Coal is cheap. Anywhere between $10 and $55 PER TON. Okay? PER! TON!
Subject to conversion losses, you get around 2 kilowatt-hours per kilogram of coal; so a long ton of coal (a metric ton of 1,000 kg) is going to produce 20,000 kilowatt-hours of electricity, more-or-less.
At $55 per ton, that's a fuel cost of $0.003 per kilowatt-hour.
Add all the overhead, including shipping and handling, mining, the power plant, employees to run it, etc, etc--and the power companies manage to make a profit while selling power for around $0.10-$0.15 per kilowatt-hour.
Renewable energy is not cheaper than coal.
It just isn't, and it's not going to be cheaper than coal for a long time.
How do you get power from renewable resources? You must first make machines to extract the power, and that costs something. In most cases, the machines cost more to construct, in aggregate, than the single power plant they would replace.
In most cases, the machines also are subject to the whims of weather--particularly in the cases of solar and wind power. You can't just flip a switch and have your solar panel start generating electricity; the sun must be shining on it. Wind must be blowing past your windmill and it must be enough to turn it at or above a certain minimum speed.
Coal, on the other hand, is relatively simple to control: you need more power, you put more coal into the boiler--subject to certain limits, of course. You can start and stop a coal-fired plant as needed. It's a little harder to make the sun shine and the wind blow to order.
And making solar panels is dirty. You use a lot of toxic chemicals to make solar panels; making enough solar panels to satisfy even a major fraction of the United States' current power needs would result in a serious environmental crisis wherever the panels were made. You can't make them without generating some toxic waste, and if your goal is to be "eco-friendly" you have to consider that--and the costs of dealing with that waste will inevitably increase the cost of making the solar panels. (And the cleaner you make your manufacturing plant, the more it will cost. Guaranteed.)
Coal: we know coal power is dirty. But we can control most of the pollutants.
* * *
Google's idea is to present us with ways to make power that are cheaper and cleaner than coal. But I don't think it's possible; "renewable" resources simply cost too much to exploit, both in terms of dollars and "hidden costs".
Of course, the Ars Technica article I linked above uncritically mentions the "social costs" of "climate change" and "pollution". "Climate change" is a bugaboo; pollution--if you want power, you're going to have pollution, regardless of what method is used to generate it.
Wind power? You think wind power won't cause environmental problems? Assume the issue of the variability of wind is solved, however--what do you think happens when you start extracting a lot of power from the movement of the atmosphere? Guess what--climate change!
Put windmills on a hill: wind blowing across those hills has some of its energy of motion extracted by the windmills. The air doesn't move as far, or maybe as fast, as it otherwise would; will this have any effect on the environment downwind of the windmills? How about 50 years down the road? 100? Put up enough of them and you'll make the change obvious in a handful of years. And windmills kill birds. What will you do about that?
Solar panels--making them makes pollution. They degrade over time, so you'll have to replace them, and they can't be made more than about 20% efficient at best right now, so you'll need a lot of solar panels to make as much power as a 10 megawatt coal plant.
Let's see, a kilowatt per square meter times 0.2 is 200 watts per square meter. 10 megawatts divided by 200 watts per square meter makes 50,000 square meters--a square 223 meters on a side.
Assume that a square meter of solar cells costs $1000. (Maybe it will, maybe it won't; certainly if you want to buy a solar panel that big for your house, $1000 is cheap.) That's $50,000,000 worth of solar cells. And it only makes power when the sun is shining. And when there are no clouds in the way. Clouds attenuate sunlight, reducing the power output of the solar cells; so for cloudy days you need even more solar cells to make up for the difference. Let's say a total of 100,000 square meters of solar cells will do the job; that's $100,000,000 worth of solar cells.
10 megawatts will supply a small city. Maybe 100,000 people, maybe a bit more. Or maybe I'm overly optimistic--but let's say I'm not.
There are 290,000,000 people in the United States. The annual power usage of the USA comes to around 28 terawatt-hours; to supply that energy with solar power would take 140,000,000,000 square meters of solar panels--that would make a square almost 375 kilometers on a side.
Hell, cover the state of Arizona with solar panels. That ought to do it.
And it would cost $140,000,000,000,000.
...actually, it wouldn't, because that level of production would ensure that solar panels were dirt cheap. But the "environmental consequences" of such a move would make the current flap about "climate change" look like a birthday party. Not to mention what it would do to our economy. Maybe if you factor in the opportunity costs of spending the entire GDP for the next 14 years on making solar panels, then it actually would cost that much--so I think my estimate is pretty reasonable, actually.
My point is that Google's plan is pie-in-the-sky. It's nice, and I'm sure it makes them all feel good, but it's not doing to do jack squat to fix the problems they're trying to address.
But, you know, we do actually have a renewable power source--right now!--that we could be using to generate power cheaply, cleanly, and with little waste left over.
This is it. All we need is for the eco-loonies and the Luddites to get the fuck out of the way.