Microsoft and Mules: Why the Company Won't Lead on SOA

Share it on Twitter  
Share it on Facebook  
Share it on Linked in  

Ron Schmelzer is on a quest to get Microsoft to cowboy up and lead the way on SOA.


I spoke with him last Friday about his recent ZapTake, succinctly titled, "The Inadequacy of Microsoft's SOA Message." He'd just published the piece and e-mailed it to several tech publications, and now he was on to the interviews.


"We definitely got Microsoft's attention with our published note, which was I guess, part of the reason why we wrote it," he said.


Yeah, no doubt. Microsoft can be pretty stony silent when you want to talk. I've interviewed governors and members of Congress, and I can tell you -- it's harder to get an interview with a C-list person at Microsoft. But publish something negative, and you'll hear from them loud and clear.


Schmelzer wasn't writing the piece to be mean. He sincerely wants Microsoft to wake up and smell the SOA potential. After a briefing with Microsoft's Burley Kawasaki on the company's SOA messaging, Schmelzer was frustrated. He was also convinced Microsoft was missing the boat. In his ZapTake, Schmelzer wrote:

"Bottom-line: the official message coming from Microsoft about SOA is that SOA is just Web Services-based integration. Even the customers they trot out on their conference calls and in their events claim to have implemented SOA on Microsoft products, but when pressed admit that it's nothing more than Web Services-based EAI."

When Microsoft speaks about SOA, it focuses on integration, but Schmelzer contends that's such a small piece of what SOA could be. Integration, open standards -- these are small potatoes compared to what Microsoft could offer if it service-enabled Office, Silverlight, even the Xbox, and you could call any service from any of its products. Forget the idea of integration -- Microsoft could create a SOA on its own platform, revolutionize computing and lead the way on SOA.


"What -- a proprietary SOA?" I asked Schmelzer. No, that wasn't what he was advocating -- but yes, he said, Microsoft could service-enable its own platform without open standards and have something really powerful. He expained:

"Microsoft is not a collection of products, but Microsoft is a collection of capabilities. Anybody who has invested in Microsoft products probably have the most diverse set of capabilities than anybody in the market. That is saying a lot. There are very few vendors - other than Microsoft, perhaps Adobe - that can execute on the ultimate service consumption platform vision. However, there are hundreds of companies that are executing on the SOA web services integration vision that Microsoft is currently pursuing."

In short, it's not that Microsoft is wrong, he added. It's just that they could offer so much more. They could be SOA market leaders, rather than followers.


And here I started to see the problem. Schmelzer isn't Don Quixote, titling at windmills. And he and Microsoft aren't having a communications problem.


Schmelzer is trying to push a mule.


The problem is, mules go when and where they want -- they hate to be pushed or prodded. The more you want to move them, the more they dig in their heels.


And Microsoft -- perhaps the biggest mule of all tech companies -- isn't budging, particularly since Schmelzer wants it to move in a direction that is just not where it typically does business.


I told Schmelzer I thought he was pushing a mule. Since he's not from the South, he found my metaphor confusing, so I reworded. "I haven't followed Microsoft as a beat, but I've never seen Microsoft lead the way on anything," I said. "Have you?"


He paused and then said:

"No, I think the problem is Microsoft as a company is at a crossroads. I don't think this is even SOA-specific. I think the challenges that they are having with software and service and the Web 2.0 marketplace, and competition from all these different sectors, Open Source, Linux, I think these are all affronts to Microsoft. And I think the reason why Microsoft is sort of struggling is because they still have their '80s, '90s perspective on the market, which is that if they can just get people to buy into a platform and focus on that platform, and then basically leverage that platform for continued benefit, they'll maintain a competitive advantage."

The problem is, Schmelzer contends their old business model won't hold up in a service-enabled, Web 2.0-kind of world.


I see his point -- and I respect him for trying. I think he really wants to see Microsoft do well in this space.


But he's still he's trying to move a mule. Microsoft has legions of Microsoft-certified IT staff in small- and mid-sized businesses and government organizations all across the country. And all of these techies are watching to see what Microsoft will do with SOA.


Microsoft doesn't have to be IBM -- it's not necessarily counting on the big companies for its business model. Microsoft aims to be McDonalds: It doesn't make millions by offering the best, gourmet hamburger, but it does offer a standard, reliable, good-enough hamburger that everyone understands and can stomach.


Microsoft doesn't have to lead the SOA market, it just has to stay slightly ahead of its technology late-adoption customers. And by offering anything SOA -- even if it's just Web services and repackaged EAI -- Microsoft accomplishes that goal.


My guess is, when customers expect more, then we'll see Microsoft start adding the SOA goodies to the Happy Meal.