John Robb's Weblog

Airbus

There's some interesting information filtering out from the Air France Airbus crash. Essentially, when the pitot tubes (the device that senses airspeed on aircraft) fails (sometimes due to ice) on the Airbus, the aircraft's automated flight system goes haywire. Airbus aircraft are known for "glass cockpits" (all computer displays) and highly automated flight controls/autopilots (the pilots only indicate intention with inputs to their controls, the computer flies the plane). So, when the airspeed indicator fails, it can trigger the plane's automated flight system to fly differently (which can crash the plane given that the 'flight mode' selected isn't appropriate for the actual conditions) or it can cause the flight system to crash/reboot (there are indications from radio transmissions that this was going on). Regardless, we can learn from this that highly complex systems can be driven into catastrophic failure when the assumptions change (i.e. that sensor inputs are always correct), reactive systems that respond to minor failures can often lead to catastrophic failure (i.e. this is similar to the systems that guard against local power failures on the electricity grid, which can lead to catastrophic system failure as attempts to "save" the system from minor breakdowns overwhelms the entire system), tight coupling (the speed of event onset overwhelms human control), etc. There's lots more.

Decision making and Wealth

The Financial Times: Argentina. The superpower that never was by Allen Beattie. An excellent extract from a new book on economic history. The extract compares the US and Argentina, which although they started at nearly similar places in the 1800s, ended up with vastly different outcomes.

On the face of it the economies of the two countries also looked similar: agrarian nations pushing settlement westwards into a wilderness of temperate grasslands. In both nations, the frontier rancher – the gaucho and the cowboy – was elevated into a national symbol of courage and independence. But there were big disparities in the way this happened. America chose a path that parcelled out new land to individuals and families; Argentina delivered it into the hands of a few rich landowners.

It's been my belief that economic vitality, the ability to generate maximal growth in wealth over long periods of time, is a function of the degree two which ownership over the capacity to produce things is dispersed. The more decentralized ownership over productive assets is, the faster wealth grows. It's a function of decentralized decision making (something I call, "open decision making").  

In our wage driven society, where productive assets are concentrated, we've attempt to game this. We've focused on distributing the ownership of non-productive assets (homes or the financial equivalent in pensions/retirement savings), instead of productive assets. That approach is proving to be a failure. It is too easily gamed by financial elites, who use debt to assume ownership and corporate ownership to freeze/reduce wages. It also forces economic sclerosis since decision making on how to allocate capital and improve its application is centralized in the hands of a few.