There are at least six different meanings of "cloud". All of them have to do with servers and elasticity of resources, but the impact of each one on the architecture and revenue model of your application can be very different. We'll visit each of them, and see which of them are really very different from when computing was about mainframes and terminals. Roughly speaking, PHP was born in the PC era, got big during the e-commerce boom, survived web2.0, and is now getting ready for post-PC. During the next two years, the web is facing a fierce battle against native apps. It's sometimes hard to keep up with all these differences and constant changes in the landscape. That's why it's often easier to call it all just 'cloud'. But cloudy wordings lead to cloudy judgement, so in this overview we'll apply a bit of structure to the different sections of the skyline again. Besides all the many benefits of different kinds of cloudiness, we'll dwell on some of the very dangerous side-effects of what Richard Stallman calls "careless computing". We'll discuss why "open source cloud" is a contradiction, and how the 'decentralized web' movement is working on achieving freedom in the cloud, in much the same way as linux has given us freedom on the desktop and freedom on the server. We'll end with a demo of how to shift the pivot-point of your cloud-based applications from the server to the service end-points, which is something we will all have to do if the open web is to stand a chance against Apple's iCloud.