Jean Laherrere is one of the geologists who's been talking about Peak Oil for a while, including this Scientific American article back from 1998. Here's a new country by country analysis of his, suggesting a total liquids peak in 2010. Since there are other liquids that can substitute for crude oil, total liquids may be a better way to look at the peak rather than purely crude oil.
"Annual liquids is modeled with such ultimate and will peak if there is no demand constraint (unlikely) around 2010, but if there is some constraint from the demand (likely depression) the peak will be a bumpy plateau and oil price will be chaotic."
"Demand constraint and exploration cycles lead to several peaks or a bumpy plateau."
"Forecasts of how supply may meet demand need to cover all liquids demands including crude oil, extra-heavy oils, natural liquids as well as refinery gains as synthetic oil as GTL, CTL, and BTL."
GTL = gas to liquid
CTL = coal to liquid
BTL = bitumen to liquid? (tar sands, oil shale)