Layering MDA: Applying Transparent Layers of Knowledge to Platform Independent Models

Abstract

Model Driven Architecture is a system that holds promise for software engineers, but intimidates them also. This is due to the fact that in order to fully take advantage of MDA engineers must be fluent in UML as well as the languages used to translate their PIM’s into executable. Though UML has been in existence since the mid 1990’s its use for communication among people is eased by shared understanding and inferences, these two things are not available amongst computer interpretation units. That means that in order to satisfy automated translation units SE’s must generate PIM’s and in some cases manually manipulate PSM’s that are obfuscated in the eyes of not so well versed individuals. That is why we propose a system of dynamic markups that can provide details that would otherwise be incomprehensible, by people outside the circle of those highly educated in UML and MDA. These markups are non-translatable and can be specialized to certain groups of shareholders. Finally, a small demonstration of this system is performed in order to allow for a fuller understanding of the process and concept, while clarifying the benefits, also to set the grounds, by which it can be empirically proven.