Thankfully patent examiners understand there is a difference between obvious after the fact and obvious before the fact.
When no one was doing it, then suddenly everyone wants to be doing it, that's a pretty good example of something that was clearly not obvious before the fact and was after.
That's true of any patent. To anyone mechanically inclined, a huge percentage of mechanical patents (say, as an example, rack-and-pinion steering) are totally obvious once you've been shown there was a problem and have seen someone'e solution. It doesn't mean, a hundred years ago, that rack and pinion steering wasn't patentable -- because the examiners know if it was obvious and there were a hundred inventors looking at the problem, they'd be sitting on a hundred patent filings.
Multi-touch is an obvious solution to how you provide more complex gestural indications to a touch device ... now. But five years ago when there were gobs of touch applications in industry, and gobs of touchpads on laptops there were gobs of people looking at how to provide better gestures, and not one of them came up with that *even though the hardware supported it in many cases*.
That tells you something about the patentability of multi-touch. Apple released it and suddenly everyone was wanting to duplicate it on phones, touchpads and touchscreen computers.
Patents are made to cover exactly that situation -- where someone finds a solution to a problem that no one else has *especially* where its obvious after the fact (since the obviousness makes it easy to copy).