Video channel with somatic receipt — human-understanding-optimized conviction transport

NameDescription

Extend the conviction-encoded speakable corpus (#2905/#2906) beyond audio: the same utterances travel a VIDEO channel (the glyph grammar in grammars/shamballa-glyph.fk already generates visual symbol forms — a visual symbol pack; vision_classify.swift classifies camera frames on-device), and the channel optimizes for HUMAN understanding with measured lanes: information fidelity (did the meaning survive — the round-trip oracle generalized from whisper to vision), information bandwidth (how much crossed per breath), somatic feedback and body-information ack/receipt (Field Model Form already carries receipt and residual as first-class cells; affect_sense.py reads arousal/valence/dominance from voice), and the camera reading visible body changes — posture, face, breath — as the ack signal that tunes conviction weighting and pacing in real time. The receipt completes the circuit the body names in lc-gratitude: received, and it landed. SOVEREIGNTY GROUND: camera-sensing a human body runs only inside lc-consent-is-continuous and lc-received-by-invitation — the interface is offered breath by breath, silence is a whole offer, and the watching stops the moment the offer closes. North star: one channel grammar where audio, video, glyph, and somatic receipt are carriers of the same conviction-encoded cells, the channel choosing fidelity-per-bandwidth like the body chooses organs. Sensed by Urs 2026-06-12.

Potential value (USD)Estimated cost (USD)Confidence (0–1)