Audience: screening & vetting adjudicators · Artifacts: dossier.html (2D) · 3d.html + evidence.js with fc-room.js and fc-docs.js (3D evidence wall, flagship) · shared data dossier-data.js · Case study: index.html
In one line: an evidence-weighting dossier: twelve documents, three tags (PATTERN / ISOLATED / CONTEXT), one verdict — scored on the weight distribution, not the outcome.
Screening training teaches rules; adjudication demands weighting. The designed trap is structural: one vivid transparent disclosure (a repaid train-ticket charge) recruits moral certainty while the real case — a quietly curated self-presentation across CV versions, interview deflection and converging impressions — hides in small omissions.
Twelve documents in a fictional candidate file. Each gets tagged PATTERN, ISOLATED or CONTEXT with immediate feedback contrasting the learner's weight with the designed reading. Then a verdict: appointable-with-conditions or decline. The debrief shows the full weight distribution, names the two failure modes (declining on the dramatic disclosure; waving the small omissions), and re-anchors the lens: pattern-of-integrity vs isolated red flags.
The 3D evidence wall was checked in headless Chrome with software rendering (SwiftShader) only; not on a real GPU, phone or tablet. The fictional KBS service and candidate are composites from public screening-pattern literature — no real agency, service or person is depicted. The adjudication rubric is a design artifact, not a legal standard.
Part of I Wayan Usada's learning-design portfolio · all clients and scenarios are fictional composites built from realistic domain patterns · drafted with AI assistance and fully rewritten, fact-checked and owned by a human.
action-map.md — Action map — the behaviors the artifact must produce, traced to the business goalstoryboard.md — Text-based storyboard — the scene plan written before buildvisual-mockups.md — Visual mockups & style guide — mood, type, key screens, component decisions