how it all started

The original reason to start this blog is leaving a “paper trail” of my design decisions around a project I’ve started,

Provide an Eclipse IDE generation environment derived from a language grammar

What’s it about? First go to the ongoing Google Summer of code (2007) projects for Eclipse . From there, just dig into this project description

The motivation for the whole thing is lowering the learning curve to tooling a custom DSL. For example, a DSL more readable than BPEL, or a DSL that wall-papers over the many EJB3 APIs, and so on. It’s not that difficult to design one’s own DSL (especially if one pays attention to abstract syntax, well-formedness rules, and type checking, all achievable with metamodeling and OCL). What’s missing then is the IDE for one's brainchild.

There’s growing interest in Domain Specific Languages (fuelled by Model Driven Architecture). Martin Fowler describes in “A Language Workbench in Action” the kind of DSL definitions that current tools support, plus or minus some features.
http://www.martinfowler.com/articles/mpsAgree.html

Before starting this project I just wanted to be a user of (let’s call it) language engineering technology. So I reviewed what research teams with a lot of experience had to offer for Eclipse, and I realized that my needs were not high in their priority list. In the end, someday that will change, so check from time to time what’s new at places like: http://www.meta-environment.org/ and http://harmonia.cs.berkeley.edu


Back to my project. From the start, I can identify two problem areas:

1. the mechanics of programming a text editor under Eclipse, and
2. other prototypes nearby this technology niche
   

that are described in more detail in the next two blog entries.