The full catalog of field notes and long-form essays. Filter by category, sort by date or title, or search by keyword and tag.
The discipline of withholding an answer you hold, the four conditions for handing it over anyway, and why the model belongs after the commitment rather than before it.
Four instruments a small pilot can actually run — unprompted falsifier production, the confident-wrong-authority task, authorship equality, and transfer — each with its weakness named.
An applied extension of the Inner Direction series: the Organizational Singularity context treated as a contested premise, an interface requirement on agentic loops, six auditable procedural steps, the instruments transposed with their weaknesses, and the attention-inequality objection stated with more force.
A plain definition of the Hofstadter butterfly, and the first attempt to overlay electromagnetism on a Riemann model built for gravity and light. Downward vorticity, open versus material regions, and the questions that remain open.
The continuing scientific revolution is not a courtesy to prior certainty. It is a correction of it. On the experimental contact points now forcing revisions, and why the origin of an error matters less than the test that retires it.
Seven recent observations — cardioid envelopes, braided arrows, continua instead of tools, missing moats, engineered societies, resource thresholds, and high-dimensional light — restated briefly and mapped to the Twelve Theoretical Minimums and to the Theory of Everything.
"Artificial intelligence" is a naming error that decides the argument before it starts. A better pair of terms — evolved intelligence and engineered intelligence — plus what actually overlaps between the two substrates.
Quantum hydrodynamics, the Riemann, and a mental model with its falsifiers.
The coda and through-line of the project: when every theory and word has been spent, something remains that both humans and machine intelligence recognize. What that recognition is, and why the answer is left to each reader.
UMAP buys a picture by imposing a plane; Bonsai buys one by fitting a structure the data can support. A representation upgrade is a flattening until its discarded structure is named and priced.
A Nature Communications study finds cortical dynamics largely explained by inter-regional distance, with connectivity needed only in association cortex. What that constrains, and what it does not.
A drain vortex is not a black hole, and the analogy is still worth something. What acoustic metrics and analogue horizons actually buy, and where the borrowing stops.