Both have been hammered, but patience with these names is warranted.
The reason why the ETFs make more sense than the stocks here is that there is a great dispute over the best way to move forward. Does it involve the current mix of materials? Should the aim be to increase yields of cut costs? There are no clear answers yet.
The action is happening in labs around the world. The goal is to cut production costs below the 50 cents per watt needed to compete directly with grid energy in the United States. Current technology only allows direct market competition in a limited number of markets, like Hawaii and California, where utilities face so much solar power that they're trying to end "net metering," in which they agree to take however much power consumers offer them. Once the goal is met, that controversy, and its opportunities, will become a nationwide phenomenon.
The "traditional" solar cell based in polysilicon is getting new life from start-ups like Crystal Solar, which believes it can cut costs below 50 cents/watt with a direct gas to wafer technique it says cuts capital costs (and energy costs in production) by 50%. Rather than going to market directly it has signed Hanwha SolarOne (HSOL) of Korea as its manufacturing partner, aiming for production in 2014.
But U.S. companies like First Solar (FSLR) and General Electric (GE) don't use silicon, preferring a thin-film technology using rare earths. They're looking to new manufacturing techniques that can reduce costs as opposed to just increasing yields. There is also interest in less expensive manufacturing using microwaves.
Other thin film companies are looking to make solar ubiquitous, so it can not only go on shingles but on any usable surface.
Meanwhile, start-ups like Silevo are focusing on increased yield-per-square-meter with hybrid technologies that combine aspects of crystalline silicon and thin film techniques.
While the politics of solar shows an industry on the defensive, the economics of solar are slowly changing, and adjusting to that political reality. The industry now has a goal, 50 cents per watt, that it's determined to meet, and those who beat that goal by the most will win a growing market.
Disclosure: I am long GE.