What Are Program-to-Program Communications?

The general idea behind program-to-program communications is to allow one application to work cooperatively with another, usually over a communications network. To do this, applications must be able to:

• Share data. This necessitates that a common content/format be used between each application. Past examples of this common content/format include IBM's Document Content Architecture (DCA) and attempts at creating industry standard forms using Electronic Document Interchange (EDI) formats with ASCII content.

• Initiate "calls" to each other. Calls are usually based on verbs, such as "send," "receive," "open," "close," and other action-word activities that enable applications to share files and information. These calls are implemented using a common programmatic interface known as an Application Program Interface (API), consisting essentially of verbs that applications use to communicate with each other.

• Communicate with each other over a network. Programs need to be able to send and receive data and information over a common network (which involves using a common communications protocol such as TCP/IP, SNA, or other networking protocols).

Traditional distributed applications have communicated in the past by using a common format, a common API, and a common network, as illustrated in Figure 2-1. This figure shows that:

• both systems support the same format/content (in this case it's XML or HTML);

• both systems support the same application program interface; and

• both systems support a common network (in this case the Internet) in order to communicate.

Figure 2-1. Program-to-Program Communications Basics.

In Figure 2-1, application "A" sends a request for data or information to application "B." Because the computer platform that "B" runs on uses the same underlying network (the Internet), the same programmatic interfaces (APIs), and uses the same way of reading and presenting content (based on XML or HTML), the two applications are able to easily communicate and share data. Web services use this same basic approach (common format, common APIs, common network) to enable one application to provide messaging, transactional, or computational services for another.

Web services behave very much like traditional program-to-program communications. A "requester" application sends out a request for a service (in the Web services world this request is sent to a UDDI registry that contains information on where that service can be found). Once the service is located, WSDL negotiates between applications and establishes the rules for transferring/sharing information. SOAP binds the communications session, and the HTTP protocol is used to send information back and forth over the Internet. Chapter 3 covers Web services architecture in greater technological detail (for readers who wish to understand how Web services APIs, content, and registries actually work).