A couple of months ago I got an e-mail inviting me to keynote an SOA/Web Services conference in Beijing. My immediate reaction was - "Good. China has reached the stage where it's hosting international conferences on the subject." Actually, 2005 marks the fourth time this particular conference is being held. While China is growing rapidly in IT, it's fairly new to the services game. My subsequent thought was - "This is great, the pitch for service orientation is becoming a reality."

I say this because although SOA is rapidly becoming reality, and Web Services are proving to be the ideal platform to host SOA, SOA's true potential will only be realized when the service itself is distributed geographically, and can be realized not only by architecting solutions that are technology-agnostic, but also by implementing geographically distributed cross-cultural solutions, thus truly uniting the distributed enterprise. If it smells like utopia, that's because it is; we can but aspire.

The prospect of going to China and talking about SOA and Web Services was very exciting, so I accepted the invitation. Now came the hard part - what to talk about? I could talk about WS standards, security, governance, and strategy - there's no shortage of topics to present. But somehow none of these really hit home. I wanted to talk about a subject and offer a unique perspective on something I had a deeper perspective on. Then it suddenly struck me. I'm going to China, a country ripe for a revolution that will support SOA initiatives around the world.

Most of the initiatives in place today target the development of services and their deployment in an organization. The actual development of these services lends itself to outsourcing - primarily because of economies of scale, the resource crunch, and skill set availability - and all these lead to the main pain point of organizations - the cost of operation. SOA is causing a major change in the development of services and is inherently promoting outsourcing. The current outsourcing model has a mechanism - the Global Delivery Model (GDM). GDM is basically working in orthogonal directions. Established system integrators based in the US such as IBM and Accenture are offshoring development, and companies that started at the delivery end of the spectrum, such as Infosys, TCS, and Wipro, are moving towards a higher level of strategic engagement. But the model is the same - offshoring development of services that are deployed here in the US.

At Infosys, we are evolving the GDM that we put in place for traditional projects such as platform migration, development, and maintenance, and are applying the same model to developing SOA strategy and blueprints for our customers. In other words, GDM is being leveraged at the strategic, not just the tactical, level. It can be called a technology competency offering. The skill sets for these distributed groups are in architecture, analysis, IT strategy, and mainstream development. Of course, for an end-to-end offering, development has to support the downstream strategy.

However, all this is still focused on developing at an offshore location and deploying at the client. The next revolution will be distributing the hosting of these services in a shared services model. The infrastructure to support this in terms of people, process, and management is there. The business model is not.

Coming back to China, the economies of scale will let China replicate the GDM model and offer similar services at a global level. All the pieces are there. They just need to be connected.

So I'm en route to China as I write this. On to SOA, Web Services, and the Global Delivery Model!

Resources