SOA: Wanted Dead or Alive

The Burton Group’s Anne Thomas Manes set off some serious buzz when she proclaimed in a 5th January blog post that SOA is dead, and especially the term service-oriented architecture.

In the post entitled SOA Is Dead; Long Live Services, Manes said:

“SOA met its demise on 1st January, 2009, when it was wiped out by the catastrophic impact of the economic recession. SOA is survived by its offspring: mashups, BPM [business process management], SaaS [software as a service], Cloud Computing, and all other architectural approaches that depend on ‘services.'”

Manes certainly knows her SOA stuff, as she was formerly CTO of SOA governance software provider Systinet, now a part of Hewlett-Packard. And she helped shape some of the specifications involved in the Web services and SOA world while working at Sun Microsystems and Systinet.

Although Manes said the term “SOA” is no longer viable as a selling point to businesses because it connotes big, expensive projects, she noted that the basic tenets of SOA remain critical.

Said Manes in her thoughtful post:

“Once thought to be the saviour of IT, SOA instead turned into a great failed experiment – at least for most organisations. SOA was supposed to reduce costs and increase agility on a massive scale. Except in rare situations, SOA has failed to deliver its promised benefits. After investing millions, IT systems are no better than before. In many organisations, things are worse: costs are higher, projects take longer, and systems are more fragile than ever. The people holding the purse strings have had enough. With the tight budgets of 2009, most organisations have cut funding for their SOA initiatives”.

Moreover, Manes added: “It’s time to accept reality. SOA fatigue has turned into SOA disillusionment. Business people no longer believe that SOA will deliver spectacular benefits. ‘SOA’ has become a bad word. It must be removed from our vocabulary.”

In an interview, Manes said she wrote the post because she felt it needed to be said that the recession has essentially killed SOA because “nobody is going to buy SOA initiatives this year … But meanwhile, services and service orientation are more important than ever before. The catastrophic impact of the recession has killed SOA as we know it. But it doesn’t change the need to service-orient. The need is more apparent than ever because of the budget tightening.”

Indeed, Manes said she thinks organisations are in “desperate need” of application redesign, and services and service orientation is the most efficient way to approach rearchitecture.

Meanwhile, Manes said she has received all kinds of reactions to her post. She said about one-half of the responses have been in agreement with her. About one-third of the responses were in disagreement and “the rest said that maybe ‘big’ SOA is dead but REST [Representational State Transfer] is going to take over with REST-based services.”

In the comments responding to Manes’ post, Tim Vibbert, a software architect and SOA implementer at Lockheed Martin, said, “I totally agree with your assessment of the state of the SOA union. Without a commitment to changing the status quo, ‘SOA’ efforts fail. Most efforts focused on the platform stacks and not the enterprise. I’ve began to ask, ‘What is your SOA motivation,’ if not the business/enterprise then why do it at all?”

Dana Gardner, an analyst with Interarbor Solutions, wrote in Manes’ comments: “Could be that the enterprises that dismiss SOA, nomenclature or computing shift, will be all the more ready to hand off more of their IT functions to the SAAS and cloud providers that do do SOA well and pervasively. In other words, you’ll do SOA one way or another … it’s just whether it’s your competency or your cloud provider’s.”

John Crupi, CTO of enterprise mashup software provider JackBe, said, “SOA for SOA’s sake is dead. SOA with a purpose is the key. We believe SOA will be relegated to data services and the focus (aka purpose) will be the user and mashups.”

Jason Bloomberg, an analyst with ZapThink, commented, “We’ve been seeing this trend for a few years now – the business doesn’t want ‘SOA,’ never did. The business wants solutions to problems. ‘SOA’ has become a bad word. But SOA – that is, the best practices of service orientation – are vitally important, broadly adopted, and now mainstream. After all, you can’t do Cloud Computing without SOA – but you can definitely do it without ‘SOA.’ The bottom line is that SOA is a somewhat loose collection of architectural best practices. It’s up to the architects to know which best practices are appropriate for solving the business problem at hand. And as it happens, one of those best practices is not to call what you’re doing ‘SOA.'”

Bloomberg’s ZapThink partner, Ronald Schmelzer, said, “To say SOA is dead, either as a term or as a concept is misguided.” Schmelzer said SOA never had sex appeal, but it was never meant to have it.

“So, is ‘SOA’ dead either as a concept or simply as a term?” Schmelzer asked in an interview with eWEEK. “Of course not. What has replaced it? What explains its lack of value? The concepts we now discuss are other ones, or ones that simply build on the ideas of SOA. SOA has simply faded from the limelight as other new topics command our attention. But as a foundational concept of computing, and one that companies still struggle to implement, SOA is alive and well.”

Schmelzer took his argument a step further, noting that there has been a surge in SOA training and SOA engagements over the past year. He added:

Anne is simply in the echo chamber, and the chamber has stopped echoing on the topic of SOA. So what? Now, let’s get down to details and make this thing work. Show me a company that has truly mastered agility and flat cost of change in the face of continuing heterogeneity, and I’ll say we have achieved SOA success. All the rest are on that path. To say it’s not worth it to pursue that path is misguided. To say that since the topic of SOA has faded from the pundits’ view is evidence of its demise is equally misguided.

Perhaps the company that has made the most hay, and perhaps money, out of the SOA model is IBM. Sandy Carter, vice president of SOA and WebSphere strategy, channels and marketing at IBM, said, “We are seeing growth in SOA adoption in this economic client because of the value it delivers. In fact, the number of customers leveraging SOA with IBM has grown 28 percent since October of 2007.”

Carter said IBM has 430 customers that started with a technology improvement project “and they started with SOA.” Refuting Manes’ theory on the use of the term “SOA” being a core issue, Carter said, “A rose by any other name is still a rose. I don’t know that what you call it is important; I just don’t know if that’s relevant. The point should be what is the value SOA is bringing.”

Moreover, Carter said several IBM customers are “working with services that are mashable and they’re doing a lot of experimentation with [that] model, and because of SOA they’re able to mash up new applications and test them.”

Carter also said attendance for the IBM Impact 2009 “Smart SOA Conference” is expected to be strong even in the weak economy, and due to heavy demand IBM has had to put a cap on the number of customer speakers at the event to be held in Las Vegas in May.