I wonder what kind of cost they're using, though. Is it marginal cost (on which solar already wins) or return on capital, and does the cost of capital include the cost of research?

The only way I can think of actually trying this is to sell future contracts for, say 1% of Google's alt-power production in 2008, 2009, etc., and to declare the project a success iff the value of 1% of the power, indefinitely, X 100, is less than the amount invested in the project.

Solar doesn't win on marginal cost. What are you talking about? The marginal silicon based photovoltaic system has higher $/W than marginal increases in capacity of coal burning plants.

Why sell futures? Can't they sell energy or at least shares of companies they are investing in?

Marginal cost as in the cost of generating X+1 Watts versus X Watts. With coal, you dig for more coal; with solar, you wait for more sunlight. The systems (coal plants, photovoltaic systems) are capital costs.

Then to answer your question, people look to capital costs for the cost of energy. Considering the rate of expansion of energy demand, it is the only number that matters.

> People whine and whine about responsibility for energy, while everyone chooses the cheapest energy available. This is a feature not a bug

I'm not sure I get what you mean. Is it that self-interest leading people to buy the cheapest energy form, environment be damned, is the way to go? What if we find out that the cheapest energy form is polluting just as much (or more) as coal? We can't escape some form of collective responsibility.

An externality is when a choice has an impact outside those making an exchange that isn't factored into the price. You can't expect people to rationally account for often unknown externalities.

Hope that alternatives that are expensive will win out of shear good will is a losing option because it requires everyone to be aware of the externalities of their choices.

Regulation to deal with externalities can work, but my faith in the competence of government to do well in regulating energy is close to nil, given past performance.

The only solution then is an engineering solution. Make alternatives that people would choose without knowing the externalities.

One side benefit is that you don't need the government and rhetoric about "collective responsibility" that too often approaches totalitarianism when enacted.