Chapter 13. The Future of MDA

"Prediction is difficult," Yogi Berra is reported to have said, "especially about the future." And so it is, especially with the future of technology. Even so, we're willing to place our reputations in an unattended picnic basket and risk a look at the future of model-driven development.

The value proposition for MDA is that the cost of building and maintaining systems is significantly lower. The combination of the following makes a powerful business argument:

• Raising the level of abstraction so that people can express themselves and communicate with man and machine alike more productively

• Raising the level of reuse to a higher level of granularity so that bigger systems can be composed from ever-larger elements

• Relying on the power of design-time interoperability, so that models can stand alone, ready for combination with other models in an additive manner

Before we dive too deeply into speculation, however, we should be clear about where model-driven development is today. It is not just a vision: Many systems are being built using models, albeit not as many as we'd like, or as many as one might expect given the value proposition. Far too many folk are working at a lower, less-productive level of abstraction (read: "hacking code"). The question is, Why?