Congratulations to Armadillo Aerospace for winning the Level One Lunar Lancer Challenge. John Carmack has a full report.

Bucky paper looks like ordinary paper but is made with carbon nanotubes. It potentially has up to 500 times the strength of steel at one tenth the weight, has good thermal conductivity, and can be an electrical conductor or semiconductor. Initial uses will likely exploit it's light weight and conductive properties in applications like electromagnetic interference shielding for aircraft and lightweight electrodes for batteries and capacitors, but we could eventually see paper airplanes, cars, and motorcycles made from the stuff.

Nassim Taleb has an essay exploring the limits of statistics over at Edge. It's about the contribution to the meltdown of the financial system by bankers and regulators putting too much faith in statistical models of risk. He emphasizes the importance of knowing what we don't know, and planning accordingly.

Taleb constructs a map which divides decision making into for quadrants in a two dimensional space. One axis divides randomness into two categories and the second divides outcomes into simple and complex.

He calls the first type of randomness Mediocristan and can be described by well defined mathematically with a strong tendency toward a particular value such as the Gaussian (or normal) distribution and exponential distribution. The second type of randomness, which he calls Extremistan, is less amenable to static analysis and may be described by such things as power law (or fractal) distributions and are likely to be sensitive to initial conditions and detailed history. The first type of randomness is dominated by typical values, while the second is dominated by extreme values.

The first quadrant describes games of chance and is a predictable type of randomness that provides the livelihood for casino operators. The third quadrant represents real world domains such as quality control where statistics is generally useful, though some inconvenient "outliers" may have to be ignored. The second quadrant doesn't seem to have much real applicability, though there may be some examples in decision theory literature. The forth quadrant is where statistics breaks down and we get into trouble when we rely upon it. Evolution, politics, economics, wars, etc. lie in the last quadrant.

The fourth quadrant is were we most want to assess risk, but are least able to predict it. As Will Rogers said: "It ain't what we don't know that gets us in trouble. It's what we know that ain't so."

I've been playing with Google Spreadsheets. Putting my Highland Games results in a Google spreadsheet is pretty easy and seems like a good way to share data. I'm still waiting for results from the Long's Peak games.

Free piston engines could potentially be used to build generators for hybrid vehicles that rival the efficiency of hydrogen fuel cells while begin multi-fuel capable. Drawbacks are immaturity, more noise than conventional internal combustion engines, and difficulty in controlling the free floating piston.

Greg Pfister is looking for killer apps for the wave of multi-core processors that is beginning to appear. I think the embedded computing space could soak up a lot of multi-core processors and a couple of start-ups are betting on it.

Intellasys is sampling a 40 core version of its SEAforth processor and Xmos has begun shipping chips based upon it's XCore processor and XC programming language. The SEAforth architecture is close to the simplest possible multi-core design. It's a mesh of stack machines programmed in a dialect of Forth. The XCore is a reincarnation of the Transputer microprocessor and Occam programming language. Both implement a CSP processing model.

It'll be interesting to see if either of these is a commercial success. SEAforth is simpler and offers more MIPS per buck and per watt, but the XCore has more RAM and looks like it would be easier to program.

A hydrogel will soon be used to assist in articular cartilage repair. It promises to make micro-fracture techniques more effective.

I competed in the RMSA Fall Fling Highland Games yesterday. I'm happy with my performance, improving in several events. My knees have continued to get better and I've had fun. The Highland Games atmosphere is competitive, but friendly.

I finally turned a caber, so I'm beginning to feel like a real heavy events athlete. Now, if I can just get the spinning down for the weight for distance events.

Jerry Pournelle's Exile -- and Glory was originally published in the 1970's as two books, a collection of short stories set starting about 2020 and a novel set in about 2050, all in the same "universe". It's interesting to see how well the projections of one of the premier hard science fiction writers fit the way things are playing out today. Computers have advance more, and space travel less, than he imagined. Highly recommended for aficiados of hard sci-fi.