[C3 HL-019] Build relational color, geometry, and perceptual field experiments #29

Outcome

Use restricted variables to produce selected original visual studies.

Placement and activation

• Circle: C3 — Composition studio and governed generative experiments
• System responsibilities: Product / Assurance
• Capability coverage: 01 Core Domain / Proprietary Engine; 02 Product Applications & Interfaces; 08 Quality, Testing & Evaluation
• Planning status: C1 three-field instrument verified/deployed; C2 adoption and C3 extensions remain planned.
• Activation: Optional expansion after core acceptance and a demonstrated authoring need
• Parent epic: [EPIC C3] Composition studio and governed generative experiments #6; master roadmap: [MASTER EPIC] Hole-loop: concentric evolution within Persona / Visual Form Canon #2.

Dependencies

C1 color instrument: uses verified rendering/configuration from #17/#19; already completed mechanically, independently of C3 patch/quality work.

C2 field adoption: uses that instrument's captures and relevant artist/reader observations from #20. Those later decisions are not prerequisites for #20 to verify the baseline.

C3 geometry/perceptual extensions: require the relevant versioned parameter contract from #28, changed-risk release checks from #25 and an artist-reviewed question; no full studio closure is required.

Acceptance evidence

• Compare identical target color/text/geometry in different surrounds; verify identical target values and vary adjacency, area ratio, hue, saturation, and value independently.
• Compare orthogonal, diagonal, and variable-interval grammars with identical corpus/route rules; validate each declared constraint.
• Compare static perceptual interference, slowly modulated fields, and plain navigation of the same graph; movement can stop and controls remain legible.
• Record parameters, renders, observation notes, and artist selections; repeatable mechanics do not prove that a perceptual effect occurred for every reader.
• Provide non-motion and non-color-only equivalents and prevent unreviewed flashing patterns.
• Review compatible Glyph Cascade rendering classes through HL-045 #14; preserve hole-loop source wording/room semantics and verify imports/builds before selecting an adapter.

Artistic experiment

Albers suggests relational color; Malevich/Concrete Art suggests reduction; Mondrian and Mohamedi suggest distinct constraint inquiries; Op Art suggests figure-ground instability. Do not collapse these into one lineage.

Completion and scope

Attach the actual artifact or commit, relevant verification, limitations, and an artist decision where a composition changes. This issue plans authorized work; creation does not claim implementation or deployment. Preserve independent work/source identities and private source boundaries. Repository creation requires the extraction evidence in HL-004 #13 and HL-045 #14.

Completed baseline and open decisions

The implemented instrument fixes 042115 wording, foreground #292b29, Arial/sans-serif, weight, size, placement and actual bounds; only the field changes between #ede6d6 / #b8d6c4 / #d3bde2. Reset and configuration recording work. This baseline is part of C1 #56; current head/CI/deployment evidence is recorded in #20.

No winning palette or universal color response is established. Current green Moment and peach Progress fields are proposals. Collect figure–ground/legibility/reading observations and an explicit artist selection or rejection before score adoption; version any approved configuration and rerun fixed-property/score checks.

The original acceptance list above is the unimplemented C3 expansion: adjacency/area/value experiments, constraint grammars, perceptual interference and any selected adapter. Three background toggles do not complete those capabilities.

Artist-requested generative pilot · 8 October 2026

The old palette values above describe the initial 687bd3e deployment, not the current edition. Protected score configuration 1.1.0 restores source cyan/black/pink fields and white Futura-preferred type. Its independent color study fixes that white text while comparing #ff00aa / #006c53 / #492475. The new generative configuration 1.0.0, explicitly requested by the artist, couples entrance black/cyan/magenta/yellow to loop fields and selects three seeded RGB word colors with calculated opaque-color contrast >=4.5:1. This is not a winning-palette finding, a universal color claim, or completion of the broader C3 controlled perceptual studies.

Implemented and locally verified on PR #59, branch experiment/six-encounters, commit c5add03c4d458e3c84fe1fcf21f612adb8d8d1d4. See experimental/six-encounters/docs/GENERATIVE.md and verification/generative/. Evidence: 19 pure state/rule tests and 30 Chromium browser groups across desktop/mobile, including 320px keyboard/reduced-motion controls. No artist approval of the resulting render, physical phone test, or full accessibility certification is claimed. Deployment receipt follows in the PR. Wider issue acceptance remains open.