And how DeepMind might be doing it
It’s great to hear about Gerald Holton. He was my advisor in college, where I majored in chemistry and physics. I’ve been a biologist at universities since then, and spent huge amounts of time writing grant applications and running small research projects in my lab or labs of other professors. I’ve dreamed of working at an institute that got a lump of cash that its scientists would vote on how to spend. Scientists in the institute would vote to pick one or several projects to focus on for several years, contributing their time to each project based on their specialization.
If a project looked like a dead-end, they’d vote to switch to a more promising area. If a scientist wasn’t pulling their weight, they’d be voted out of a job and a new scientist would be hired. Without a need to spend a large portion of their time on grant applications, and without the need to buy their own equipment that will sit idle the vast majority of the time, scientists could enjoy the benefits of specialization and economies of scale, along with the fun of working in a team for common goals, instead of alone.
This utopia sounds similar to Compton’s idea of department funding. Probably the optimum for science overall would be to have a range of lab sizes, with some big labs like these and some traditional labs led by individual professors. I hope some government or billionaire tries this big-lab or autonomous institute approach.