In 2010 Markus Völter, Eelco Visser, Steven Kelly, Angelo Hulshout, Jos Warmer, Pedro J. Molina, Bernhard Merkle and Karsten Thoms discussed the idea of creating a Language Workbench Competition at the CodeGeneration 2011 conference. Yesterday, the submission period closed and saw the registration of 11 participants:
- An Xtext based effort, for which code is available here
- An MPS based effort, for which code is available here
- A MetaEdit+ based effort, not made available yet
- An EMFText/JaMoPP based effort, which can be tracked here
- An OOMEGA based effort, which can be tracked here
- An solution based on the Whole Platform, described here
- A solution based on Essential, can be found here.
- A solution based on Spoofax, is available here.
- A solution based on Intentional, information for which will be online soon.
- A solution based on Rascal, can be found here.
- A solution based on Atom3, for which the platform is available here.
The competition will evaluate carefully all workbenchs from different point of views, ranging for instance from code generation, to multi-language integration, to language versioning. InfoQ spoke briefly with Angelo Hulshout who explained:
Over the past 5 years, a lot of attention has been given to MDSD, not only as a hype, but also in terms of real developments, on the side of tool development as well as production use. This has resulted in a set of mature and rapidly maturing environments, like the once participating in the LWC. These language workbenches combine the tools needed to define and use DSLs in such a way that MDSD can become mainstream. The LWC11 workshop will allow the participants to show the strengths and weaknesses of their current workbenches in realising a common set of problems found in all projects. The differences and commonalities will be discussed, to the benefit of participants and attendees.
Most recently, several products that use Language Workbenches have emerged to help buid mobile applications: Mobl, Applause or Moppr to name a few. In that particular field, the complexity of vendor SDKs (iOS, Android, WM7...) seem to be unrelated to the simplicity of the solutions people are trying to build. Beyond mobile applications, Language Workbenches combined with the emerging Web based Code editor Framework such as Ace / Cloud9 or Concrete could well signal a new era in software engineering.
Are you using a Language Workbench? do you plan on using one? which one? what kind of problem do you plan on solving with it?