Future proposals: terrestrial sky, planetary maps and orbits

These are proposals for 0.4–0.6, not implemented features, release dates or frozen public API names. They extend the existing roadmap; GPU, volumes, large OSM PBF, universal grid-based CRS, global overlay/topology, geocoding/routing and TeX/complex shaping remain planned and are not displaced. Scope must be revisited after 0.3 release evidence and maintenance feedback.

0.4 proposal — terrestrial Sun and Moon

• Own scene objects for Sun/Moon; manual solar azimuth/elevation and positions computed from explicit UTC instant, observer longitude/latitude/elevation.
• Solar lighting of 3D terrain/buildings, incidence, physically defined shadows and temporal shadow animation. Apparent daily/seasonal paths, sunrise/sunset, subsolar point, day/night terminator, solstices/equinoxes.
• Moon position, Earth-relative orbit, temporal playback and 3D Earth–Moon view; sublunar point and phases from Sun–Earth–Moon illumination geometry.
• Eclipses only where spatial/temporal precision and extended-source/occluder geometry justify them; a decorative overlap must not be advertised as prediction.
• Initial scope remains terrestrial geography, not generic solar-system simulation.

Acceptance requires declared ephemeris/model accuracy, validity dates, UTC/time scale conversion, reference frame, coordinate units and observer/refraction assumptions. Validate against independent authoritative fixtures whose source and license are recorded. Renderer light vectors alone do not establish solar/lunar positions, shadows or reliable eclipse timing. No model/dataset is selected yet.

0.5 proposal — Beyond Earth cartography

• Public body model: name/identity, size/shape, sphere/ellipsoid/triaxial model, prime meridian, axes/orientation and coordinate conventions. Earth becomes a preset; initial candidates Moon, Mars, Venus and Mercury. Extensible to moons, dwarf planets/asteroids and explicitly supplied irregular meshes.
• Planetocentric versus planetographic coordinates, longitude conventions, body-specific global/polar projections, geodesy (distance, azimuth, routes, lengths, area), physical units and surface elevation profiles between points.
• Extraterrestrial DEMs, orbital georeferenced imagery/mosaics, terrain textures, 2D/3D relief, hillshade, vertical exaggeration and geological legends/features: craters, volcanoes, mountains, canyons, maria, caps, faults and named regions.
• Extend illumination/terminators/incidence from 0.4; atmospheric models only with declared data/physics. Compare worlds in a Figure with explicit physical or normalized visual scales; whole body → region → feature → local terrain.
• Unwrapped global maps; irregular-mesh unwrap remains separate research work.
• Landing sites, rover tracks/waypoints/explored regions and temporal seasonal caps, dust storms and illumination when appropriate data are supplied.
• Existing scatter/line/raster/terrain/labels/scales/geojson operate on the active body; no second parallel plotting library. Earth defaults retain compatibility.

Acceptance requires authoritative body constants, datum/epoch/frame identifiers, units and source licensing. Triaxial/irregular geodesics, projection inverses and area require their own numerical limits; an Earth formula with a new radius is not automatically a correct irregular-body implementation.

0.6 proposal — Orbits and spatial visualization

• Scientific 2D/3D solar-system scenes: Sun, planets, moons, dwarf planets, asteroids/comets; explicit hierarchical objects/referentials.
• Keplerian orbit elements: semimajor axis, eccentricity, inclination, ascending node, periapsis argument and anomaly/epoch; finite temporal updates/play/pause/ slider/speed/steps/animation/export. Specify elliptical/hyperbolic/parabolic domains and precision rather than treating every conic as an ellipse.
• Heliocentric/barycentric/geocentric/body-centric reference frames; changing center must transform the whole scene consistently. Ecliptic/orbital planes, inclination, perihelion/aphelion/nodes and past/present/future tracks.
• Data-supplied spacecraft trajectories, flybys/encounters and comparisons; no implicit mission optimization or operational navigation.
• Explicit physical versus visual scales/exaggeration for both distance and size; asteroid/Kuiper belts and trans-Neptunian objects when licensed data permit.
• Geometric lighting, mutual shadows, alignments, eclipses/simple occultations with precision limitations; spatial camera orbit/pan/zoom/follow/focus/reset.
• Picking exposes name/position/distance/period/velocity/elements when valid. Possible high-level APIs include time, focus, orbits, bodies and trajectories; names such as space3d/solar_system are proposals, not reserved promises.
• No initial promise of high-accuracy N-body, professional full propagators, maneuver planning, transfer optimization or operational astrodynamics.

Shared architecture and sequencing

The sequence is terrestrial sky → body-general cartography → orbit/reference connections. Adopt compatible internal body/frame/unit/time abstractions before 0.4 creates Earth/Moon objects, then stabilize the public Body interface in 0.5. 0.6 must use the same bodies and scene/Artist/picking/temporal/export mechanisms, not parallel representations of planets. Current 3D/temporal APIs remain bounded. Celestial work complements urban/cartographic maintenance; it does not replace it.