially far into the future. I'm sure some of these cards represent less than 30 seconds reviewing time over a lifetime.

Thanks to JS Denain and Léo Grinsztajn for reading drafts of this.