Track linkage between data package properties and established standards

frictionlessdata / datapackage

Data Package is a standard consisting of a set of simple yet extensible specifications to describe datasets, data files and tabular data. It is a data definition language (DDL) and data API that facilitates findability, accessibility, interoperability, and reusability (FAIR) of data.

https://datapackage.org

The Unlicense

501 stars 114 forks source link

Track linkage between data package properties and established standards #936

Open pschumm opened 6 months ago

pschumm commented 6 months ago

@pschumm "linkage between these metadata properties and other established standards such as Dublin Core" is true for all properties in Data Package and are maybe better defined at the common dictionary level, so they apply to all Data Package elements that use them. In Camtrap DP, I have added those as skos:exactMatch (or alike) properties, see here. Maybe we should open an issue to do discuss this further?

Originally posted by @peterdesmet in https://github.com/frictionlessdata/datapackage/issues/63#issuecomment-2109525613

pschumm commented 6 months ago

Specifically, the proposal here is to:

Determine whether additional metadata properties should be defined at the common dictionary level for consistency and maintainability, so that they apply to all data package elements that use them (as suggested by @peterdesmet above)
Add to those definitions linkage to established standards such as the Dublin Core or others (cf. what @peterdesmet has done in Camtrap DP)

Note that (2) would really help when pushing greater use of Frictionless within NIH-funded (and related) initiatives and projects.

I don't have time to work on this right now, but will do so later on. This issue is intended to serve as a tickler for that, but also for anyone who may wish to contribute to this in the meantime (which would be terrific).

khusmann commented 6 months ago

Note that (2) would really help when pushing greater use of Frictionless within NIH-funded (and related) initiatives and projects.

Agreed! I don't have bandwidth for this either atm, but would be happy to help out later on as well.

As you say, I think the ability to link to Dublin core, etc. would be a great selling point in our eventual publications to the NIH et al. audience.

roll commented 1 month ago

It's not exactly related to ontologies but please take a look - https://github.com/frictionlessdata/datapackage/pull/988 (CSVW comparison guide!)