Research & Writing

The Romeo–Juliet Framework is a critical risk-analysis framework for any coordinated system operating under information asymmetry: AI, blockchains, politics, business, engineering. It maps the “deathly embrace” between problem-solvers and the capital order onto four squares — preparation (green), action (yellow), access (red), and status (blue) — and models the actors and sequestration channels that determine whether a project can move from zero to one.

The actors — Oscar, Mallory, Eve, Sybil, Olivia, and Walter — and the three sequestration channels (diagonal, left, right) describe how value is captured, how access is blocked, and where a project can move. The framework draws on Schumpeter, Kondratiev, Henry George, Gambetta, Nitzan and Bichler, Thiel, Nash, and the ρ-calculus/process-calculus tradition.

It is a critical risk-analysis tool first: it reads the failure modes of a system before the architecture is fixed, not after the fact. The paper applies the same structure to centralised AI, state power, intelligence agencies, rent collection, voting, law, engineering, and business. The CoEngineers experience (2017–2019), where I was an early investor, was one origin of the framework, not its scope.

Romeo–Juliet Framework archive (ArDrive; videos and three papers). YouTube playlist.

UNISS is the simulation layer: containment before a system exists — what moves, who can see it, and where a boundary has to sit. It models how influence, secrets, and layered compromise move through networked communities. The series extends through ESN, GX, GX-SD, and Ingenesist scenarios, including the Darkweave-Ingenesist endowment model.

The Ingenesist archive preserves Dan Roble’s presentations and papers published with NAIC, ASME, ASCE, NSPE, and others. The “blockchain for engineers” line is the lineage for UNISS. Ingenesist archive.

Big Data Bowl entry. A judged entry to the 2027 NFL Big Data Bowl. A 10 Hz Combine burst of speed and acceleration predicts interior pressure and sacks; the 40-yard dash and 10-yard split do not. The signal is position-specific — real on the defensive line, marginal at receiver, absent on the offensive line and at tight end. The entry’s second finding is an allocation gap: the cap pays for performance, but the depth chart follows prior snaps and draft position, not measured performance. The preregistered RJF/Yellow Square test found no detectable effect and is reported as a perspective, not confirmation.

The dollar findings. The follow-on valuation prices that gap. On the actual 2026 free-agent class, the median team carries about $7.0M a year of uncut role premium — pay that tracks playing time beyond measured performance. At a 0.40 capture assumption, that is $2.8M gross / $1.8M net after a $1M implementation cost; the conservative EDGE/safety/corner case is −$0.45M net. The role premium is +0.179 log points on 2026 APY (95% CI 0.022–0.336, n=306), close to +0.185 on the 2025 cap ledger, but it falls to +0.049 with contract-length and guarantee controls. The team distribution runs from $1.8M at P10 to $22.5M at P90, and the top 10% of contracts carry 51% of the premium. These are inferences from the 2026 class, not forecasts or causal estimates.

The longer record is in the archive at technologytruth.substack.com.