Also, I wonder about the browsers that come with game consoles like the Wii, are they any good? They might be easier to use than full-fledged computers?
This is a good observation. Anyway, the issue here is, whether that massive number will have some common notion of what the simpler computer actually is.
So the problem of "simpler browsers" will likely only be solved by solutions that redefine the experience and goal and it will likely be very close to what we now call a smartphone.
For those of us that are computer literate, it's hard to imagine that even basic abstract concepts like email are completely foreign to the majority of the population. I would guess that over half of people don't know how to answer the question, "what do you use for a web browser?".
So you have a blank slate, a free pass to design anything you can imagine. If you can deliver it to these people, and if they don't find it intimidating, or hard to learn, or unpredictable, then they'll like it.
Let's take the "simplified browsing" problem, for example. I worked phone technical support for an internet service provider for a while, and it wasn't long before I saw the need for exactly such a thing. But, it runs deeper than just web browsing and email: there is a huge, absolutely massive number of people out there that need a simpler computer. They don't understand things like firewalls and security, and aren't inclined to ever understand them.
The solution? Take a Linux distribution and hack it heavily; simplify the desktop layout, hide all the settings, and set it up so that immediately after booting, a full-screen web browser appears. The web browser defaults to a very simplified portal page; the user logs in to their "computer" exactly once (on the portal page), and from there they have access to simplified email (which, for example, doesn't have things called "Reply to All"), chat, word processing, and other services.
Thing is, the previous services that have tried to create such a thing have done it wrong; they tried to release their own computer, hardware and all. That doesn't work, at least not now. This would work though because you could sell a CD which would make the installation process a breeze. You wouldn't have a huge initial development cost for the operating system; a good Linux hacker could probably make the necessary changes very quickly.
After, say, 6 months of development by a few people, and with the aid of a crack marketing team, you could start distributing copies of this thing for around $75. You'd be tapping in to an under-served market not just of senior citizens but of every average family that's frustrated at using their computer.
You could even build in a secure remote desktop protocol for the operating system, and make tech support -- if you wanted to offer it -- the easiest it's ever been.
So, there ya go. Probably a hundred-million-dollar idea, with most of the framework. Lots more details, too many to list here.
Got time to build it?