Thoughts

One of the few annoying things I find with `dplyr` lately is the addition of an experimental option where if you use something like, group_by(X) %>% summarize( Val = mean(Y)) it gives you an error message with something like. `summarise()` has grouped output by 'X'. You can override using the `.groups` argument. But why would I want to 'override' that grouping dplyr::authors? Didn't I **just set the grouping**? I would suspect that the majority of use cases are exactly l