As much as I love Linux and GNU, neither is especially innovative. Just about every FOSS tool was invented elsewhere in a commercial setting before it was clon

As much as I love Linux and GNU, neither is especially innovative. Just about every FOSS tool was invented elsewhere in a commercial setting before it was cloned or reinterpreted into open software. The question is whether novel products or services can arise among distributed cultures. In my 30+ years in this business (software R&D), I've seen lots of examples that it cannot and few/none that it can. Those few people who grok a new problem that needs a mousetrap when nobody else has ever seen a mouse, somehow they need to communicate this problem / opportunity to the scientists and engineers who must build that first trap. In my experience, this is a LOT easier done in person than remotely, especially if you don't already know your makers well. Until remote communication improves at least ten fold, addressing new problems remotely will remain vastly inferior to doing it locally.