SOA is integration. SOA is not integration. Got that?
There's a renewed debate raging about the relationship between the practices of service oriented architecture and integration.
SOA is more than EAI 2009, but where do we start?
Loraine Lawson stirred the pot up when she posted Gartner analyst Yefim Natis' statement from the recent integration summit:“SOA is integration.” (Quoted in SearchSOA.com.) Yefim was discussing the fact that few organizations have one comprehensive SOA -- most organizations have multiple islands of SOA that will eventually need to be brought together. In other words, integration of integrations.
Loraine said that ostensibly narrow view of SOA was not well received among members of the Yahoo discussion group on SOA, which includes many of the thought leaders in this space. Anne Thomas Manes, for example, pointed out that “Many organizations mistakenly perceive SOA as an integration strategy. But it is not. SOA is about architecture. To achieve SOA, you must rearchitect your systems."
But Loraine takes more of a pro-SOA-is-integration stance, noting that "most companies aren’t getting into SOA for a complete rebuild. Most companies deploy SOA because it’s so darn helpful with simplifying integration:"
"...companies don't want Extreme Makeover. They're looking for a slight update, something that ties the room together, as interior designers like to say. ... And here’s another hard truth: Although David Linthicum and others believe that agility is the ROI for SOA, many companies are realizing SOA ROI through integration."
Agility -- while a commendable goal that everyone needs to shoot for -- is a vague, hard-to-quantify state of existence that SOA-based approaches may or may not be able to accomplish. But the results of integration are often measurable. Perhaps, Loraine suggests, integration can even be a starting-off point for more complex transformation efforts down the road.
Still, in a follow-up post, Anne Thomas Manes urges practitioners to look beyond integration when planning SOA efforts. "It's fine to use service oriented middleware to implement integration projects, but then you need to readjust your expectations. Most organizations that I speak with say that the goals of their SOA initiative are to reduce costs and increase agility. Unfortunately, these organizations aren't likely to achieve these goals if their projects only focus on integration...."
I see Anne's point, as SOA needs to involve the organization in a transformative process. SOA is much more than enterprise application integration 2009. It's a new way of looking at how technology can be better leveraged to serve the business. Integration as a goal in itself can be fairly limiting, and confine SOA to the IT silo.
SOA should be about more than simply linking app A to app B, and so on. SOA is about innovation. Every SOA-related proposal out there should include the word "innovation" somewhere in the text. And, let's face it, SOA would be downright boring (make that Boring with a capital B) if it was just another means of app or systems integration, and nothing more.
But, nevertheless, integration is still an essential phase in the evolution from siloed, proprietary systems to full-functioning service oriented architecture. In many cases, it can help provide the initial justification for commencing with an SOA approach. And for many SOA proponents, especially those facing organizational resistance, the key is to just start.
And, remember, SOA may involve integration, but it truly is about innovation.