Section 18.4. Perspectives

18.4. Perspectives

This section discusses some perspectives on the future of the Web services platform. Will the platform succeed? What are the risk factors?

18.4.1. Why Will It Succeed?

If this platform is successful, it will be possible to hold secure, reliable, transacted interactions between Web services running on any platform, using open standards that are supported by all vendors. Different vendors will offer wildly different programming models for this platform, but when messages finally reach the wire and flow between services, they will interoperate successfully. The final objective is at least the level of interoperability of today's Web: clients and servers can be on different platforms without having to pay attention to the platform.

Why will the Web services effort succeed? Because of four factors:

• The unprecedented level of vendor support gathered so far, which has resulted in early availability of multiple interoperable implementations.

• The consistent focus (in spite of all the noise of competing standards) on solving the core technical problems, rather than attempting to address every possible requirement. The core of specifications reviewed here provides the core function that other domain specific specifications might extend in the future.

• The composeability of the specification set, which avoids a monolithic solution and permits flexible use and adaptation of the core specifications.

• A pragmatic specification development process in which technical and interoperability issues are resolved before standardization begins.

18.4.2. Risks

Some significant risks threaten to keep this platform from achieving the lofty goal of ubiquitous interoperability. Without that, Web services isn't very valuable. Thus, the biggest risk is an unbridgeable rift between key industry players that might end up creating two or more islands of interoperability.

The IBM & Microsoft collaborative Web services effort started with Microsoft's publication of SOAP 0.9 in the fall of 1999. Since that time, there have been several deep divisions that could have easily derailed the Web services platform. However, calmer heads have prevailed, and so far the industry has been able to stay together at the lowest levels of the stack: SOAP, WS-Addressing, WSDL, UDDI, and WS-Security, as well as at the highest level with BPEL.

That leaves reliable messaging and transactions as the two key areas of contention. The IBM & Microsoft proposals have a significant advantage here, in that they build on and compose cleanly with the other Web services specifications. Thus, with some evolution of WS-Reliable Messaging, WS-Atomic Transaction, and WS-Business Activity, it will be possible to achieve consensus in the industry.

Another commonly articulated risk is that of IBM and/or Microsoft choosing to exercise their large patent portfolios against specifications and implementations, and demand royalties. However, both companies have unequivocally stated their intention to offer these specifications under royalty-free terms.