Also, don't discount wind that fast. The typical "it will have to whole country with these 1-2 MW turbines" line of argument is old. Modern offshore turbine designs are 6MW, getting to 10MW by 2012 and 20MW designs are in the works now. Advances in the field have been massive just in under a decade, with higher output efficiency (gearless designs) and less overall maintenance effort.
It takes time for the other generators to change output levels, so during high speed wind turbines tend to get disconnect from the grid because of the harmful effects they have on it.
The net result is that they don't produce out as much energy as expected.
Traditionally, a wind turbine generator works only in narrow range of speeds (say, 2000 rpm). In older designs the variable speed of wind was addressed by using a (massive) gearbox, not unlike what you have in your car. The newer designs (e.g. what Siemens Windpower is churning out now) are gearless, don't suffer from that problem, and have in ballpark of 40% better energy efficiency.
The cap/penetration effects of wind power on the grid are being studied, but on levels below 1/5th of total consumption it is believed to not cause any substantial problems. Most of the existing grids can accommodate that. I think no country approached the 20% wind power penetration yet.
Now imagine that the outlets for charging those batteries were connected to a wind-powered grid, where the power company charged a lower rate in exchange for less reliable availability. That kind of service would suck for running, say, an Internet data center, but it would be adequate to make sure that all the batteries that came in by 6:00 pm were fully charged by 6:00 am.
Voilà! Two problems solving each other.
Yes, we need a smarter grid, we need a breakthrough in battery technology, etc., etc. The way forward is going to be hard and take serious investment in technology, but what else is new?
It seems like so many people want everyone else to be clean while being dirty themselves.
Well, recycling might use too much energy to be clean (depending on how we get the energy), but using too much water doesn't seem a useful complaint, since cleaning and reusing water is just a matter of more energy.
I'm skeptical of this figure, but it's not particularly relevant. The issue is whether directing energy towards renewable energy production has a better environmental impact vs directing that same amount of energy towards fossil fuel energy production.
Total evaluations of the costs (in dollars, to the environment, etc.) are extremely important in comparing energy production methods, but your 20 year figure is fairly meaningless on its own.
Solar? Not clean at all. (For photovoltaic which is what most people mean. Solar thermal is good though.)
And just because they are made from plastic means nothing. Plastic is cleaner than wood you know (i.e. uses less energy to make and use).
There is a fact that no one wants to acknowledge: In timeframes of under 20 or so years there are NO energy technologies that are clean. None. For timeframes longer than 20 there is exactly one: Nuclear. There is nothing else.
A solar cell will probably fail before it manages to return more energy than it took to make (it's hard to say for sure though how long they will last, none have been around long enough yet). Wind can never be more than a minor part of the energy supply. Hydro is fully tapped out. Geothermal causes earthquakes. What else is there? Maybe solar thermal, but that requires a huge amount of electrical wires (the energy source is very far from the consumption point), too much apparently since no one is doing it except for some tests.