What is application integration?

Unfortunately the term “integration” is ambiguous and can mean many things. “Integration” is defined as the process of enabling existing and new applications to easily and securely exchange data and functionality with one another on an internal and/or external basis.

EAI (enterprise application integration) refers to the plans, methods, and tools aimed at modernizing, consolidating, and coordinating the computer applications in an enterprise. Typically, an enterprise has existing legacy applications and databases and wants to continue to use them while adding or migrating to a new set of applications that exploit the Internet, e-commerce, extranet, and other new technologies. EAI may involve developing a new total view of an enterprise's business and its applications, seeing how existing applications fit into the new model, and then devising ways to efficiently reuse what already exists while adding new applications and data.

Presentation Level Integration™ -- a new approach to integration

So at the end of the day today, each enterprise has a multitude of applications deployed on its premises and they have a real business need to have these applications communicate with one another to streamline their business.

There has to be a simpler way to make the process of enterprise application integration cheaper, faster, and more predictable.

Taking every consideration aside, i.e. disregarding what layers the application touches, the operating systems it runs on, the platform it’s built on, etc. the overwhelming majority of enterprise applications work by human beings interacting with them. To “interact” with an application can mean typing on a keyboard, clicking buttons on a mouse, speaking into a microphone or other means. This human interaction point with applications is called a “user interface” and any application, whether it is mainframe-based or J2EE-based is controlled via some form of a user interface.

This is of course nothing more than the obvious – there’s no brilliance in realizing it in and of itself. However, if you follow the same realization, you can abstract any application to simply be a series of controlled interactions between a user interface and the underlying business logic. A human being does not compile code nor do we get plugged into a machine ala the Matrix. We simply see objects (text areas, buttons, windows, links) on a screen, and navigate accordingly by using the appropriate peripherals (keyboards, mouse, voice, etc.) Our interactions with the user interface of an application results in that application performing whatever functions it was programmed to do.

For example, a mainframe library system can be modeled as a human being interacting with a terminal emulator by typing commands into a text area. For every text command the user types in, the library system has a discrete response which is delivered back over the terminal. Similarly, more complex applications, like purchasing a book at Amazon, can be modeled as a finite state model of text entries, option selections, and button clicks executed via a Web browser: You enter in a title and search, you get a list of links, you select the book you want and then click purchase; each state of the application expressed in unique user interface steps.