Art is memes

An article in the NY Times criticizes the buyers of Edvard Munch's "The Scream" who spent $120 Million for  the iconic painting (they could have purchased a slide sharing website instead!).  The painting is notable for it's popularity, but it is popular not because it is especially beautiful or deft. I believe the paintings immense popularity and value are derived from it's power as a meme.

Miniature Cold War.

The brinksmanship of the cold war was based on the idea that, in a game of chicken, it was best to make the other side think you are crazy.  Even if you had absolutely no plans to nuke the Russians, it's in your best interest to make them think you really want to. This game show gives a pretty good demonstration (most likely staged) of this principle.

Thieves and Lawmen

I had the lights stolen off my bike a few months ago in San Francisco. The price of replacing them compared favorably with a parking ticket, so I was not upset. How could I be too frustrated, leaving my bike out like that in front of a karaoke bar late at night. I had it coming.

When I was a child that's how violence and theft were justified. If someone was beaten, or shoes were lifted from a store, there was always a ready reason that the victim deserved it.  I had violated the code myself by leaving the lights on my bike, the thief was just reminding me.

In this way both thieves and lawmen reinforce the current equilibrium. Lawfulness pushed on either side, with thieves punishing naivety and lawmen punishing rashness. Sometimes they push from the same side, with both the lawman and the criminal in agreement that you shouldn't have been here.

In the current discussion of Trayvon Martin, the child that was killed by a self-appointed neighbor-hood watch, the lawman/thief have fused together.  The criminal/victim should have known better than to be in that neighborhood.   The lawman/thief (like the war/lord) is a sign of chaos. That we can't tell the two apart is a sign that there is no law at all.

Crazy Maps

OpenStreetMap is a competitor to Google Maps. All of the data is free and open source, but the actual tiles that display the info need to be provided externally. At first I thought this was a liability, but as OpenStreetMap becomes more popular, new tile sets are popping up. The Watercolor version above is one of the coolest I've seen. It gives an idea of the possibilities for map customization.  I'd really like to know how the algorithm that produces the maps works.

Schelling points

Sometimes it really hurts to be out of fashion. A few examples: buying a phone with an obscure operating system limits access to the best apps. Or if a company locks into a multi-year contract with a dying software platform, it shouldn't expect a good return on that investment. Spending months perfecting a MySpace account... etc.

Any time the value of a group resource increases as the number of users increases there is an advatage in picking the eventual winner. Is there a way to know which option everyone else will pick?   One mechanism that drives collective decisions of this type is the Schelling point (or focal point). From the Wikipedia article:

Consider a simple example: two people unable to communicate with each other are each shown a panel of four squares and asked to select one; if and only if they both select the same one, they will each receive a prize. Three of the squares are blue and one is red. Assuming they each know nothing about the other player, but that they each do want to win the prize, then they will, reasonably, both choose the red square. Of course, the red square is not in a sense a better square; they could win by both choosing any square. And it is the "right" square to select only if a player can be sure that the other player has selected it; but by hypothesis neither can. It is the most salient, the most notable square, though, and lacking any other one most people will choose it, and this will in fact (often) work.

 It's a pretty simple idea, but it helps makes sense of otherwise irrational behavior.

Another example helps explain behavior that might otherwise appear irrational. Imagine the same situation as before where two people who can't communicate will win a prize if they pick the same square. Rather than have 3 blue blocks and a red block, instead the prize is awarded for picking the popular choice in black ink on identical white squares.  3 of the squares say $100 and one of them says $99 (See Figure 2). In this case, the best bet is to take the worst (but unique) prize. By exchanging value for uniqueness, the overall expected payout increases because others are likely to pick that too.

If you can think of other pure Schelling points, leave them in the comments. Here's a video if you're bored.

Strategic Marine Life

A shark can swim faster than a small fish over long distances (speed), but over a few feet the fish is faster (quickness).  The idea of a flocking or schooling is to make sure that the speedy predator can never select a single quick prey.  As long as the shark never picks a single fish to follow, each fish only has to dart a couple feet and it's out of the way.

Of course I'm not going to stop with just one.

A well-defined goal has a definite point of completion (eating a fish). At the point of completion changing conditions become much less important, and success is assured (the fish will remain in the shark's stomach). However, before that the point of completion, changing conditions (like the position of the fish) work to prevent completion of the goal (see above videos).

 If the shark had the ability to track a single fish for a long period of time, it would always tire the fish out and eat it. In that way, it could eat every fish one by one. Because it is unable to stay focused on a single fish, it cannot eat any of them.

on Big Data

The New York Times had a longish piece on "Big Data" on Sunday. The author asks,

What is Big Data? A meme and a marketing term, for sure, but also shorthand for advancing trends in technology that open the door to a new approach to understanding the world and making decisions. There is a lot more data, all the time, growing at 50 percent a year, or more than doubling every two years, estimates IDC, a technology research firm.

"Big Data" is a real thing, not just a meme, and it's different from "small data".  Big Data has two defining characteristics:

1. It doesn't fit on one computer.
2. It can't be consumed by a single person.

The first point explains the technological change that is occurring in the Silicon Valley right now. New systems like Hadoop and Twitter's Storm are popping up to process this deluge of information. Meanwhile, new classes of schema-less distributed data stores bring some structure to the huge data sets.  All of these systems are distributed across many computers. By breaking out of a single box, the Big Data Set can be scaled by simply adding more nodes to its cluster.

Big Data is a monster that wants MORE NODES.

The second point helps explain the way companies are changing to take advantage of Big Data.  small data is something that has an Executive Summary. small data is the number of widgets produced by a factory and the Bottom Line.  Big Data doesn't work like that (think of Google's data; what is the Bottom Line?). Every part of an enterprise must have access to Big Data. Each person needs the ability to answer their individual questions in a quick and easy way.  Companies need teams of Big Data Engineers and Big Data Analysts (Data Scientists!) to facilitate the connections between each person in an enterprise and the enormous, ever-growing data set.

 Big Data is an ocean. You don't consume it, you swim in it.

What's the upshot?

They studied 179 large companies and found that those adopting “data-driven decision making” achieved productivity gains that were 5 percent to 6 percent higher than other factors could explain.

This productivity gain should be even more pronounced for companies that focus heavily on Big Data analytics.

So Big Data is great for making your company more money, but what do the kids think?

“The culture has changed,” says Andrew Gelman, a statistician and political scientist at Columbia University. “There is this idea that numbers and statistics are interesting and fun. It’s cool now.”

We can dream, Professor Gelman, we can only dream.