SOA - loosing sight of the objective

I visited an insurance company today to talk about their investment in creating an SOA. They had appointed a manager to lead the project and had forty five programmers working. They have been working for over a year. So far, there has been no tangible result. Yes, they are creating services but they need a lot before they can create a worthwhile business process.

Forty five Programmers for a year? They have obviously lost sight of the principal justification for such an investment - Agility! How can a technology that's touted as the answer to everyone's need for agility need more than forty five man years of effort just to get started.

Then there's another problem - Once the new services are created, say with an average of two man years of programming a piece in them, how easy is it to modify them? They desperately need a technology that can create services in a hundredth of the time. A technology that does away with the need for specification and programming. One that allows its users to create services based applications iteratively through dialogue with the users. They do exist.

Until you've got that, you have not got anything.