The Economist has a great article on the history and near-future outlook for battery technology. Batteries have scarcely improved in the last century, and there have been too many false starts for a seasoned observer to get his hopes up too easily. But this chart of battery capacity by unit weight, in particular, is something I have been seeking for a long time. It vindicates my belief that lithium-ion technology is improving at a rate far faster than traditional nickel batteries (that have scarcely improved at all in the last half-century). Note, importantly, that if we join the multiple curves, we see a strong indication of the classic accelerating technology exponential curve. This time we know it's for real.
This is exciting on multiple levels, because it opens to door to not just mainsteam electical vehicles in the next decade, but to a variety of wearable electronic devices, 20-30 hour laptop batteries, household robotics, and other applications that have not yet been imagined.
Future projections are usually over-optimistic, you say? Let's also not forget Stanford University's nanowire research to increase Lithium-ion battery capacity, which was wide acclaimed as among the most important scientific breakthroughs of 2007.