Objective measures to judge package quality?

[DamienIrving]

From ACCESS-NRI/ACCESS-Hive-Docs#207 (comment):

It would be nice to have some metrics to judge if something were appropriate.
e.g. libraries.io has a "quality score" based on a bunch of stuff: https://libraries.io/conda/xskillscore

[aidanheerdegen]

You could have an issue/PR template with the criteria listed, e.g.

• user-facing package: xarray is user-facing, cftime is not
• actively developed: updated in last x months/years
• general audience: designed for and usable by general community, not targeting sub-section
• specific to pyAOS community: e.g. geopandas is much wider and not pyAOS specific
• unique functionality: e.g. tools like PCDMI coordinate calling other tools rather than providing their own functionality