01
Identity as evaluation criteria
Objectives, constraints, and correctness are declared, not inferred from the last chart. The score is only valid inside the Identity frame that produced it.
Overview
Score and threshold decisions under uncertain measurement.
P1-314 is the evaluation surface of the Intelligence OS. It turns the Identity frame — objectives, constraints, and criteria of correctness — into a score the rest of the stack can act on. The result is a number that is always the current evaluation, recomputed against the observations present at the moment it is issued, not a KPI held from the last reporting cycle.
CURRENT RESULTS ARE A BASELINE, NOT AN UPPER BOUND.
Continuous evaluation.
Most scorecards treat a metric as equally valid at every point in the operation. In practice, measurement quality varies by source, lag, and definition. A number that is reliable in aggregate can be systematically off at the decision that depends on it. 314 corrects for this variation before anything downstream consumes the score.
An institution is not one measurement problem. It is a field of measurement problems, each with its own error pattern. 314 is built around that structure. Where a dashboard ends, 314 begins.
Analytical architecture
Analytical architecture
02
Calibrated inputs
Upstream signals are treated as distributions with known error, not single-number estimates. 314 scores the field as corrected, not the raw feed.
03
Traceable thresholds
Every threshold is a decision: who set it, against which observations, in which cycle. The number can be replayed. It cannot be silent.
Position in the Principia stack
314 as an Intelligence OS application
314 is a domain-specific instantiation of the Intelligence OS architecture applied to evaluation. Identity — objectives, constraints, and criteria of correctness — is the final authority on the score. Calibration precision at the Chaos and Tempo layers determines the quality of the field 314 evaluates, which determines the threshold that fires, which determines the action the rest of the stack is allowed to take. Scores are traceable to the identity frame, the observation set, and the cycle that produced them.
Next
Talk to P1 314
Tell us what you measure and what a wrong number costs. We will come back on the evaluation surface.