The irreducible epistemic atoms underlying the curriculum. 4,828 atoms across 8 types and 2 molecules
Define the minimum viable output by identifying the recipient, the action they should take, and the minimum content needed to enable that action—then produce only to that threshold before shipping.
Before launching a public output cadence, pre-produce three to five buffer outputs to absorb inevitable disruptions without breaking the chain.
Complete all batch preparation (gather materials, load contexts, prepare templates) before the production session begins, so that session time is pure execution without information-gathering pauses.
Within a batch session, start with the easiest or most familiar output to build momentum, tackle demanding outputs at peak cognitive capacity, and save formulaic outputs for the end when pattern-matching remains strong but creative energy wanes.
Before beginning any significant output, define its distribution plan by identifying specific audiences, channels, formats, and timing rather than treating distribution as a post-production task.
Move outputs sequentially through pipeline stages (Draft → Review → Polish → Deliver) without skipping stages or oscillating backward except through explicit, documented regression decisions.
At each pipeline stage transition, check explicit gate criteria (3-5 yes/no questions) before advancing the output, and if any answer is no, keep the output in its current stage until deficiencies are addressed.
During draft stage, generate content and follow templates without editing or polishing; during review stage, evaluate structure without wordsmithing; during polish stage, refine language without restructuring—never mix activities from different stages in a single work session.
Before every work session, consult your pipeline board and work on the most downstream stage that contains items (Deliver before Polish, Polish before Review, Review before Draft) to pull from bottlenecks rather than accumulating work-in-progress.
Set work-in-progress limits for each pipeline stage (typical solo creator: Draft—3, Review—2, Polish—2) and do not start new items in a stage until existing items advance, to prevent attention fragmentation and increase throughput.
For high-stakes outputs with long lifecycles, implement full versioning with named versions, changelogs, and archived copies; for medium-stakes outputs, use lightweight date-stamped versioning; for low-stakes outputs, skip active versioning entirely to match overhead to value.
Before an output advances from one pipeline stage to the next (draft to review, review to polish), snapshot the current state as a version to create recoverable checkpoints throughout the production process.
Use push distribution (direct delivery to inbox/channel) for known audiences with time-sensitive needs, and pull distribution (searchable repositories) for unknown future audiences seeking information over time—deploy both for significant outputs, not one or the other.
For every output, answer four audience-mapping questions before finalizing: (1) Who specifically needs this? (2) What do they need from it? (3) Where do they consume information? (4) When do they need it?—then format and time distribution accordingly.
When translating an output to a new format, rewrite the content in the language, tone, structure, and emphasis that format demands—do not copy verbatim across formats, because each medium requires different cognitive packaging to serve its audience effectively.
Cascade derivative format releases over days or weeks after the pillar ships (rather than publishing all simultaneously) to extend the life of the original research, prevent audience fatigue, and create multiple touchpoints across time.
Build a personal output scorecard tracking four dimensions (reach, resonance, downstream action, personal growth) rather than any single metric, reviewing monthly to identify which output types score highest on the dimension that matters most to your goals.
Before measuring anything, define your personal value metric (what 'value' means for your specific outputs) as the lagging indicator that ultimately matters, then select leading indicators that predict it—never measure without first answering 'value for what purpose.'
When a heavily-viewed output generates zero downstream action while a quiet output generates high-value outcomes, increase production of the quiet output's type—optimize for measured impact, not vanity metrics, even when impact metrics are smaller and less emotionally satisfying.
In post-action reviews of outputs, terminate causal reasoning at the process and structure level, never at the personal adequacy or character level, to extract systematic improvements rather than identity-based judgments.
During output reviews, answer exactly four questions in sequence: (1) What did I produce and what results did it generate? (2) What patterns exist across successful vs unsuccessful outputs? (3) What assumption should I question? (4) What one specific thing will I change next cycle?
Close every output review by integrating the committed action from question 4 directly into your next production cycle's plan before ending the session, ensuring review insights become operational changes rather than intellectual observations.
Store every archived output with exactly five metadata fields: descriptive title containing search keywords, completion date, output type from your controlled vocabulary, project/context, and 2-5 searchable tags, establishing this as the minimum metadata standard before any file enters the archive.
Implement a two-minute archiving workflow triggered by 'I shipped an output,' consisting of filing the output in your single archive location with required metadata, treating archiving as the closing ritual of every delivery rather than a deferred task.