R and RStudio in repo2docker without waiting for packages to compile!
Jupyter+R:
RStudio:
Binder supports using R and RStudio, with libraries pinned to a specific versions.
Install R itself and your required R packages via conda packages. Installing conda packages is faster than installing CRAN packages. This is because CRAN packages need compiling during the install process and conda packages do not.
For some R packages there is no corresponding conda-forge package yet, in that case take a look at https://github.com/binder-examples/r.
Both RStudio and IRKernel are installed by default, so you can use either the Jupyter notebook interface or the RStudio interface.