contolledagility makes some great points in this post.
I guess in the end no one has the motivation to solve for INTEROPERABILITY.
It's not an Open Source thing since those that code in this space are (generally) one language oriented. It's not a Microsoft thing since they would be making it easier for customers to stay with non-Microsoft integrations. It's not a IBM/BEA/Oracle thing since they would be making it easier for customers to stay with non-JAVA integration.
Who has the confidence to step up to this? Vendors have nothing to gain. Speeding up delivery time for Integrators/Consultants costs them contract dollars. Middleware vendors make money off the complexities of the integration problem. CTOs and CIOs are trying to consolidate their platforms and not stratify them even more.
Perhaps if someone creates a tool for proving TRUE INTEROPERABILITY they could Open Source it ... Microsoft/IBM/BEA/Oracle can go fight for licensing rights.
In the meantime, its good architects supporting good developers to master the approach to "contract first".
What do you think?