I don't think we can make a more competitive system that isn't gained. My issue is that research is itself very inefficient. We need to bake new research paper

I don't think we can make a more competitive system that isn't gained. My issue is that research is itself very inefficient. We need to bake new research papers into encyclopedias at a much higher rate. We need to automate tons of research so there's no more grad students hanging around beakers. We need to make much better programming tools than python or Cuda, that allow working with high level concepts not shitty arrays. All that will require a lot of work. We can siphon would-be researches to do that instead. The remaining research will be more productive.