A lot of research has gone into flywheels over the last 40 years or so. The best that came out of that was ride-trough systems for hospitals and other large consumers (datacenters!) which catch the load until the generators can be spun up and power stabilization for wind parks (but these have now been supplanted by more solid state tech using superconductors).
There are all kinds of reasons for that much too involved to go into here but for most practical (and cost efficient) applications fly-wheels are probably not going to happen unless there is a breakthrough in materials that would allow much higher RPM.
Other interesting applications where the gyro bus (http://en.wikipedia.org/wiki/Gyrobus ) and a similar system used to power regular cars (the flywheels were in the 100K+ RPM region and the size of a large dewar flask).
Well, they just don't seem as science fictiony as superconducting batteries. Coal and oil seem like great ways to store energy until you need it. Turning electricity into coal seems pretty tricky though.
If the cost of panels keeps dropping like this, i think a lot more money will go into electricity storage. no idea what that storage system might be though. with home generation, batteries aren't as encumbered by size and weight when compared to a car or a laptop. Maybe some big heavy battery tech will come along and give us a good chemical storage answer.
Maybe in the future, power is generated on rooftops, and large centralized installations store the excess energy till night - replacing some fraction of our current coal/oil/gas/nuke generating capacity.
