[scottshepard]

I think trust is a bigger deal in the R community than say Ruby. With a gem you can tell pretty quickly if it works or not, and the community has a higher standard of testing.

With R, many of the packages are for complicated math and stats. That particular package might pass all it's internal tests, but what if they implemented a distribution wrong, or calculated 95% confidence incorrectly? That is when you have to trust the developer personally.