Is Service-Oriented Architecture the latest trap for unsuspecting CIOs?
Type the words "Service-Oriented Architecture" in Google and you'll be inundated with white papers and overviews from a myriad of ICT consultants all around the globe. To the casual reader it doesn't sound half-bad. The documentation has an academic intensity to it. It's full of promises like "application integration", "data rationalization" and "business service enablement". Surely only a Luddite would fail to appreciate the beauty of this answer to our IT prayers?
Yet something in my bones tingles whenever I hear SOA mentioned. When I read the material my BS antenna goes on high alert. To cap it all off, no one seems capable of explaining to me exactly how I can get it and how much it might cost.
The real giveaway for me is the literature. Let's start with a definition of Service-Oriented Architecture. This one is from IBM, which describes SOA as "an -application architecture within which all functions are defined as independent services with well-defined invokable interfaces, which can be called in defined sequences to form business processes". Come again! What on earth does that mean?
SOA-related literature is also interspersed with sexy business words like agility, alignment and integration, and acronyms like J2EE, CORBA, SOAP and UDDI. If these don't impress you, then you'll be battered by the analyst projections, which all highlight how the technology is catching on and warns users of the dangers of missing out. Then there are the diagrams. After 25 years in ICT, I've learned that there's always something fishy going on when the diagrams have arrows going in every direction.
SOA sounds a lot like the IT vendor equivalent of a promise to still love you in the morning. Just buy our product and somewhere down the track SOA will materialize and all your applications and data will be seamlessly integrated.
Hasn't the IT industry been down this path before? I remember back in the 1980s when all that IT managers wanted to know was whether the product I was selling could be something like an LU6.2 logical device in an SNA network. At that time the business problems you addressed with the product seemed irrelevant. But then I was undermined by one of the great marketing tactics in the history of ICT. When Systems Network Architecture (SNA) came along, IBM painted a world where every device in the network fitted snugly together. While this sounded great in principle, in reality it was a ploy to keep everyone other than IBM out of the organization. Today I worry a similar intent exists behind the SOA rhetoric devised by the marketing departments of the IT vendors.
The problem with earnestly following an -architectural road map is that it can blind CIOs from appreciating whether the product they have selected is actually the best one to do the job at hand. I believe that it was a naive belief in the potential of such architectures that placed the IT industry on the back foot for much of the past decade.
For the last six months nearly every CIO I've met has told me that they are working on a new architecture. Everyone in the industry seems to have come to the conclusion that the secret to good governance is to provide a framework where everyone in business can see how everything in ICT fits together. I only hope that this recent enthusiasm for architectural standards doesn't undermine all the efforts IT professionals have made over the years to earn the respect of the business
Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.