Kind of thankful that #RStats is not a system dependency that is pretty much embedded in the OS, and that I can upgrade it even on older software (like running R 4.3.0 on Ubuntu 20 LTS) without worrying about other things depending on it.
In contrast to #python 3.6, that I'm not keen to try and force an upgrade to 3.8 so I can get access to a newer version of SciPy that exposes some underlying functionality that one of our python packages depends on.
Before anyone suggests Docker or VM, this is already a VM running on top of a cluster of machines, and it has another big project for a manuscript that is currently under review inside of it, so I'm very hesitant to break it in any way before the manuscript is completely done.
Add comment