This is far from true. Example: Given the amount of resources Microsoft has and all the effort they put in UX it seems reasonable they have made the best UX decisions
No, that's provably not true.
It seems logical that the large resources of big companies would work that way but that's not how the real world works. It always comes down to just a few people and their taste. Rarely are these kinds of decisions made based on anything more than a few people's opinions whether it's Microsoft, Google, Apple or Facebook. Whoever is the lead programmer for a particular project or lead UX designer for another or one of the 2 or 3 people around them decide this stuff based on nothing more than their professional opinion. That opinion might be more informed than your average joe but it has nothing to do with the size of the company and their "considerable resources".
I think the biggest problem with HTML5 is development obstacles, and mobile browsers being under-powered.
Think of it this way: creating a fluid GUI can require AJAX, and a number of other tricks to streamline loading. But a native app can load the whole GUI at once instantly. To make a fast and responsive native app is hard NOT to do, while doing the same in HTML5 is HARD to do.
That being said, the example such as Microsoft does ring true, and perhaps Facebook didn't give HTML5 a fair try. I still think that overall evidence has been pretty clear that native can still provide better performance.
I'm not sure we can mark this as a win for HTML5, but what Sencha did is impressive. HTML5 will certainly continue to grow, and this is a fantastic step forward.
You are also correct that HTML5 is cross platform, and Sencha has proven that HTML5 can be taken into serious consideration when starting a new project.