Writing & Thinking

Writing as a thinking tool, the xettel method, publishing workflows, and the practice of public thinking.

Writing practice

Everything Should Be Written — write everything down; ideas grow from other written ideas
Why I Write Posts Myself and Don T Use AI Generation — writing preserves identity and voice
Better Written Than Lost — imperfect and published beats perfect and lost
Done Today Beats Perfect Never — streamlined workflow for shipping posts
Hello World — introducing the blog and its purpose

Thinkbox: an LLM Knowledge Base with a Public Surface — how xettel, LLM wiki, and the public website work together

2037821667188387946 — Xettel = X + Zettelkasten: one card = one tweet = one atomic thought
2037821910034313401 — connections between cards are separate cards, close to Luhmann’s original
2037827169259585982 — xettel is for yourself first; it becomes a mirror of your mind
2037827760534798439 — writing regularly self-corrects what you consume
2037827907536720083 — don’t fear being public; obscurity is worse than exposure
2041783535233814738 — thoughts in the box are atomic, but connections give them a world
2041783732001239420 — a walk through connected cards produces a new view, not new thoughts
2041785087185055803 — a collection of aphorisms is not a thinking tool; connections and counter-arguments matter
2041786419665109251 — better to save too many thoughts than to lose one important one

2041770909640618337 — comfort of being unchallenged is a widening gap between self-model and reality
2041771064766919075 — the earlier you learn you’re wrong, the less it costs
2041774198138532061 — learning in public is not sharing results; it’s exposing your reasoning to challenge
2041785187806408849 — what matters is becoming smarter and closer to truth; method is secondary
2041786177263694116 — becoming smarter in private is safe but unchecked; public invites challenge

2041782058150961294 — disagreement is the most valuable thing in science, but only when argued
2041782260282880403 — science without bureaucracy is argumentation, evidence, and changing your mind
2041782482945888763 — anyone can do science; you need a question, evidence, and willingness to be wrong

2041161600103801063 — problems outlast tools; problem structures are the durable object of study
2041162396941832642 — selection criterion: not “is this interesting?” but “what problem does this solve?”
2041407270291153388 — problem formulation is irreducibly human; responsibility requires skin in the game
2041407424922562636 — solution methods are the most volatile layer; replace them perpetually
2041407557726867556 — a problem journal is both a reflective tool and a tradeable economic asset
2041783918085714125 — learning is not collecting information; it’s integrating it into your worldview
2041786300714606600 — learning is two simultaneous acts: extracting and integrating
2041784038571307023 — explaining your position to an opponent clarifies it faster than explaining to yourself
2041784174298984697 — responding before formulating teaches you what you actually believe
2041784653976367268 — a complex question paralyzes; a simple one creates entry point and momentum
2041784803612356890 — once momentum is there, depth follows naturally
2041784925687566493 — a fast read is not a replacement for a deep one; it’s triage
2041771208736465155 — you can read a text or read yourself reading it; second produces self-knowledge

2041784373276766651 — a raw record is worth more than a polished post; from raw you can cut many outputs