Programming Opportunities for the Global Biodiversity Observation Network
summary
A platform called BON in a box packages community-contributed scripts plus a YAML description into reusable pipelines that compute EBVs and CBD indicators; the codebase is open source, pipelines are transparent with intermediate results, and a peer review (3 reviewers) label vets workflows. It runs on laptops or servers via Docker Compose and is adding Slurm/Apptainer for HPC. Known gaps: dependency drift breaks reproducibility, expensive full re‑testing, need for impact analysis (call‑graph diffs), pluginable WASM visualizers, and metadata‑driven chains of trust for sensitive occurrence data (some indicators today only one lab can calculate).
AI-written summary · about
chapters (11)
- 0:00Geobon introduction
- 2:29Geobon’s role in monitoring
- 4:36Biodiversity pipelines in BonBox
- 7:01Pipeline engine principles
- 8:46Programming pipeline engine
- 9:34Peer review for pipelines
- 12:43Pipeline visualization challenges
- 14:13Trust for sensitive data
- 16:03Dynamic data flow modeling
- 17:27Q&A data spaces mismatches
- 21:38Q&A dependency changes updates
related talks
01h 11mProgramming for the planet
32mDiscussion on multidisciplinary PROPL-work
35mStreamlining the Cosmos: Pythonic Workflow Management for Astronomical Analysis
34mOptimal Observability
36mThe OpenClimate Network: Nested and Connected Climate Accounting for a Digitally...
17mThe Biodiversity Risk Filter - A Tool to Save Lives
42mOS-Climate: A Breakthrough Open Data and Open Source Analytics Collaboration for...
05mCollecting Massive Data via Crowdsourcing