A better way of thinking of it is to compute the break-even time. Consider the Prius vs. Matrix comparison in the study. Even if you accept the 20,000km (12,000 mi) assumption and and the 1.17$/liter gas price, the difference in fuel cost accumulates year over year. According to the EPA, the Matrix has an average efficiency of roughly 30mpg (12.75km/liter) whereas the Prius is around 50mpg (20.28km/liter). Using the numbers in the study the Matrix annual fuel cost is 1.17 $/liter * 20,000 km / (12.75 km/liter) = $1835. The same calculation for the Prius gives $1154. This is a difference of $681. According to the study the 5 year difference in cost was only $1718. Therefore based only on the gas cost break even would be another 1718$/ 681$/year = 2.5 years.

In other words the break even time, using the numbers in the comparison, is roughly 7.5 years. The break even time drops considerably if you drive more than 20,000km per year. Add to this that the Prius is a much nicer car than the Matrix and you get the conclusion that if you keep your cars for a reasonably period of time, the Prius is MORE economical, and a better value.

> The awful bit was that he got delayed coming back, wasn't sure how far it was, went to the limit of his air, turned the little knob that gives you the last five minutes, and used that time scratching out a goodbye to his family on the air tank. I find the scenario as you describe really hard to believe. What you describe is called "J Valve" tank, which had a spring valve that would artificially stop the air flow at around 500psi. The idea was that you would use the tank until it got hard to breathe, then pull a lever that would let you use the last 500psi. It was a great idea in the early days before people started using pressure gauges. J-Valve tanks have been obsolete for years! I agree that cave diving is extremely dangerous, but no one in his right mind would do it with vintage gear. Unless, by a "few years" you mean 30 or so.

Deep diving is extremely dangerous and requires exotic blends of gases that vary with depth range. The deepest ocean dive that has ever been recorded was in 1988 to a depth of 534 meters. It was done in the Mediterranean by a team of divers from COMEX using a mixture of Hydrogen, Helium, and Oxygen. The gas mixture must balance the physiological effects of each gas component in terms of narcosis and metabolic needs. All gases, even "inert" ones have effects on the nervous system. The physical nature of these effects is largely unknown, but they likely stem from the differing solubility of each component in various parts of the body. There is some literature that makes the case that the narcotic effect of a gas depends on its lipid solubility with the suspected mechanism being that the dissolved gas molecules lodge in cell membranes and change their physical properties affecting neuro-transmission. The deepest "chamber dive" was to 701 meters which took 43 days to perform the decompression. You get more into at http://en.wikipedia.org/wiki/Saturation_diving and http://www.techdiver.ws/exotic_gases.shtml It is possible (likely?) that these records have been surpassed by the military, but as far as we know it is physically impossible for a diver to survive at the depth of Deepwater Horizon rig, let alone do useful work.

I have always wondered why people deride "teaching to the test". The real question should be how to structure the tests so that they measure what we want them to. Perhaps we should be teaching to Torrance's test?

it is clear who has really benefited: Lawyers. Seriously, the piracy industry keeps a large number of lawyers well fed, both prosecuting and defending.

When this system crashes... it REALLY crashes. Having a network control your accelerator and braking: what could possibly go wrong?

cobwebs on the brain.

Here are two that I haven't heard mentioned. 1-2-3 Infinity by George Gamow (ISBN-10: 0486256642). The physics is a little dated, but the math is elegant, engaging, and accessible. I read it when I was a child and it was one of the reasons I went into physics. It is easily enjoyed by any motivated high school student. Another good one is The New Turing Omnibus: Sixty-Six Excursions in Computer Science" by A. K. Dewdney (ISBN-10: 0805071660). This is a little more advanced, but most of it is well within the abilities of a good high school student. The nice thing is that it is written as a series of vignettes of applying math to compute science. If one is a little too advanced, just skip it and go on to the next.