SOA: Dead as Elvis


There's just something dull about January. Oh, sure, it's fun for a day or two, what with the champagne and the black-eyed peas and collard green dishes. But once the corks are popped and you realize the awful reality of actually keeping a resolution-well, the month starts to look like a long drag.


I was feeling the ennui of the month last night, already regretting my resolution to never again write about the whole "SOA is/is not integration" debate (not an easy resolution to keep, when you need a post for the next day and David Linthicum's just reopened the fray with a new blog entry and podcast). Then my e-mail flashed.


"Burton Group: SOA is Dead."


Anne Thomas Manes, the Burton Group analyst whose research first raised questions about SOA's ability to deliver, had declared the whole thing mute by writing an obituary-complete with a really cute graphic of a doomed SOAsaurus-for SOA.


Well, shut my mouth.


Wrote Manes:

"SOA met its demise on January 1, 2009, when it was wiped out by the catastrophic impact of the economic recession. SOA is survived by its offspring: mashups, BPM, SaaS, Cloud Computing, and all other architectural approaches that depend on 'services.'"
Of course, you have to keep reading, because-of course-she's not actually announcing the end to the concept of building an architecture based on services, so much as the hype and drive to promote SOA as an end in itself:
"The demise of SOA is tragic for the IT industry. Organizations desperately need to make architectural improvements to their application portfolios. Service-orientation is a prerequisite for rapid integration of data and business processes; it enables situational development models, such as mashups; and it's the foundational architecture for SaaS and cloud computing. ... Although the word 'SOA' is dead, the requirement for service-oriented architecture is stronger than ever."
In other words, a service-oriented architecture may still be a good thing, but it's just too hard and too divisive. It was never supposed to be an end in and of itself, but part of something bigger, the kind of transformation most companies need, but few are willing to invest in.


And just in case anyone might forget that Manes is against integration as a driver for SOA, she throws this in, "Incremental integration projects will not lead to significantly reduced costs and increased agility."


Last time I twittered her name, Manes' post had drawn a good 14 or so tweets. Most people seemed to agree with her, and many were quick to add their +1 to her comments about the continued importance of services with cloud computing and SaaS.


I suspect the hailstorm's just winding up, however. Joe McKendrick has already written a response, agreeing that the term "SOA" gets in the way, but noting that this is not the first time SOA's been declared dead. In fact, he questions the very premise: Can a "philosophy, methodology, and set of best practices or patterns" even die?


Linthicum, too, had already posted his agreement by day's end, adding four fatal factors he sees as killing SOA, including big consulting firms, too much vendor hype, and too few qualified enterprise architects.


One of the first to respond was Steve Jones, who writes an SOA blog and works mainly on SOA for a systems integrator. Jones took the news in stride, giving Manes credit for writing a great headline.


He took her post as more of a death knell for using SOA for vendor marketing than an actual declaration of SOA's demise. Like Manes, he sees services as having particular vitality in light of the recession.

"So in a recession you need to Identify your services, understand the business value that they deliver, understand the cost model to deliver that value and then decide on the right technology approach. If that isn't SOA then I don't know what is. So in reality its the 'other' SOA that is dead, not the SOA of today."
So-what does all this mean for the average IT leader, trying to make it through the year?


First, it's nice to see everyone be so candid about what's gone wrong with something that seemed so promising. Second, you now have someone officially saying what should have been obvious all along: You can focus on building services and find business value to that, without investing in a complete overhaul of your current architecture.


And it would be nice if vendors would suddenly reprint their marketing material and we could recapture the good of SOA without the hype. But to be honest, I don't think Manes was just talking about the hype dying.


I think she meant what she said-people are tired of the whole concept of SOA and it's just not something they're willing to struggle toward. That part, you'd probably already worked out for yourself.


But for my part, I think SOA may have the undead power of Elvis - there will still be sightings. Based on the e-mails in my inbox, SOA may be dead, but it's far from gone.