To make a successful open-source project is not easy and is about balancing a range of factors, including balancing their work/life. Not everyone can do this, or wants to do this. I think rock stars earn their status.
Yeah, that's true, that they invest a lot of personal time in this stuff, and they do a very good job in balancing, but the problem is that people trust them in important questions like performance even despite the fact many of them don't know CS basics properly (like data structures, etc).
I don't want to say all of them are this way, there are some well-educated people who definitely know a lot and deserve all of this. But as it was mentioned earlier, a lot of them are just 22 years old hip guys who released something what became mainstream/were involved in it and became ambassadors.
I think this is part of the tradeoff that they excel at. There is a huge difference between perfecting and optimizing vs. shipping and growing an ecosystem.
I think that's mostly bunk. For every rockstar there's 100 equally talented individuals who will never get recognition because they were not in the right place at the right time.
Rockstars are incredibly lucky.
IMO, social media and blogosphere driven complex work cannot be deep, just confused.
Any kid with a new idea can have his ideas listened to. This means we don't get stuck.
I would claim that any technical undertaking without understood goals using a process whose main design feature is not efficiency but sluggishness is a pathological one.
A computer system is far more simpler than any countrys political system - it's tractable and optimal solutions can be found. Unless, of course, unless the design process is lost in confusion and bafflement.
I would claim a place where any kid is listened to is precisely the one where everyone will get stuck. If people spend more time reading blogs rather than on CS basics they cannot understand a great ideas because they are on the Kruger-Dunning level of skill development.