While is may be true, the problem is extracting the oil is getting more difficult due to no new technological breakthroughs since the 90's. In the 90's I knew guys (fellow grad students working on their doctorates consulting for petroleum countries) who were applying multi-sensor DSP to find oil. Basically, you set a grid of sensors out, set off a shockwave in the ground (explosion, thump, etc.), measure the returns (time delays, return amplitude/phase). This data allows one to recontruct they layers of material underneath the ground without having to drill. Think of it as a geological radar or ultrasound. This technique aided in the discovery of easy-to-extract oil reserves. The "known" oil reserves (known and known to be extractable using today's technology) can't meet increasing global demands. That doesn't mean there won't be a breakthrough in technology that unlocks more reserves. You can extract the difficult-to-reach oil...it'll just cost a lot more money. $50/barrel is probably more of the floor now, and that assumes no terrorist activities that either damage oil infrastructure or set oil traders into a panic (think $80-100/barrel or $4+/gallon for gas). Hydrogen fuel cells sounds great in theory, but it takes a lot of energy to generate the hydrogen, and that energy's gotta come from somewhere, whether oil or nuke.