SD-165
Layer model v0.2 — refined from empirical data. Map at docs/lexical-harness-not-prompt-harness.md updated. Changes: (1) L6 decomposed into three operational modes: L6a direct/interruptible, L6b dispatch/queued, L6c override/kill (from Captain’s rubric, SD-160). (2) L4 extended with reasoning_tokens as observable alignment channel (SD-162). (3) L3 extended with compaction as human-controllable lever (SD-160). (4) Temporal Asymmetry added as cross-cutting concern. (5) Loading Points section added: CONVENTION, CONVERGENCE, DIVERGENCE, ATTESTATION annotations at each layer. (6) L12 extended with human’s instruments and empirical data sources. (7) All annotations back-reference SDs and Lexicon entries. Map header references SD-162 (The Map Is Not The Territory).
← all decisions