Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ


Choose your language

InfoQ Homepage News Combining General Purpose Languages and Domain Specific Languages for Model Driven Engineering

Combining General Purpose Languages and Domain Specific Languages for Model Driven Engineering

This item in japanese

In his latest blog post, Johan den Haan , CTO of Mendix, asks one of the key question of Model Driven Approaches (Architecture, Development, Engineering...):

The basic principle of Model Driven Engineering is that "everything is a model". Models and model elements are given a first-class status. The essential change is that models are no longer used only as mere documentation for programmers, but can now directly be used to drive software development.

... When defining a language for an MDE approach based on the concepts presented above, the question arises for particular elements whether to define them in the [domains specific language] or in the [general purpose language].

Johan argues:

In most approaches a choice is made between a general purpose or a domain specific language. I think both approaches can be combined using the best from both worlds.

The article is also educational. It starts with a clear definition of a model, a metamodel, showing clearly the four M0, M1, M2 and M3 levels of a model driven architecture defined by the OMG. Johan also establishes the difference between a linguistic metamodel and ontological metamodel:

While linguistic metamodels define the language of a model (including the abstract syntax, type semantics, static semantics and dynamic semantics), ontological metamodels define the inherent semantics of a model. Inherent semantics describe the "inner meaning" of modeled resources and provide the basis for reasoning about concepts

He concludes:

Ontology's are essential for defining declarative languages, e.g. languages defining the "what" instead of the "how"... [and] both dimensions [linguistic and ontological] are needed for a complete modeling approach

Johan also explains the importance of ontological metamodel hierarchies. He argues that the hierarchy should start at the technology layer, from which the domain ontology depends on, from which models depend on. He then links linguistic metamodels to the generation of software artifacts.

When looking at Object Orientation, Johan sees that:

  • keep the language definition small and simple,
  • keep the virtual machines stable even in the event of considerable (library) concept restructurings,
  • give users maximum tailorability (users may change or extend a library but cannot change the core language concepts)
  • .

Then Johan takes the example of "Business Process Modeling":

A lot of discussions are going on about the use of BPMN (OMG, 2006) and BPEL4WS (OASIS, 2007) as general purpose languages for business process modeling. While BPMN is used for defining business processes for analysis and documentation (in general), BPEL4WS is used to define executable business processes.

Johan describes a precise metamodel hierarchy relating a model layer, a domain layer, a Services layer, BPEL as the abstract, platform independent concepts, and BPMN as the language layer. He argues that:

defining a transformation [between BPMN and BPEL] is made very easy by using this metamodel hierarchy.

Johan concludes that

[the] design decisions [he has] made are very powerful for defining domain specific languages (DSL's). First, the resulting DSL's are easy to transform into working software. Second, the approach [he has] shown enables the combination of concepts from general purpose languages and domain specific languages.

Rate this Article