Ecologists, researchers working with geological or geospatial data use R and have been using R for a long, long time. The 'old 'standard library 'maps' is decades old by now, with maps v3 now being 8 years old. maps v2 goes back to 2003, about the time of Python 2.3?
I can't even find the sources of maps v1. https://cran.r-project.org/src/contrib/Archive/maps/ I believe that maps v1 was for R's predecessor, S, so from the late 90s, about the time Python 1.5 came out. Was anybody even using Python then?
OP's post uses the slightly fancier tidyverse-versions of these decades-old libraries. I think it's safe to say that map-making libraries are older in R than in Python.
Most of the R packages in this post aren't that new either.
Python for everything, but if I need to make something nice to look at, R makes my life so much easier, and the final touch in Affinity Designer/Photo.
In jupyter it's trivial to mix and match python and R anyway, all the wrangling in python and once it's clean, just basic tidyverse in R
"More and better" is simply subjective and not substantiated. In fact detailed and objective comparisons of capabilities, performance etc between these ecosystems are sorely lacking not just in map-making but practially concerning every comparison.
Still, the "waste of open source energy argument" is quite bit more complex and I'm frequently wondering about this.
For sure this "waste" happens everywhere in open source. About the only thing that approaches some optimal level of efficiency is the Linux kernel. From there on, distributions, desktops, toolchains, apps, libraries etc. is the epitomy of chaos.
This diversity creates resilience and a sort of competition for the survival of the fittest. The openness also facilitates fast diffusion and borrowing good ideas. But it is also true that open source resources are still scarce so development happens at maybe much lower speed than what would be possible if more coordination / less fragmentation was practiced.
ggplot2 is an obvious heavy-weight, but other packages such as mapsf and tmap also are top notch in this regard.
on the other hand, my general map making experience in Python has not been very pleasant when I had to work towards creating a map following acceptable cartographic practices, ie. anything beyond throwing data on a webmap layer with subpar legend and symbology.
....something....something...requirement.. different approach...