Post #3186969

2025-02-06 06:04 UTC

@jospueyo@fosstodon.org oh absolutely, I would take CRAN over Python any day - the quality of R packages, documentation, and ease of access is amazing. It's just so disappointing when one misunderstanding or (human) error can kill a package dead. This should be fixable!

Replies (2)

@JorisMeys@mstdn.social 2025-02-06 06:08

@jimbob@aus.social @jospueyo@fosstodon.org it is fixable. Just add a dontrun to the example. That's 2 minutes max. If the maintainers can't be bothered, that's not on CRAN. And you can't expect them to make little changes to all the packages. CRAN can be improved for sure, but this is not a problem CRAN caused.

Open ##3186970
@HydrePrever@mathstodon.xyz 2025-02-06 06:14

@jimbob@aus.social @jospueyo@fosstodon.org I understand that they don't want to have to read the warnings each time a package is tested and to decide if it's harmful or not. In this case there certainly are easy ways to avoid the warning - iirc it's not difficult to tell curl to skip SSL validity checking. But the way they communicate it... It's death by a thousand paper cuts

Open ##3186971