Web services promises to finally provide a universal mechanism for connecting loosely coupled systems. If the dream bears fruit, it will represent a huge advance in enterprise computing, facilitating distributed, transaction-centric collaboration in an inexpensive, quick, and reliable manner.

As with every "next big thing," the combination of analyst buzz, Silicon Valley expectations, and incessant media attention has led to a manic state of frenzy. A heady mix of confusion is rarely a good thing and can derail businesses and get CIOs fired. Fast.

To date, we face several issues that may lead to the implementation failure or even premature demise of Web services. As all of this unfolds, technology professionals will need a game plan to separate the hype from reality.

The Premise and Promise of Web Services
Web services is neither a single technology nor a new concept. Instead, it is a portfolio of methodologies and standards that extend and build on Internet technologies. Historically, loosely coupled, highly distributed systems have enjoyed marginal success due to the very complexity required to design such systems, as well as the sheer cost of supporting private networks and proprietary interfaces. The premise - and the promise - of Web services is that the use of open and shared interfaces, e.g., WSDL, atop the Internet will make it easier and cheaper to connect, integrate, and collaborate.

While the Internet is a mature networking platform, higher-level protocols, such as XML, are still relatively new. Moreover, the standards that use XML to specify Web services interfaces are at the very first stages of their life cycle. The ensuing state of flux results in a number of issues that an aspiring adopter must consider.

Competing Platforms
Two major platforms compete for your Web services business: .NET and J2EE. Platform choices and the closely associated tool(s) selection are significant decisions that go far beyond the scope of this article. It should be noted, however, that Web services implementations are eventually constrained by the underlying platform, which is of particular concern in high-volume transaction environments.

Maturity
Several important functional components of the XML substructure for Web services are either just under development or in their early stages of release (see www.w3.org/2000/03/29-XML-protocol-matrix). Moreover, fundamental Web services specifications and interfaces are also immature and fall short in terms of functional completeness, e.g., robust security and authentication mechanisms.

The lack of maturity in these core standards currently forces adopters to either do without certain functionalities or pursue proprietary extensions. While the former can compromise the completeness of a solution, the latter generally impedes interoperability and extensibility.

Right of Ownership
Although the majority of Web services protocols are open, they are not unencumbered intellectual property. Some major computer companies today hold rights to various pieces of SOAP, WSDL, and UDDI, and seem intent on capitalizing on their IP through the "reasonable and nondiscriminatory" (RAND) licensing framework (see www.w3.org). This may have costly implications to users and fundamentally undermine the adoption of Web services.

Tackling Web Services
So what can a CIO and IT staff do while these issues evolve? Following are several practical tips on how IT professionals can educate themselves and test the premises and promises of Web services - a do's and don'ts guide: