Originally, Java was developed to avoid doing things server-side. With Java, the browser takes care of processing so that the code is only complied once so it can be platform-independent. Node doesnít seek to change that, but is instead useful when you have an application that is I/O heavy, but easy to compute. Node runs on a single thread so that it isnít going to consume a lot of your local computerís resources. This will make it ideal for a lot of different applications. The first chapter of Learning Node is dedicated to the installation solely. After that, the next 3 chapters go over basic concepts of how to use REPL, TCP, HTTP, Modules, and many other useful concepts to start you going with Node. If you already know this beginnerís stuff, you can skip or skim these chapters and move on to the more advanced topics.
After Chapter 4, I found myself skipping around looking for more specific concepts rather than going in order. Being more interested in the administration of the system, I moved to Chapter 15 to work on security. There is an amazing amount of customization you can do to control who is accessing specific pieces or all of the scripts you have running. I really liked playing with the difference security concepts available in Node. Other than security, Node has pretty much everything you need for a fully usable server. Learning Node has chapters on routing, using templates, graphics, sockets, and most importantly, debugging for when you mess up somewhere along the line.