going beyond DwC-A

[pieterprovoost]

There are several recent developments but also legacy situations that may require us to look beyond Darwin Core Archives and think about implementing connectors for other data formats. We are well aware of the limitations of the DwC-A star schema, and while we have developed a stop-gap solution, there are other more generalizable solutions being developed for this problem. This is a non exhaustive list of formats and protocols I think we need to look into:

• BioCASe: Senckenberg is exposing several of their collections through BioCASe/ABCD. While it is technically possible for them to generate DwC-A, it seems that it will be a challenge to ensure they stay up-to-date.
  • https://www.biocase.org/products/protocols/
  • https://www.biocase.org/whats_biocase/index.shtml
  • https://www.gbif.org/installation/60383ab8-f762-11e1-a439-00145eb45e9a
• Frictionless Data: this would allow us to publish Darwin Core data using a more relational schema. I believe Darwin Core Archives can be made compatible with Frictionless Data packages by adding an extra JSON file.
  • https://frictionlessdata.io/
  • https://data-blog.gbif.org/post/frictionless-data-and-darwin-core/
• BIOM: We were considering converting BIOM files to DwC-A for use in OBIS, but there's a suggestion in the Publishing sequence-derived data through biodiversity data platforms guidelines document to skip that step (at least for ingestion into GBIF).
  • https://biom-format.org/

[meliezer]

@pieterprovoost Very interesting! Do you have updates about the following Frictionless Data test case? C2:Phytoplankton-BG Black Sea-2007-2016 by OBIS Event core + 2 extensions (ExtendedMeasurementOrFact & Occurrence) (DwcA) ad mentioned in https://github.com/frictionlessdata/FrictionlessDarwinCore/blob/master/testCases.md ?