The challenge is that Google search is a self-perpetuating machine that is able to improve because of how well it’s collected the world’s data. It’s able to do this in ways competitors just can’t. They don’t have the data and they can’t gain that data any faster than Google can. So how does someone come along and make something better?
I think the only answer is the hardest one: someone needs to do to Google what Google did to AltaVista back when it was new…someone needs to completely change the game in what we expect from search. But again, Google has done well here not to give future innovators much to grab onto.
Google used to show you sites that had the information you need. Then they took that information (stealing it sometimes) and just showed it to you without the need to go anywhere else. They then made it so you could ask your questions in natural language. They’ll even sell you a cheap little puck to answer those questions around your house. What’s the next thing? I don’t know, but I would put good money on Google already exploring it and releasing it before anyone else gets a foothold in the market.
Google's biggest weakness is its willingness to throw up dozens of new ideas a year, let them run for a while, then lose interest in them and shut them down. That is why I steer clear of most of their services. You never know when they'll decide to shutter a popular product because they can't be arsed with it any more. (RIP, Google Reader.)
Google has turned into Microsoft. Microsoft has turned into IBM. IBM has turned into a fossil. Such is the march of progress, and no number of patents or acquisitions will stave it off forever.