Allow to create Darwin Core Archive for non-published dataset

inbo / movepub

R package to prepare animal tracking data from Movebank for publication in a research repository or GBIF

https://inbo.github.io/movepub/

Other

3 stars 1 forks source link

Allow to create Darwin Core Archive for non-published dataset #72

Open peterdesmet opened 1 month ago

peterdesmet commented 1 month ago

[ ] Don't call datacite_to_eml()
[ ] Allow to provide dataset_id = package$id
[ ] Allow to provide dataset_name = package$title`
[ ] Allow to provide rights_holder
[ ] Allow to provide license
[ ] Indicate this in the documentation, cf. https://github.com/inbo/camtrapdp/blob/f48d89eef6abc07e44bbd164aab05f772590a672/R/write_dwc.R#L36-L45

sarahcd commented 1 month ago

@peterdesmet what is the use case? Let me know if you want me to check code or examples for this.

peterdesmet commented 1 month ago

write_dwc() currently demands a DOI to exist, but only in order to populate datasetID. I think that's too strict of a condition. I've noticed people also use the write_dwc() function to transform Movebank data to Darwin Core (e.g. to downsample), without the necessity of having it published, see e.g. http://doi.org/10.1111/ecog.07246

sarahcd commented 1 month ago

I agree. In terms of Movebank the source study/url (like you use in the metadata for your Zenodo data pubs) is an easy place to start. I'd just want to be sure that we guide the user through appropriate use, e.g., they might translate and resample data for various uses, but should not then publish it on GBIF if it is not 'their' data (even if it is CC0). This could get a little fuzzy.