By dkl9, written 2024-228, revised 2024-228 (0 revisions)
For over two years, I have fully memorised much of what I learn, with the help of my spaced repetition program Memoire, a substitute for Anki. Mostly, I prepare my Memoire cards in secret entries of Arbitrary Notes.
I give here some of my better reusable note formats, those most suitable to be shared, those which avoid issue 4 in two senses:
In a document like this, issue 1 is moot, and I can filter out truly secret notes for which issue 2 applies.
Whichever of these templates you use, remember to properly understand a fact or concept before throwing it in as a card, and feel free to make changes as you see fit.
In general:
words XY: {WORD} (WZ) = {TRANSLATION} (XY)
Works well for vocabulary of foreign language or shorthand.
Examples:
words fr: {glass} (en) = {le verre} (fr)
words ase: {science} (en) = {two open-As, palms out, neutral space, S-circled past each other} (ase)
words ds: {year} (en) = {an} (ds)
I use ISO 639 standard abbrevations, when defined, to specify languages XY and WZ. This format is less effective for logographic languages such as Chinese.
In general:
quotes: ;;SPEAKER on TOPIC: ;;CONTENT_A ;;CONTENT_B ;;CONTENT_C Attribution: ;;
Works well for poetry and long aphorisms.
Examples:
quotes: ;;B. Pascal on lazy verbosity: ;;I made this one longer only sith ;;I have not had the leisure to make it shorter. Attribution: ;;
quotes: ;;U. Neisser on spaced repetition: ;;You can get a good deal from rehearsal / ;;If it just has the proper dispersal. / ;;You would just be an ass / To do it en masse: / ;;Your remembering would turn out much worsal. Attribution: ;;
quotes: ;;Sun Tzu on employing those of various skills: ;;[Good leaders] employ the intelligent, the brave, ;;the greedy, and the foolish. The intelligent are glad to establish their merit, ;;the brave like to act out their ambitions, the greedy welcome an opportunity ;;to pursue profit, and the foolish do not care if they die. Attribution: ;;
When reviewing, you'll see a fragment of the quote and be asked for the next fragment. For the last fragment, you're asked for who said the whole quote, in what context.
In general:
chars zh: ;;char, meaning ;;definition (en), pinyin ;;pronunciation, hanzi ;;
Works well for associating shape to pronunciation to meaning of Chinese or other logographic characters.
Examples:
chars zh: ;;块, meaning ;;lump/dollar (en), pinyin ;;kuài ("lump"), hanzi ;;
chars zh: ;;教, meaning ;;to teach/class (en), pinyin ;;jiāo, hanzi ;;
chars zh: ;;密, meaning ;;secret/thick/dense (en), pinyin ;;mì, hanzi ;;
These prompt for (translated) meaning given the character, pronunciation given meaning, and character given pronunciation. If multiple characters are pronounced the same, but mean different things, disambiguate them by including a hint at the meaning with the pronunciation, as I did for 块.
In general:
words TOPIC: {WORD} = {DEFINITION}
Works well for a great many topics, especially chemistry, culture, linguistics, maths, medicine, mathematics.
Examples:
words med: {reduction (of a bone)} = {movement of bone or fragments to proper position}
phrases society: {excise tax} = {sales-like tax added to socially-unwanted goods to discourage quantity supplied}
words: {histrionic} = {acting-related/dramatic/attention-seeking}
The topic marker disambiguates words and aids recall; it should be short and systematic, such as "art" for art, "chem" for chemistry, "med" for medicine, "psych" for psychology, "society" for contemporary culture. I replace the label "words" with "phrases", "notation", or "synonyms" where appropriate.
In general:
TOPIC: {THEOREM_NAME} asserts {ASSERTION} with variables {DEFINITIONS} where {CONDITION}
Works well for maths, physics, and psychology.
