Opinion: The seduction of the systems integrator

Although, during the course of the year, systems integration and consulting has proved recession-proof, this is not likely to be the case a year from now.

In good times, people rely on systems integrators to help them deploy applications more quickly. In bad times, they rely on systems integrators to help integrate existing applications, thereby getting more value out of their existing IT investments. In either extreme of the economy, it has generally been a very good thing to be a systems integrator – unless you were overly reliant on dot-com companies when the Internet went bust.

All that said. A fair amount of resentment toward systems integrators exists within the IT community. Time and again companies have been seduced into hiring systems integrators managed by people with impressive resumes only to be serviced by busloads of college kids getting on-the-job training at their expense. However, rather than relying on systems integrators to get more value out of existing IT investments, leading-edge IT organizations are experimenting with Web services technologies.

The whole idea behind Web services is pretty simple. Instead of building and deploying large-scale, monolithic applications wrapped in arcane proprietary interfaces, people are beginning to experiment with component architectures such as J2EE (Java2 Enterprise Edition) and Microsoft Corp.’s .NET. These architectures rely on technologies such as XML, SOAP (Simple Object Access Protocol), and UDDI (Universal Description, Discovery, and Integration) to create industry-standard interfaces for programming logic. It is not a particularly new idea, but the industry-standard interfaces that will make the idea work are.

This means that a core revenue stream is slowly about to erode for systems integrators. Despite all the claims around business expertise, most systems integrators generate the bulk of their revenue from point-to-point application integration projects. These projects, although important and necessary, are what Web services will automate because the hooks to integrate any given application will be built in, as opposed to being added on later using EAI (enterprise application integration) middleware.

For integrators, the advent of Web services means they will have to add value by tackling complex business-process integration projects rather than rely on low-level application integration projects to drive revenue. The companies that cannot make this transition will simply disappear, and when that happens most people in the industry will say good riddance.

The best way for IT organizations to facilitate this transition is by giving some serious thought to the word architecture, which, during the recent wave of e-business hysteria, fell out of favor in boardrooms bent on implementing anything that higher-ups could label “Internet strategy.” When rational thought returns, IT organizations and the people for whom they work will once again see the value of having an IT architecture that minimizes integration costs. At the core of those architectures will be Web services technology.

Michael Vizard is editor in chief of InfoWorld and InfoWorld.com. Contact him at michael_vizard@infoworld.com.