1 · Map
Place the topic. See where it lives, what feeds it, what it feeds.
Learning design, made visible
Map, study, lab, retrieve, bridge. Repeat per topic. That is the whole method.
No magic, no megacourses. The site’s job is to make the high-leverage habits easier: study in sequence, manipulate models, retrieve from memory, revisit weak spots, and link ideas across fields.
★ the five moves
None of these replace the others. Doing one is fine; doing all five over time is when ideas start to stick.
Place the topic. See where it lives, what feeds it, what it feeds.
Short ordered pages. One idea each. No overload.
Manipulate a working model. Make it react. Find the limits.
Pull it back from memory. Flashcards, quiz, critical prompt.
Follow one edge. Same pattern, new domain. That is transfer.
⌁ touch the algorithm
Rate a card — see how the next review date moves. This is the same idea running underneath the review page.
How well did you recall it?
The taller the bar, the longer the gap before the next review. Good ratings stretch it; “again” resets it.
SM-2 style spacing · 7 most recent reviews
✦ why these moves
Pulling information back is what builds the durable trace — reading the same page twice mostly builds familiarity.
Reviewing right before forgetting is more efficient than reviewing while it’s still fresh. That’s why the demo above grows the interval.
Switching between related ideas forces discrimination, which makes each one cleaner. The bridge step is for this.
Sliding a parameter and watching a model break teaches the limits of the model. Every topic gets a lab for this reason.
Short, ordered pages keep cognitive load low. Each lesson page introduces one move, not five.
Entropy in thermo and entropy in information are the same shape. Bridge topics make that explicit instead of leaving it to luck.