Monday, October 5, 2009

On Big "M" Modeling...

Speaking of Wayne, I, too, was recently interviewed by a German magazine, JAXenter. Following his example, I thought I'd post the English translation of the interview for the benefit of those (like me) that can't read German. It's a little dated (especially since our Helios plan is now in place and we've already started development), but hey, better late than never, right? ;)


Can you describe the Eclipse MDT Project in a few words?

The Model Development Tools (MDT) project focuses on big "M" modeling within the Eclipse Modeling project. Its purpose is to provide implementations of industry standard metamodels and exemplary tools for developing models based on those metamodels.

Can you give a typical use case for the project?

Developers use MDT as a framework for building professional-grade modeling tools based on industry standard languages such as Unified Modeling Language (UML), Object Constraint Language (OCL), and XML Schema Definition (XSD), among others. End users use MDT's exemplary tooling (currently provided by the UML2 Tools and Papyrus components) to develop models based on those languages.

What's new in the Galileo Version of MDT?

Four of the nine components in MDT released as part of Galileo. OCL 1.3 saw the addition of finer-grained control over debug tracing, support for big numbers, and an extensible type checking mechanism. UML2 3.0 subscribed to the EMF "ultra-slim diet", migrated to the latest released version of the UML specification (2.2), and made some extensibility enhancements in the area of profiles. UML2 Tools 0.9 includes enhanced support for working with profiles and stereotypes, vastly improved look and feel, provisional support for sequence and timing diagrams, and the ability to specify which elements are included (synchronized) when creating a class diagram. Finally, XSD 2.5 added support for using an ecore:nsURI attribute on a schema element to specify the namespace URI for a schema without a target namespace.

Where do you see the MDT Project in one year? How will the project evolve after Galileo?

We have yet to prepare our plans for the next major release of MDT (dubbed "Helios"), but I can give you an idea of what I'd like to see happen. First, I'd like to see revival of ODM (Ontology Definition Metamodel) support in MDT (in fact, we already have a proposal), especially given the resurgence of interest in RDF (Resource Definition Framework) in the industry of late (e.g. Linked Data). I also anticipate compliance with the latest releases of industry specifications, particularly BPMN, OCL, and XSD. Integration of EMF-based support for things like searching, comparing, and indexing of models in our end-user tooling should also be a priority. Finally, I hope to see some innovation in modeling tooling and techniques, including support for such things as facet-based metamodels, task-focused tools, and collaborative design.

No comments: