Oslo: Microsoft’s SOAP opera, continued

Back in June 2000, I went to Microsoft’s Forum 2000 event in Redmond, Wash., where the company rolled out its strategy surrounding SOAP and the .Net platform. It was Microsoft’s first big step toward interoperability, taking the work of people such as Dave Winer and Don Box and moving Windows’s COM+ middleware architecture beyond the boundaries of the operating system in a way that seemed genuinely exciting.

Now, Microsoft has committed itself once again to reaching beyond Windows with a renewed investment in the tools that make up the core of its middleware strategy, labeled Oslo. The .Net platform has exceeded the expectations of many. SOAP now extends far beyond what Microsoft had in mind when it called it “simple,” having become the core of the WS-I’s (Web Services Interoperability Organization’s) standards.

And BizTalk, the “orchestration” server that many analysts marginalized as being strictly for entry-level enterprise application integration, is second only to IBM’s WebSphere in terms of market share (according to WinterGreen Research) and implementations of “mission critical” application integration (according to an IDC study commissioned by Microsoft [PDF]), published last August.

Still, there’s good reason for Microsoft to be concerned about its place in a world driven by SOAs. For one thing, there’s the risk of commoditization. The market force that used to make Microsoft successful now works against it to some degree, with BPM systems now available from JBoss and others as open source. Despite .Net’s success, integrating it alongside Java-based services and applications and other SOA systems is hardly a cakewalk for Microsoft customers. BizTalk doesn’t speak the same modeling language as other BPM environments.

And with a lack of SOA management capabilities, Microsoft’s been mostly a side road on many enterprise buses and not the main track. Despite the .Net plug-in profligation, Microsoft has been vulnerable to accusations that .Net and BizTalk are “lightweight.”

Microsoft’s commitment to business modeling language interoperability and its Internet Service Bus concept are essential to the future of Windows as an architectural element in the enterprise. The big problem is, other than a couple of quick demos, there wasn’t a lot of substance on which to base strategic decisions now.

So it wouldn’t be overly cynical to say that Microsoft is pledging to play on a bigger field partially to prevent customers from shifting their focus to someone with product already in place. Much of what was discussed as part of Oslo was technology that sounded a lot like the “magic software” already mentioned in the form of the Dynamic Systems Initiative, which Microsoft rolled out in 2003.

Will the technology in Oslo finally make Windows an equal citizen with other application platforms? Maybe. But it’s not as though Microsoft really has any other choice in the long term, because if the company doesn’t do that, someone else will for them.