I remember at Uni we had a massive and complex assignment due, and many of the people in our class had been out of academia for a long time. Many didn't know whether or not they'd get the assignmment done up to scratch, and they just weren't bothering, which went right against their work ethic. All sorts of problems ensued. People were depressed, considering dropping out, and so on. Nothing was being done.
The lecturer got up two weeks before the due date, and told us this--"If it's worth doing, it's worth doing badly."
In other words, a low 'D' is better than an 'F-no submit' especially when resubmissions were allowed.
Everyone did what they could, and all passed first time. None of it was perfect, but that wasn't required.
Linus does the equivalent of this with his 'release early-release often' philosophy. It's better to have something out there that can be improved by the community than hidden away waiting for improvements that will never come. As long as 'good enough' will still be improved, I think there's room for both.