An engineering culture that breaks APIs is bullshit. Taking 48+ hours to fix show-stopper platform errors is bullshit. Their 'innovative approach' is not working and the proof is in the pudding. They have not reached 500 million users solely on the success of this engineering approach, and I would vehemently argue they have reached 500 million users in spite of it.
The price will become too high to pay, if the future they envision involves third party developers at all. 500 million users is a hell of a lot of momentum, but it won't carry them as far as they want to go.
Saying that you think they have reached 500 million users inspite of their engineering approach is far from a common-sense argument. It's clear that Facebook ran technical circles around MySpace. And it's clear that they have launched new features at a breakneck pace throughout their history that have fueled growth at every turn.
I personally will avoid working with Facebook as much as I can, but will that hurt them? Not in the current business climate.
The reason is very closely tied to Facebook's engineering success: there is no top-down control over anything. All the engineers are encouraged to break things and to keep the pace of progress rapid. For developers on their API this is a nightmare scenario. However your claim that this is not sustainable may just be wishful thinking. Facebook craps all over its developers, but it's not clear at all that that price is too high to pay for the continued agility that Facebook exhibits. It's hard to argue with that kind of market dominance. Plus, Facebook is so nimble that if API stability does become a problem they can easily switch gears and focus on it.
In other words, I wouldn't hold your breath for any change at Facebook.