Skip to contents

Getting help

Start here before opening an issue:

  1. Read the Getting Started vignette: vignette("geefetch").
  2. Browse the pkgdown site: https://aagi-aus.github.io/geefetch/.
  3. Search the open and closed issues.
  4. Use GitHub Discussions for usage questions and ideas; reserve Issues for bugs and concrete feature requests.

Authentication and setup troubleshooting is covered in ?gee_setup and in the Getting Started vignette.

Maintenance capacity

Please read this before filing feature requests, opening large PRs, or assuming a response time. Honest capacity disclosure protects both users and maintainers.

Time budget

Maintainer Role Capacity
Max Moldovan (Adelaide University) Primary / cre ~2–4 hours per week
Adam H. Sparks (Curtin University) Secondary / review ~1 hour per week

Capacity can drop temporarily during teaching, fieldwork, conference travel, or grant-writing cycles. During those periods, issue triage may lag by up to 10–14 days.

Response expectations

  • Bugs with a reprex: acknowledged within 7 days. Fix cadence depends on severity — see SECURITY.md for the security-specific SLA.
  • Feature requests: triaged within 14 days. Not every request is in scope — see API_STABILITY.md and DESCRIPTION for scope boundaries. Low-effort, in-scope asks land fastest.
  • Pull requests: initial review within 14 days. Complex PRs may take multiple review rounds.
  • Discussions: best-effort, no SLA.

Bus factor

Bus factor: 2. Both maintainers can triage, review, and release. The competence-matrix table in CONTRIBUTING.md documents who holds domain knowledge for each subsystem.

Abandonment / sunset protocol

If geefetch receives no maintainer activity for 6 months (no commits, no issue responses, no release):

  1. The AAGI-AUS org steward (GitHub @AAGI-AUS owners) opens a lifecycle:seeking-maintainer issue and notifies the mailing list.
  2. If no maintainer steps forward within 3 months, the lifecycle badge on every exported function flips to superseded (if a replacement exists) or deprecated (if not), and a final CRAN archival release ships.
  3. The GitHub repo is archived (read-only). Historical source remains available; no new issues or PRs are accepted.

This protocol is documented here so it isn’t ambiguous when it’s needed.

What we don’t support

  • Non-GEE data sources. This is deliberately scoped to Google Earth Engine. For Australian sensor-network data, use nert. For SoilGrids, WorldClim direct access, GBIF, etc., use the specialised upstream packages.
  • Server-side GEE computation. geefetch extracts; it does not orchestrate server-side mosaicing, classification, or band algebra beyond the scale / offset / filter primitives needed for extraction. Use rgee for complex server-side work.
  • Visualisation. Use terra::plot(), ggplot2 + tidyterra, or leaflet.
  • Modelling. Use nlme, mgcv, ranger, brms, etc. geefetch stops at the data.table / SpatRaster output.

Commercial support

Consulting engagements are case-by-case via Max’s consulting practice. For organisational-scale deployments, data-pipeline integration, or custom GEE workflow development, open a discussion thread and we’ll scope it.