Examples:
complex analysis: {Cauchy-Goursat theorem} asserts {int[C, dz * f(z)] = 0} where {C bounds a simply-connected region where f is analytic}
physics: {time-independent Schrodinger equation} asserts {E * Ψ = -ℏ^2 / (2*m) * ∇^2 Ψ + U * Ψ} with variables {E = total energy, Ψ(x, t) = wave function, ∇^2 = spatial Laplacian, U = potential energy} where {the system is modelled as a quantum wave}
psych: {Purkinje effect} asserts {we perceive luminance more sensitively for bluer colours} where {it's dark}
The "with variables ..." clause is only useful when the assertion is expressed symbolically with obscure variable names. For complicated assertions or conditions, it helps to split them into smaller segments.
In general:
TOPIC: what's wrong here? STUPID_ASSERTION {EXPLANATION_OF_PROBLEM_AND_CORRECTION}
Works well for grammar, maths, programming, and science.
Examples:
grammar zh: what's wrong here? "路有几个人" {"on the road" translates as "路上有...", not just "路有"}
python: what's wrong here? `if not l:` to check for blank line under `fd.readlines()` {line-strings include final newline, so try `if not l.strip():`}
chem: what's wrong here? SN2 m-R-bromomethylcyclohexane into m-R-cyanomethylcyclohexane {SN2 inverts stereochemistry, so this produces m-S-cyanomethylcyclohexane}
Make the "stupid assertion" a mistake you've actually made. The explanation you memorise should give the right answer to the original problem, with reasoning, so it can generalise.
In general:
TOPIC: ALPHA relates to BETA via {RELATION}
Works well for maths, philosophy, and science.
Examples:
graph theory: edge count relates to planarity via {planar graphs have at most 3 * V - 6 edges}
philosophy: "predicate" in "predicate logic" relates to "predicate" in "subject and predicate" in that {predicate's subject (language view) is argument (logic view)}
thermo: "heat goes from hot to cold" entropy relates to "ΔS = Q/T" entropy in that {small local heat transfers make more arithmetic entropy (ΔS > 0) when the positive heat (recipient of heat) is the cold object}
Most of my uses either describe why two things are kinda equivalent, or suggest what theorem to apply when studying two objects together.
In general:
SYSTEM: GOAL_RESULT with {COMMAND}
Works well for computer application techniques and programming.
Examples:
spreadsheets: {convert value to string} with {TEXT(x, fmt)}
git: show summary of which files changed when with {git log --stat}
python: iterate over key-value pairs of a dict with {for (k, v) in d.items()}
If, as with TEXT, it's hard to figure out what a command does from the command itself, put braces around the goal to memorise it separately.
In general:
TOPIC: ;;TERM, made of ;;COMPOSITION, found ;;CONTEXT, it's called ;;
Works well for almost anything, especially hard science.
Examples:
plants: ;;persimmon, made of ;;orange spheroid fruit, ~5 cm, found ;;from ebony trees, esp. in East Asia, it's called ;;
physics: ;;Compton wavelength, made of ;;h / (m * c) (Planck constant over particle mass times speed of light), found ;;as the scale below which particles stop making sense, it's called ;;
geology: ;;gypsum, made of ;;calcium sulfate dihydrate, found ;;as fertiliser, drywall, and chalk additive, it's called ;;
These prompt for substance given name, significance given substance, and name given significance.
In general:
TOPIC: PROMPT_HAS_COMPONENTS (mnemonic) ;;ACRONYM, i.e. ;;ALPHA, BETA, GAMMA, DELTA, EPSILON, ZETA, ETA, THETA
Works well for lists.
Examples:
psych: main contributors to suicidal intent are (mnemonic) ;;BAIT, i.e. ;;"burden" feeling, agony, impractical goals, "trapped" feeling
economics: demand is affected by (mnemonic) ;;STIPE, i.e. ;;substitutes/complements, tastes, income, population, expectations
physics: electrons in orbitals are distinguished by quantum numbers (mnemonic) ;;PAM'S, i.e. ;;principal, angular momentum, magnetic, spin
If the order is arbitrary, re-order the items in the list to make a more pronounceable (hence memorable) acronym.