I'm a firm believer that all of this climate change thing is going to be solved by the application of human ingenuity. You might call me a Simonite on that point.I'm also absolutely certain that solar power is going to play a large part in that solution. There's just so damn much of it available that it would be near mad insane of us not to use it. At which point my insistence that we should not be subsidising the installation of current solar power is going to seem most odd. However, I refer you to Mike Munger [3]:

In 20 years, solar will be useful, and used. But it's a mistake to spend our money now on an immature and still not well-engineered solar generation system.

I'd argue on the 20 years: it's going to be much sooner than that. Solar power depends upon a variant of Moore's Law (in part, at least, and then further on the efficiency with which silicon metal can be made, something increasing by leaps and bounds as well) and it's getting more efficient faster than most realise. Or more productive perhaps, to bring the falling price of it into play.

The usual argument at this point is that since solar will become efficient at some point in hte near future then we've got to subsidise the installation of it right now. Which is absurd of course: that it will be grid comparable in general (rather than just in specific locations, as now) in the near future is exactly why we shouldn't be offering any subsidiy at all for installation of the current, not efficient, generation. And the closer that near future is the stronger the argument against subsidy. If the next generation of solar, available in, say, 2015, will be cheaper than coal (a claim some make although I'm not sure it will be that quick) then why one Earth would we waste money installing not efficient solar in 2014?

Save the money and install the efficient stuff in 2015. This is true whatever your timescale for solar becoming efficient is. The more anyone insists that it will become efficient the more they ought to be arguing against the subsidy of the installation of the current generation of inefficient solar.

Subsidy for development, for R&D work, that's different, with a different set of arguments. But subsidy for the current installation, for 25 years of subsidy through feed in tariffs, when we're all also arguing that unsubsidised efficient kit will be available in 2 or 5 years, is simply ridiculous. Wait and install the good stuff instead of littering the countryside with the current bad kit.