I agree that it's better to reduce complexity and that browsers have been subject to mission creep, but a.) I'm not really convinced that adding more integration points will reduce complexity and b.) the mission creep already happened - I don't think we get to reverse it. Creating an application platform _is_ the mission of a browser now. I think there would be value in an alternative such was just a simple document viewer and form filler, but it'd be a different kind of application and customers would need to be sold on it from scratch.

Is your position that no project needs to be large if we employ the right strategy, or that we should look at our complexity ceiling and say, "okay, that's how large a project we can take on?" Or something else?