A fuel cell converts hydrogen+oxygen into water+electricty. Most fuel cells require very pure hydrogen, so you require a very expensive infrastructure to generate, clean, store, and transport the hydrogen.
Storing hydrogen is fraught with difficulties. Cyrogenic cooling takes a large amount of energy and the hydrogen boils away relatively quickly. Compressing takes less energy but the hydrogen still leaks through the tank... A tank of hydrogen also stores relatively little energy and is very heavy compared to the amount of energy it contains.
All in all it's simply impossible to get the same efficiency as an EV. You also need to use oil/gas to generate the hydrogen as splitting water using electricity is too inefficient. This is why I don't think hydrogen will ever become popular in the future.
The article ponts out that FEVs will represent a small percentage of auto sales, because the assumption is, I/m sure, that they require a hydrogen disrtribution infrastructure. It wouldn't be the case if they can run directly on gasoline, diesel, and/or biofuel. Still, it's really great to see the automokers working hard at developing the most promising EV technologies, rather than falling back on the tired old battery-powered EV so predictably.
Related to this, I hope people are also paying attention to the U of Waterloo research on global warming. Another body of work that finally runs counter to the popular orthodoxy du jour.
NASA's Orion Flight Software Production Systems Manager Darrel G. Raines joins Planet Analog Editor Steve Taranovich and Embedded.com Editor Max Maxfield to talk about embedded flight software used on the Mars on EE Times Radio. Live radio show and live chat. Get your questions ready.