The rats always find a way in

I have a theory. In spite of all the moral and idealogical bullshit about different forms of government, there's one determining factor that leads to their eventual failure. That factor is susceptibility to corruption. Some systems are more resistant than others, but all nations get sucked down sooner or later. It's just a question of how long it takes.

• How real-world corruption works

I've got a fat finger for ya, Wall Street

Just when you thought it couldn't get any weirder out there...

The DOW dropped almost 10% in 7 minutes. Some stocks even went to zero. For a while they were trying to blame a "fat finger event" in which someone typed "billion" rather than million into a sell order. Another theory is that high-speed trading got out of control -- the rise of the machines scenario. Positive feedback, anyone?

This bit from Tech-Ticker Trader Goes Bonkers is priceless.

I can picture it clearly. A Goldman executive in immaculate pinstripes saying, "This is the new HAL 9000. It's going to do all our trading for us based on precise mathematical models at millisecond speeds. What could possibly go wrong?" Later, the same executive, pleading with the computer, "Stop! Stop! The market's plunging!" HAL 9000 calmly says, "I'm sorry, Lloyd, I'm afraid I can't do that." And, then becoming frantic and agitated, "Tell Ben Bernanke to kiss my silicon ass! Fuck your pod bay doors!! Suck my Extended Binary Coded Decimal Interchange Code! Your shit's going to ZERO, BITCHES!!!!"

Of course, the conspiracy theorists are going wild. But, if you did manage to snap up a chunk of blue chip at a penny a share, don't look now. They're canceling orders deemed by the exchanges to be "clearly erroneous". Fuckers.

• Washington's War With Wall St. to Blame for Panic Selling

Update 5/13/2011: Donald MacKenzie writes in the London Review of Books in How to make money in microseconds: (edited)

The bulk of the research also suggests that automated trading makes the buying and selling of shares cheaper and usually easier. What needs weighing against this, however, are the implications of one strange and disturbing episode that lasted a mere 20 minutes on the afternoon of 6 May 2010, beginning around 2.40 p.m. The prices of US shares fell by about 6 per cent in around five minutes, a fall of almost unprecedented rapidity, then recovered almost as quickly. Shares in the global consultancy Accenture dropped to a single cent. Sotheby’s suddenly jumped to $99,999.99.

With the rise of electronic trading, the stock market (especially in the US) has become a tightly coupled complex system. Systems that are both tightly coupled and highly complex, organisational sociologist Charles Perrow argues in Normal Accidents (1984), are inherently dangerous.

The rest of the article is a fascinating look at programmed trading, concluding that the "flash crash" was indeed a case of computers run amok.

Update February, 2012

Looking at millisecond scale market data, researchers found 18,520 sub-950-millisecond crashes and spikes. According to a Wired article about the study, Nanosecond Trading Could Make Markets Go Haywire. The research, led by Neil Johnson, a complex systems specialist at the University of Miami, and simulation engineer Brian Tivnan of the University of Vermont, is published under the wonderful title, Financial black swans driven by ultrafast machine ecology. Johnson says:

“There’s this whole world below 650 milliseconds. It’s like landing on another planet,”

“We are certainly witnessing one of the major transitions in the history of financial markets,” said automated trading researcher John Cartlidge of the University of Bristol, who was not involved in new study. “Economic theory has always lagged behind economic reality, but now the speed of technological change is widening that gap at an exponential rate. The scary result of this is that we now live in a world dominated by a global financial market of which we have virtually no sound theoretical understanding.”

Book review: The Tipping Point

Malcolm Gladwell's book The Tipping Point (2000) is mostly about viral marketing and social networks. I'm surprised I read a book about such things, but it wasn't bad. It's written to be easy and popular rather than rigorous, so if you're familiar at all with the theory of networks, then there might not be a lot of truly new information here. The charming anecdotes that illustrate the ideas are the reason for its popularity.

I found myself recalling, with dread, my high-school years; navigating various social cliques, trying to emulate the cool kids, then shortly there-after consciously trying to be as opposite as I could. There's a lot of ugly stuff wired into our instincts as a tribal animal. But, like it or not, that's how we are.

MG identifies three special types of nodes in these social networks. Most obviously, there are connectors, people who seem to know everybody. They are the classic highly connected nodes. Messages propagate quickly through social networks with the help of connectors. Mavens are experts in some area who act as information brokers. Salesmen are charismatic persuaders. Due to the special properties of these personality types, MG derives the "Law of the Few", which boils down to this: Get connecters, mavens and salesmen on your side and the sheep will follow. Well, OK, he doesn't come right out and say, "Sheep".

• connectors, highly connected people
• mavens, people who are especially rich in information and who act as facilitators
• salesmen, persuaders

Many of the folks we know and love in technology fit these personality types. The maven appeals most to me because they have a genuine interest in being helpful and in subverting the hype and finding the real deal. Most of us engineers are a little more cynical about the salesmen.

Messages themselves have properties that help them spread. Stickiness might come from the message itself or its packaging. Our psychology is such that we're much more receptive to ideas delivered in the right context, as exemplified by the broken-windows theory of law enforcement.

Most of MG's examples are from fashion and marketing. Two are about trendy shoes. DeeDee Gordon, a marketer that helped the Airwalk brand gain popularity in the 1990's sounds like the role model for William Gibson's cool-hunter Cayce Pollard from the novel Pattern Recognition. Now that I mention it, that's another book I might not have read if I knew how much it had to do with marketing, but a good one.

Just for the record, I rode a skateboard and never owned a pair of Airwalks, though I don't imagine I'm immune to manipulation. And it is manipulation because it's seeking to influence you for someone else's gain, rather than your own. That's the parasitic aspect of marketing.

MG tries to transcend marketing. He brings in suicide, smoking cessation, and cancer awareness. But, he keeps circling back. Even the one political theme he talks about, freeing tibet, worthy cause though it is, is sufficiently distant to be safe. It would be too bad if the main result of these ideas was more effective ways to sell trendy consumer crap to lemming-like mainstream losers. But, I don't mean to sound harsh. It's a good book. I just don't like marketing.

How about tapping our social networks and tribal instincts to spread messages about individual liberty, being a good person, working for justice, living beyond consumer junk-culture, or making the world a better place? Too bad, MG avoids any really revolutionary tipping points.