You can't really hyperscale research I think. Throwing more money at a problem isn't a solution. A viable approach is probably to try and remove as many adminis

You can't really hyperscale research I think. Throwing more money at a problem isn't a solution. A viable approach is probably to try and remove as many administrative barriers as possible so that researchers can focus on interesting problems instead, which is what they seem to be doing.