Why MDx has failed (and how it might yet prevail)

Abstract

In the early 2000s, Model Driven approaches to software development were all the rage. (We will collectively refer to such approaches as MDx). Parts of the software industry, and certainly many software engineering researchers believed that MDx would greatly improve the level of productivity in industrial software development. Then, this approach fell out of favor. But why? Did it not deliver on its promise, and if so why? Or where there external factors that stopped the dissemination of this approach? And, what do we consider to be a „failure“ to begin with? In this talk, I will visit the research on this question, and specifically add an industry perspective. I will also present an idea of how modeling may rise to greater prominence, once more.