# Accessibility

Status: cross-cutting implementation contract · Version 0.1.0 · 2026-08-21

The Scene is not exempt from interface accessibility because it is spatial. Every meaningful spatial Component has a semantic representation that can be focused, named, understood and acted upon without precise pointer control, motion perception, sound or color discrimination.

## Required for interactive Components

- concise accessible name and optional description;
- semantic kind and current state;
- Focus Anchor and ordered navigation hint;
- Action names, consequences and permission constraints;
- hit target of at least 44 × 44 CSS px at the active zoom tier;
- selected, unavailable and consequential states conveyed by shape or text as well as color;
- reduced-motion representation;
- visible keyboard focus with sufficient contrast.

## Scene reading order

The default semantic order follows Purpose, current Area, current Avatar, nearby relationships, focal Components, Portals, distant Areas, then ambient description. Visual depth order does not determine reading order.

## Motion and sound

Ambient animation pauses in reduced-motion mode or resolves to a stable state. No effect flashes above 3 Hz. Spatial audio supplements but never replaces visible and semantic cues. Captions and transcripts attach to the relevant Component or Scene event.

## Color and contrast

Gold, teal, sky and ember accents are never the only state carrier. Text and critical linework meet WCAG contrast against their immediate composited background. If atmospheric imagery undermines contrast, a world-material plaque or restrained Overlay may provide separation without turning the Scene into a conventional dashboard.