And in biology (which is the context of the original post) you often need uninterrupted time, as you can't pause a live experiment if you get busy in your day job, no matter if it's bacteria or mice.
If we don't consider funding, we're effectively saying that science should be the domain of independently rich aristocrats living off of inherited wealth, as it often was a few centuries ago.
Given the existing model has brought us to a point where there are several fields in which almost no published studies can be successfully replicated, at this point I'm not even sure that it would be a bad development to go back to having the likes of Gauss and Leibniz running the show. Dependency on external funding and the pressure of publish-or-perish seems incredibly destructive to meaningful scientific progress.
The vast majority of academic scientists don't have access to particle accelerators, can't send probes into space, and so on. Regardless of who you are, there are always going to be experiments that are inaccessible to you.
