It is a known fact that embedded system complexity doesn't stop increasing and model-driven software development turns more and more out to be an adequate approach to keeping this complexity under control. But it is still a major hassle to provide corresponding integrated tool environments which are sufficiently domain-oriented, scalable, and robust. Eclipse as an open and extensible tool platform is a very good starting point but not a solution per se. The Eclipse Modeling Project provides many of the additionally required building blocks but is very fragmented and not up to the point where it could provide the level of tool support embedded system designers expect.
The recently created Sphinx project is meant to significantly improve this situation. It will be seeded with code from Artop – a platform for creating modeling tools supporting the AUTOSAR standard which has already been deployed at several larger automotive companies. Complementary building blocks will come from the Papyrus project. Sphinx will provide a modeling tool platform enabling IDE-like tool support for specific modeling languages to be developed at reasonable effort and cost. It will leverage and foster the industrial strength known from Artop and head towards a domain- and vendor-independent interoperability layer (backbone) for modeling tools of any kind.