As a whole, the BON in a Box platform is designed to facilitate the establishment and operation of
Biodiversity Observation Networks (BONs). It does so by providing a pipeline engine aiming to enhance the
capacity of BONs and countries to report on biodiversity effectively. It runs user-contributed
code on a shared infrastructure, the outputs of which can be shared for auditability, reproducibility,
and transparency.
BON in a Box’s pipeline engine is
-
an open-source repository of pipelines
to transform data into biodiversity indicators and EBVs
- a modeling tool to turn your workflows into automated pipelines, or edit the existing pipelines
- a form to adapt the pipeline to a local context and a results viewer to present them
- community-driven, aiming to create a peer-review process to enhance quality and collaboration.
BON in a Box’s pipeline engine is designed to be
- modular and expandable, with a specific emphasis on empowering users to bring their pipelines and indicators to the platform
- language agnostic (currently runs R, Python, Julia, although more can be added), allowing users to collaborate on science regardless of their software stack
- user-friendly, with a visual pipeline editor allowing a no-code approach to pipeline design when relying on existing components
- data-source agnostic, accepting a mix of user-provided and globally available data
- fully transparent, allowing all intermediate results, logs, and code to be audited
- portable (can be ran on on personal computers or on the cloud, with capacity to be integrated to high-performance computing resources)
BON in a Box is not
- a data repository
- platform or computer language dependent
- a ready-made solution showing pre-calculated global maps
- domain specific
- closed box
- for profit
