[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[news.eclipse.technology.ldt] Re: Beyond textual represenations...

Guillaume,

We'd be ever so grateful to increase the sales of the EMF book:
Eclipse Modeling Framework
Eclipse Modeling Framework Sample Chapter
I personally don't like that word meta-model much, but we did use it in the overview paper:
However, to avoid any confusion, the MOF-like core meta model in EMF is called Ecore.
I find the term is kind of confusing and I tend to think that a model is a model is a model.  I'm pretty sure that focusing on the Ecore meta-model in the overview paper would just turn most people right off because most of our clients care only about their own models not about Ecore (until sometime later).  But Ecore is indeed a UML-like model, and UML itself is a meta-model, right?  EMF does support strong Java binding for Ecore instances but that is in addition to being able to emulate the all model instance behavior dynamically without generating any Java code; this is because the full EObject API can be supported either reflectively or with generated code.

I'm rather fond of the model diagrams in the JavaDoc.  Almost everything you need to know about the Ecore API  is available directly from the second diagram:
API

Guillaume Pothier wrote:
Ed Merks wrote:
Guillaume,

I'm curious why you consider the EMF overview paper to be misleading?   I.e., how might we improve it? Perhaps I'm biased, but I think that upon close examination you'll find that EMF provides a very good basis for modeling abstract syntax trees and graphs.  In fact, I have a hard time imagining what an AST node would need to provide that isn't already in EObject...



The document I'm talking about is "The Eclipse Modeling Framework (EMF) Overview". Maybe I missed something but I did not read anything about EMF being a meta-model. Reading this document, it looks more like EMF is a UML-like model with strong Java bindings...
But I will definitely read other documents. Any reading suggestion in particular, which will help me understand EMF's 'meta' nature?

Regards,
Guillaume