Jeff Vail talks about Energy Return On Energy Invested (EROEI) in a detailed post
The basic idea is to account for the energy used to produce wind turbines, for instance, and determine if the energy they generate is more than the energy required to implement them (build, transport to location, and install). Obviously if it's equal or less than you shouldn't even be doing it. In fact, if it's not significantly more (by say a factor of 10) then maybe you'd be better off not doing it.
Vail points out that this calculation is very dependent on how wide you define the scope of implementing the technology. For instance, do you also need to add in the fuel needed to raise the rice to feed the worker that built the turbine blade? How about the infrastructure to educate and pay the engineer that designed it? He concludes that it is impossible to accurately account for this, and suggests a different approach, based on price of the technology in the market. He intends to evaluate Wind and Solar power using this methodology in upcoming posts.
The point to take home here is that if the energy alternatives we're considering do not have much of an EROEI ratio then we should not be wasting our time on them. If there is no alternative with a good EROEI then we should start figuring out how to get by on less energy. I think there are reasons why we should greatly lower our energy use even if a good alternative energy source exists; this is a topic for a different post.