This is the system that is the farthest along that I've seen in terms of solar.
http://www.stirlingenergy.com/whatisastirlingengine.htm
It's my understanding is that california is planning to deploy to a massive array of these for multi-megawatt power production.
Powering your house off rooftop solar costs ~$7000 and used to be tax-deductable. Some energy companies pay you for excess energy production from your solar array, some don't. Here in Austin, TX - the municipal energy company rebates 75% of the cost if you hook your solar into the grid, so they get the excess energy. It's like a mini power plant for them. Without rebates or tax deductions, solar can pay itself off within 5 years or so.
Regarding geography, solar energy is effective at some level in most areas outside of the arctic circles. Near the coasts, you can expect increased production because of solar reflection off the ocean. Decreased power plant emissions increase solar emissions.
I think the stigma of solar energy is one of its ownly problems. Energy production doesn't have to be the responsibility of a large organization. Solar energy has been a worthwhile home investment since the mid-1980's, it just hasn't been a state/corporate solution and won't be for another 10 years.
The best way to gather solar energy, of course, is in space. A massive array could gather the energy and beam it down via microwaves to massive collection areas on the planet. It would be very effective - 40-50 years off though, and the collection arrays would have to be isolated.
Inevitably, if we try everything and it all generates power, that's good. When it comes to energy production, you don't need a unified solution. Nuclear, Wind, Solar, Biofuels - let's do it all. We can sort it out once we've ditched the coal plants and it's safe to eat fish again.
More info:
http://en.wikipedia.org/wiki/Solar_energy
The best way to gather solar energy, of course, is in space. A massive array could gather the energy and beam it down via microwaves to massive collection areas on the planet. It would be very effective - 40-50 years off though, and the collection arrays would have to be isolated.
I have read that proposal when I worked for Spectrolab back in the early 80's. Very interesting stuff but it dosen't address what I think is the core issue with energy generation. Supplying energy to the masses from a central generating source of some kind/any kind vs point of use generation. The only true energy independance will be when you generate what you need from where you are. Anything else will allow somebody else to control haow much and where it will go. I like the point of use idea myself.