Interaction

Interaction and multiple views

For Canvas views, drag empty space to pan and scroll to zoom around the pointer. In 3D, left-drag orbits, right-drag pans, and scrolling zooms. Each space retains its camera independently of scene time.

Opt a point into dragging with drag: { parameter, coordinate?, min?, max? }. For a Point, the selected coordinate must directly name the bound parameter. For a PointOnCurve, its parameter must directly name that parameter. PointOnCurve values use normalized sampled arc length, not the original graph's x coordinate.

Dragging pauses playback and creates a parameter override. Overrides survive seeking until cleared or another document loads.

PictureInPicture displays another space inside the current view. Its space, x, y, width and height define the view and rectangle in CSS pixels. Insets have independent cameras, share the clock and parameters, and support visibility events. Use qualified inset IDs with the player's inset-camera methods.

SwitchSpace changes the active space; it supports cuts and fades. A plain-text object caption supplies a hover tooltip. Hover highlighting is temporary; the player does not automatically maintain application selection state.

Inset composition and camera behavior

Insets appear when their containing space is in the main view and inherit its transition opacity. Later objects stack above earlier ones. Insets do not recursively embed other insets. Each 3D view uses a WebGL context, so practical view counts depend on the device.

Authored camera events progress in hidden spaces and reconstruct on seek. Manual exploration is an offset on authored framing; resetCamera clears it at the current scene time. Camera windows refit when their view resizes. During a space fade, the destination owns interaction.

---

Wiki home · Quick start