What is Fedora?

If stability is the number one reason people switch from Fedora to another distro, then why isn't the solution a more robust and thorough automated testing process for admission of updates into Fedora?

As a software developer, I understand the complexity of testing an entire ecosystem, but it's not really an unsolvable problem. If it means we need to put pressure upstream on developers to produce more unit-tests that are exercised as a part of a holistic testing approach, whenever a new release of a package is considered for inclusion... well, then that's a discussion that ought to happen.

The problem isn't "defining the target user". The problem is that Fedora eats itself sometimes, and yum update will crater your day. That is a fixable problem. But the fix is to avoid releasing broken software, not "marketing" spin.