Build trajectory

Status

Astrolyte is being built in phases, starting from stable real-data lanes before expanding into broader synchronization and richer products.

Phases

Build phases

The roadmap follows the same logic as the source repos: start narrow, stabilize with real public data, then expand the product surface only after validation is visible.

1

Core ingest and structure

Current

Core ingest and structure: define lanes, preserve raw truth, and add metadata indexing on top of stable real-data workflows.

define stable source lanes
preserve raw source records
add metadata indexing
publish first dataset surface

2

Validation and serving

Validation and serving: expose checks, provenance, status, and cleaner lane-level inspection surfaces.

add validation checks
show provenance state
expand dataset pages
surface freshness notes

3

Multi-source synchronization

Multi-source synchronization: align sources and push beyond isolated lanes into richer downstream products.

cross-source alignment
richer derived outputs
consistency rules
deeper audit logic

4

Product maturity

Product maturity: tighten documentation, changelog discipline, and the public-facing product surface.

cleaner product surface
lane changelog
clearer documentation
public serving polish

Direction

Current direction

The current build posture is shaped less by interface novelty and more by reproducibility, provenance, and lane hardening.

Pattern

IRIS pattern

Start narrow, stabilize one scientific lane, then harden it across multiple real observations.

Pattern

Rubin pattern

Ship a measurable baseline first, surface failure modes, then scale after the evaluation is honest.

Pattern

T CrB pattern

Keep lineage reproducible before layering in historical context, archive support, and richer products.