Suggest softening the advocacy for `package::function()` syntax as a blanket rule

[grahamison]

It is a valid convention, but not the only one, and might confuse beginners who will probably be used to seeing library() used in reliable sources. I think library() can be a valid choice, e.g. an analysis script with extensive use of dplyr, or for ggplot2 functions.

There is a later section "Examples why you should avoid library() calls". Again, my preference would be to soften this - "you should avoid" is too strong for me.

[pbbgss]

I strongly disagree on this. The impact of library() calls on reproducibility is too high.

[dussaf]

I think if people have been attending the intro to R course and to some extend the ggplot one they should be very familiar with the package::function syntax as we introduce and use that a lot; so it shouldn't be too confusing for them.