"Over the past decade... the number of annual major accidents has averaged around 15, with year-to-year fluctuations that reveal little or no pattern." - Patrick Smith

If aircraft accidents are independent of one another (e.g., there are no common defects in the aircraft) and if there is no underlying trend in the frequency of accidents, then the frequency distribution of these accidents is described by what statisticians call the Poisson probability distribution.

If the average number of annual major accidents is 15 and, again, there is no underlying trend in this frequency, then the probability that there will be 10 or fewer accidents in a given year is 12%. The probability that there will be 20 or more accidents in a given year is also 12%.

Patrick Smith reports that in the past decade the number of accidents was 11 in the best year and 24 in the worst year. This is consistent with the range of random variation described in the previous paragraph.

What this means is that random variation alone can account for wide fluctuations in the annual number of accidents. Or, more to the point, a relatively large change in the number of accidents from one year to the next is not a particularly good indicator that the safety of air travel has changed.

"One can't change basic human values. We are products of our evolutionary history, end of story."

At the risk of reigniting the nature versus nurture debate, it seems self-evident that the human brain has evolved to learn values beyond what I would call innate values. These learned values vary sufficiently from culture to culture to support John Roemer's idea that learned values can change within a culture.

Yes, communism failed for a reason. It was incompatible with the innate values of fairness and justice. The task for social engineers like Roemer is to design an "optimal" system that respects the constraints imposed by innate values, while treating learned values as design parameters.

"An addition of perhaps 6 million supplemental barrels per day from Iraq would make a striking difference in the energy equation. In fact, it might prove the difference between squeaking by and a catastrophic worldwide [oil] shortage." - Michael Klare

Professor Klare's article is a model of concision and clarity. But his talk about a catastrophic worldwide oil shortage doesn't address the economics of energy. As oil production peaks out, the resulting rise in the price of oil will make oil at $100/bbl seem cheap by comparison. This, in turn, will make alternative energy sources and conservation efforts much more attractive. And we should be weaning ourselves off of oil anyway to reduce CO2 emissions.

The key to avoiding an economic shock when oil production peaks out is to radically increase the pace of innovation right now. The governments of advanced industrialized nations should be massively subsidizing alternative and environmentally benign energy technologies and conservation strategies. As with universal health care, the U.S. lags behind other advanced countries and is perversely loath to learn from them. This speaks to political leadership and if we can't get action from the Obama administration, our future will indeed look bleak.

Andy Borowitz has a brilliant take on Goldman Sachs in his current column: http://www.borowitzreport.com/

Beyond her amusement value, Michele Bachman also plays a useful role for the Democratic Party. As Rush Limbaugh's hand-puppet in Congress, Bachman frequently reminds everyone just how wacky the wingnuts are.

"The [Canadian] government wants to can AECL's now old reactor design. So the bid is off the map. That will kill it. What Ontario will get is likely a Wesinthouse or or other competitor."

AECL's generation III CANDU reactor, the "Advanced CANDU Reactor", isn't expected to be in-service until 2016. So why would AECL essentially close up shop until then? Any why would Ontario take a step backwards and accept Westinghouse's problematic PWR reactor design?

The Toronto Star story contained this line: "Much of the dramatic price increase relates to the cost of labour and materials, which have skyrocketed over the past few years." These costs were skyrocketing because the global economy was booming prior to the economic meltdown that began last fall.

I suspect that the bids for Ontario's nuclear power plants were prepared when costs were at their peak. If so, then the Ontario government should consider asking the bidders to resubmit bids that better reflect costs under the current economic conditions.

So the Democrats are going to help those who Leona Helmsley infamously referred to as "little people" not because we "little people" deserve a break, but because the Democrats don't want to risk taking a political hit. This nation really needs a third political party, a party genuinely responsive to all of us "little people" out here.

It is obvious by now that Wall Street has rigged the financial and political systems to suck wealth out of the economy. There is a simple solution. Apply a federal tax on every single trade proportional to the value of the trade.

If computers can handle the trading, this would be no extra burden. The tax would discourage precisely the kind of trading that serves no useful economic purpose. And the tax would help compensate for the loss of federal revenue that the Wall Street debacle caused in the first place.

You asked why there aren't more movies about math. It is probably because of the requirement that movies attract enough of an audience to make money. One notable exception to this is "A Beautiful Mind" (2001). Other recent math-themed movies are "Proof" (2005) and "Pi" (1998).