http://www2.parc.com/csl/groups/sda/

"Our goal is simple and long-standing: to make it possible to cleanly capture complex design structures in software implementations....

"The thread that has been common to our projects is ... that the normal rules of modularity do not suffice for describing quality software."

More to the point,

http://www.cs.ubc.ca/labs/spl/projects/aspectc.html

"AspectC is a simple aspect-oriented extension to C, intended to support operating systems and embedded systems programming."

Bill Cox should be playing with something like ANGIE or GSLGen, his troubles are probably better served by such tools. http://www.delta-software-technology.com/neu/pages/pageseng/common/prod_frmset_nge.htm http://www.imatix.com/html/gslgen/

Although the aspect folks do make interesting remarks on their Explicit Programming page,

http://www.cs.ubc.ca/labs/spl/projects/explicit.html

Mark Evans wrote:
> More to the point,
> 
> http://www.cs.ubc.ca/labs/spl/projects/aspectc.html
> 
> "AspectC is a simple aspect-oriented extension to C, intended to support
> operating systems and embedded systems programming."
> 
> Bill Cox should be playing with something like ANGIE or GSLGen, his troubles are
> probably better served by such tools.
> http://www.delta-software-technology.com/neu/pages/pageseng/common/prod_frmset_nge.htm
> http://www.imatix.com/html/gslgen/

Well... these tools are interesting, but DataDraw serves my needs in this area better.  Aspect programming, while interesting, is also not what I've been looking for.

I've come to the following conclusion:  no language, is going to include all extensions I need.  D is the closest thing out there from a traditional compiler point of view.

XL, on the other hand, let's me add the features I need.  While I'm not excited about the XL implementation, or the way they've based it on ML, the basic idea of an eXtendable Language is right on.

What I would like is an extendable C.  In this language, C++ could be a library (or at least something close to C++).  Or, you could pick and choose your features: GC, dynamic class extensions, templates, framework inheritance, and on and on...

If someone were to write it, I'd call it Z, since it would encompass virtually the entire space of C derived languages that can be written well as translators into C. (Z is taken, but not by a general purpose computer language).

Bill

...and they have publications on a framework called 'D' dating back to 1997...