Comment by ModernMech
2 days ago
> The ratio has just been going up and up and up
That's because organizations get bigger as projects become more complicated and varied. Larger organizations require more overhead as a percentage of the operating costs. 30 years ago many schools didn't even have Computer Science departments. Today schools are now starting to stand up Artificial Intelligence departments. It's not cheap to maintain these organizations.
Anyway, it really comes down to a simple tension: you can have big science, good science, or cheap science. Choose two.
For a long time we've optimized for big and good. This has yielded dividends in terms of science and technology output, but it's very expensive. Yet, the ROI is decidedly, emphatically positive.
For some reason people seem to think we can do this all cheaper, somehow, by pulling funding and making all these organizations smaller. I don't see how this is possible, because it relies on an uncanny ability to predict which projects will succeed and fail ahead of time.
What I think will happen is the money will dry up, the talent will go to places that want to spend the money, and the remaining programs will be cheap, small-stakes research better suited for the 20th century, unable to compete with countries that actually want to invest in the future.
I don't believe that the feeling is that it can be done "cheap", but rather the inevitable draw towards privatizing previously government functions. This drives towards a more immediate profit motive, which inevitably pushes research towards a more applied focus.
Who's doing the research? Who's benefiting?
VC backed companies are often slotted in as the for profit version of academic R&D without the "encumbrances" of non-profitable blue-sky pure research.