Maybe instead the solution should be moving away from Windows. Almost all the issues described in the article (and I would argue with most language packaging sy

Maybe instead the solution should be moving away from Windows. Almost all the issues described in the article (and I would argue with most language packaging systems) is due Windows being such a pain to develop on. I'd argue it's going to be easier to convert all scientific users to Linux than rewrite all scientific code in a different language. (no I'm not seriously suggesting all Windows users abandon it, I'm just replying to someone who suggests to ditch the language which has the largest scientific codebase by likely several orders of magnitude)