Energy Storage is EASY

I was astounded by another article I saw this morning talking about the problem of storing energy that photovoltaics, wind-power, etc. all face. The article went on to talk about truly massive (city-sized) electrical batteries, ultracapacitors (which'll be great for cars) and other devices. For small-scale use, the problem remains, but for broad-scale power production where energy is only produced during the day, but the amounts of energy being produced are massive, why hasn't anyone thought of digging a big basement and almost filling it with a massive concrete block? When you need to store power, you lift the block up to ground level, and when you want to draw power, you lower it. If you want your power I/O to be very sharp-edged, you use a simple pulley (or massive rotating beam serving as a pulley); multiple pulleys can be rigged for any level of I/O you want. Now, granted, conversion back and forth from mechanical to electrical energy will have losses, but given the non-linear response to compressing a gas (not to mention keeping a pressurized cavern airtight), friction losses from flywheels, etc., it can't be that bad.

But what really gets me is that no-one's mentioned it before even though potential energy is part of high-school physics -- could it be that hard? Really?