Sure, spinoffs are nice, but they happen with all types of public access research. Why not create a project of a unprecedented magnitude and complexity that als

Sure, spinoffs are nice, but they happen with all types of public access research. Why not create a project of a unprecedented magnitude and complexity that also creates very important results directly? It's like saying "spend vast amounts of resources on pursuing random goals, and eventually some of the resources, by accident, will create a ROI". Yes they will, but is it really the best way to do this? It's not like some version of hypertext wouldn't happen without CERN. I'm not saying CERN wasn't useful. It was. And the budget wasn't that big. But this kind of logic is not very sound. If we have a lot of money and a series of concrete problems, we should spend the money to solve them. Most of the money are to be spend directly on the problem, less to develop and build new tools to tackle the problem, and finally even less to discover new possible mechanisms that might or might not allow us to improve our tools in the future. There's a lot of uncertainty in the future, and it's best not to bet a lot of money on it.