The main thing to consider is that, until revolutionnary discoveries in wind and solar, their power output is just too ridiculous (and expensive) to be considered as a large scale energy production.
Untrue, both have entered an exponential growth curve and are going to be major players by 2020.
For example, a solar pannel wouldn't be rentable (in money) over the course of his life if it wasn't subsidized.
People wouldn't pay for oil and coal if they weren't subsidized far more than they wouldn't do so for renewables. Oil and Coal get 60 billion a year in subsidized costs in the US, all renewables get 4 billion a year. Anyway, subsidization is not a bad thing if it works out well.
As said, there's also the fact that they can't deliver a constant flow of energy, which makes them mostly useless for network. And there is no way to store large amounts of energy.
Wrong and wrong. An electric network needs both a constant baseline and a temporary spike in order to deal with fluctuating power demands. And as it just so happens, the times when solar produces the most power is the time when people use the most power. You can store large amounts of energy, not just through capacitors, but also through gravitational energy storage (using the power to move a volume of water and then releasing it onto a hydroelectric generator with a weight). The latter is already in wide use.
Wind power in particular is also not susceptible to the "but it isn't constant" argument, as the best locations for it are ones in which wind is effectively constant (ridgelines, shorelines, ect). Solar is more vulnerable to that argument, but this is offset by it being able to generate power in the morning (when it isn't being greatly used and thus can be diverted to a storage mechanism for later). Newer solar panels are also capable of generating power during cloudy weather and in moonlight.