This document will describe the typical processes used to initiate, develop, build and deploy a Google Web Toolkit (GWT) application at Emory. It will outline the general steps and link to other more in-depth documentation on the framework itself. We'll discuss how to establish a central build and package environment and discuss the typical deployment patterns and processes. Given the evolutionary nature of frameworks like GWT, this document will evolve as well. We've tried to be as specific as possible when describing how we do certain things because there are often several ways of accomplishing similar tasks.
It is also important to remember the high-level goals we're trying to accomplish from an "enterprise perspective." One goal we're trying to accomplish is to establish the necessary infrastructure that will allow us to integrate systems quickly and efficiently as time goes by. Given that, one fundamental tenant of our approach is that all web applications will communicate with back end data and logic sources via some sort of "service" layer. This layer is responsible for tying the client side of the web application to the back end services that are needed for the application. It is important to note that "services" does not always mean "web services." Sometimes, they will, but when we refer to services we're really referring to anything that exposes back end data or logic that will be leveraged by this application's service layer. Since it can sometimes be confusing, we'll provide some definitions for use throughout this document.
The strategy used in this document involves two things:
This way, you should have an idea of what the project looks like at the end as well as specific steps to get you there if you choose to follow these instructions.
Comments and Suggestions
If you have any comments or suggestions to improve this document, feel free to make those comments directly on the page.