On bullet-points and connection-heavy writing style
The top 20 hub notes in this vault are 80% bullets with 26.1 average wikilinks. 15 of those 20 are what structural analysis classified as “index” hubs—essentially connection switchboards. The medium-tier notes are 40% bullets with 6.9 wikilinks. The pattern: the more connected a note becomes, the more bullet-heavy it gets.
The bullet-heavy notes and essays read more like idea collections waiting to become arguments. Bullets are great for connecting, and show how the argument could’ve been otherwise; prose is where you test whether the connection actually holds up under sustained reasoning—a paragraph forces you to defend why this connection and not another—it’s a contingency realized [2].
Insight is constructor-theoretic
Is bullet-points and connection-heavy writing style good for insight? I think so, but the insight lives in a specific place. Naval said: “Enlightenment is in between your thoughts, in understanding why you are having such thoughts.” The insight isn’t in the prose of any single note—it’s implicit in the connections between seemingly dissimilar thoughts, and enlightenment is about understanding the decision which led to that connection. It’s similar to mechanistic interpretability research (e.g., Arshavir Blackwell)—just that its subject is not LLMs.
When you connect Gresham’s Law to AI or Elden Ring to Karl Popper [1], such links constitute a glimpse of an insight—but insight, and creativity, is in the ability to connect. Expertise is implicit.
[1]—The vault’s most interesting finding isn’t any single cluster—it’s the cross-cluster bridges, e.g., notes that appear across 3+ thematic groupings spanning epistemology, Austrian economics, and language/mind.
[2]—Claude: “If bullets preserve optionality and prose tests commitment, and your most connected notes are 80% bullets—does that mean your most important notes are also your least tested? That’s either a feature (they’re generative hubs, not meant to be arguments) or a problem (you’re accumulating connections without ever cashing them out).” [3]
[3] What matters is the truth, and not how many possibilities I can explore. I better write more essays (and now I understand Paul Graham’s disdain for bullet-points heavy writing).
Related: