Having spent thirty years working in the oil industry I can make some comments.
The price of oil (not bowser price) is determined by supply and demand, like just about anything else. The price in real terms averaged over about five to ten years has remained almost constant for the last 130 years or so (and the reserves available have been pretty constant at about 20 years over this time). It tends to have large spikes because demand is very inelastic in the short term, and so is supply - the time lag between a decision to explore and when the oil actually starts production can be decades, especially for gas (harder to transport). For example, the Scarborough giant gas field, off NW Australia, was discovered in the late seventies, but is only now being considered for development (it is in 100m of water 200km offshore).

The actual amount of oil reserves depends very much on the price of oil - as the price rises it becomes worth spending more to get more oil out, and this means that there is more oil in reserve. So there is no single figure for the reserves - it depends on the price of oil. Then there are the results of exploration, which again is very dependent on the price of oil, as well as the tax regime and the prevailing (or expected) interest rates. Exploration done today cannot be justified unless the risked rate of return after taxes and interest on the money spent on exploration is positive - and with long lag times and high interest until fairly recently, this amounts to a lot of money.

What we are seeing in oil prices is strong demand due mainly to massive increases in consumption by China and to a lesser extent India, against a background of fifteen years or more of depressed exploration due to low oil prices, high taxation and high expected interest rates. The price has been held low mainly because of massive overcapacity in production capability by Saudi Arabia (note that there are two separate questions - production capacity and reserves)

However, the price of oil has now risen, at least in the short term, to where alternative biofuels are economic, and this will put a cap on oil prices to some extent - biofuel facilities, especially biodiesel, are cheap and quick to build, although sourcing feedstock in the medium term may cause problems.

The price also makes other alternative fuel sources economic, such as tar sands, coal conversion and gas to oil conversion (although CNG would seem more sensible). It also makes alternative energy sources such as nuclear and renewable look more attractive, and for these reasons, plus the reduction in demand due to increased price (do you really need to drive to the shops four times a day?) are likely to make the more extravagant oil price predictions look a bit silly.

Incidentally, if Australia had not adopted import parity for oil when it did, there would have been little more exploration in Australia, and a much larger proportion of oil would be imported today, and there would be no LNG gas export industry to pay for the imported oil.

Bowser price is another question. Although dependent on the price of crude oil, it is also controlled by supply and demand, and as others point out, it includes a significant tax component. (despite this, bowser prices in Australia are low compared to any comparable country except the USA). Refining in Australia, and for that matter most of the world, is almost a non profit business. As a result (of this plus the NIMBY factor for new refineries) the world (and Australia) is now looking at a real shortage of refining capacity, which may be why the refiners are now feeling they can increase their margins. And most of the world's refineries are getting to be past their use-by date, but cannot be readily replaced for the same reasons.

The wild variations in bowser price are mostly the result of fluctuation in demand - refinery throughput is pretty near constant, so any increase in demand means increases in price and any drop means a decrease as refiners try to even out sales.

Diesel used to be a byproduct of refining petrol, but increased demand, especially in Europe (transport and heating) and North America (heating) plus increasingly stringent specifications mean that it is no longer a byproduct, but costs at least as much as petrol to refine and on a worldwide basis demand exceeds supply.

John