In my day job, I do a lot of work with Amazon’s cloud infrastructure arm, a.k.a. Amazon Web Services. These are the guys who sell the servers, the storage, the platforms, and the automation that powers pretty much every Silicon Valley startup you’ve ever heard of.

People buy AWS because it’s brutally efficient, particularly at scale. If you’ve ever had to manage a physical data centre–and, more importantly, provisioning in a physical data centre–being able to log into a webpage, click a button, and spin up $100,000 worth of servers in ten minutes is literally mind-blowing.

It’s even more mind-blowing to realise that if, for example, you turn all those servers off when you’re not using them–say on weekends and at night–then your costs drop to $40,000. Oh, and if you open another webpage, you get a reporting graph that tells you exactly that. In real time.

This is Amazon’s business model, and this is what is actually meant when people talk about “cloud computing”; it’s a cost and consumption model, not just a buzzword for virtualisation. Amazon, more so than any other player, made this industry. It commoditised IT infrastructure, made it one-click and on-demand. Every other player in the market is still scrambling to catch up.

The reason I’m mentioning this, is because the fact that this is my day job means reading articles about services like Kindle Unlimited can be… interesting. Interesting because it’s pretty obvious, for anyone who comes from IT cloud land, what Amazon’s new pay-per-page model is.

It’s cloud computing paradigms brought to bear against the publishing industry. It’s fiction-as-a-service, pay-as-you-go commoditised consumption in the same way EC2 is to a blade server or S3 is to a SAN. Whether it’s “good” or not is a moot point; it is, in the same way cloud computing now is.

Amazon made the game. Now it’s up to the rest of us to figure out the rules.