Could the IPT take DwCA that are updated online?

[ManonGros]

The Specify platforms can generate Darwin Core Archives and makes them available online. These archives can be updated automatically.
Usually, we ask publishers to register the Archive endpoints directly (or helpdesk does it for them).

However, the metadata associated with these datasets isn't always what the publishers would like to share on GBIF. @spalp Asked if it would be possible to have the Darwin core archives content in the IPT but have the EML from the IPT.

[mike-podolskiy90]

Thanks Marie
cc @spalp

[spalp]

Thank @marie. My question was whether some of the metadata fields could be taken as they are from the Archive, but others are automatically updated/added when ingested. The fields that are almost certain to change between archive versions are:

1. Coverage: temporalCoverage, geographicCoverage and temporalCoverage,
2. Additional metadata: dateSamp, citation

I just talked with @mike-podolskiy90 and he ensured me that even if we fetch a complete DwC from a URL, together with its EML, the option to automatically infer Geographic, Temporal and Taxonomic scope, if previously selected, should not be affected.

So, the only new feature I would like is the ability to provide a URL to a DwC-A to be regularly monitored and published via the IPT.

[mike-podolskiy90]

Thanks Salza
This sounds like a useful feature. I would like to gather more opinions on how this should be implemented.

[mike-podolskiy90]

My idea here is to create a new source type - DwCA (or URL/DwCA, we can discuss). IPT takes this archive from the provided URL and publishes it. I am not sure if we should unpack the archive and reassemble it with IPT with validation and all. I suppose it should, to ensure the quality of DwCA.

EML will be taken from the archive, but with the ability to automatically infer coverage metadata.

@gbif/dataproducts Andrea and Cecilie, maybe you have something to add here please