This paper considers that there are two important views of Model Driven Engineering. One consists in starting with a well known standard universal modeling language like UML 2.0 and to define restrictions and extensions to this all-purpose language by the way of so-called "profiles". The other possibility consists in using small, well-focused Domain Specific Languages and to deal with the coordination between these. We will not discuss here the pros and cons of both approaches but we will consider that each has many supporters while the debate is going on. UML CASE tools currently produce a significant legacy, mainly of UML profiles. Bridges between both modeling worlds will surely be most needed in the future. We consider this bridging problem to be a hard one and we propose to use some advanced modeling tools to solve it. We show how a combination of a model-transformation tool (ATL) and a model-weaving tool (AMW) may help solving this problem. We also show how to take advantage of higherorder transformations in this endeavor.
[1]
Arie van Deursen,et al.
Domain-specific languages: an annotated bibliography
,
2000,
SIGP.
[2]
Frank Budinsky,et al.
Eclipse Modeling Framework
,
2003
.
[3]
Jack Greenfield,et al.
Software factories: assembling applications with patterns, models, frameworks and tools
,
2004,
OOPSLA '03.
[4]
Jordi Cabot,et al.
A simple yet useful approach to implementing UML Profiles in current CASE tools
,
2003
.
[5]
Jean Bézivin,et al.
Principles, standards and tools for model engineering
,
2005,
10th IEEE International Conference on Engineering of Complex Computer Systems (ICECCS'05).
[6]
Junichi Suzuki,et al.
Modeling turnpike: a model-driven framework for domain-specific software development
,
2005,
OOPSLA '05.
[7]
Hiroshi Wada,et al.
Modeling Turnpike: A Model-Driven Framework for Domain-Specific Software Development
,
2005,
MoDELS Satellite Events.