The power law of cloud computing

In the future we won’t need all our programs and data to live on our own computers. They will be floating in a vast, interconnected nebula of servers and databases, and will be accessible over the Internet.

Ideally that cloud of servers would be spread across a multitude of providers, indistinguishable from one another. Ideally we would be able to choose who takes charge of running our infrastructure. Ideally we would be free to choose the provider or providers of our cloud computing service.

According to Macleod, the problem is that this scenario fits what Clay Shirky explained earlier:

Diversity plus freedom of choice creates inequality, and the greater the diversity, the more extreme the inequality.

In systems where many people are free to choose between many options, a small subset of the whole will get a disproportionate amount of traffic (or attention, or income), even if no members of the system actively work towards such an outcome. This has nothing to do with moral weakness, selling out, or any other psychological explanation. The very act of choosing, spread widely enough and freely enough, creates a power law distribution."

This power law distribution is widely known in many fields. In economics, Vilfredo Pareto established that wealth follows this distribution, with 20% of the population holding 80% of the wealth. Chris Anderson uses it to explain the concept of the Long Tail, and Shirky to explain the nature of the success of blogs.

That is what worries Macleod: the fact that, invariably, the cloud will end up controlled by a single player, just like Microsoft with desktop operating systems and Google with search. Whoever controls the cloud will end up with an enormous business.

Not many people think the way Macleod does. Even Larry Ellison has said that he has no plans to build the cloud, because he doesn’t see how anyone can make money in this business.

But the most interesting criticism of Macleod’s proposal came from Tim O’Reilly, which in turn drew another interesting reply from Nick Carr, which we will look at in the third part of this series.