Thu Mar 29 17:50:44 EDT 2007


back to the code generator. the reason i wrote this was twofold. one
is to have a portable target for brood forth. the main idea there is
to rewrite mole into someting more graceful, and have a basis for
(re)writing PF.

and two: i need a language for expressing the signal processing code
in PF. this should not be forth, but a multi -> multi dataflow
language. maybe just forth + protos?

so. i think the next step should be to transform current cgen (poke)
so it has an extensible name space.

maybe it is a good time to look into defining new languages inside
PLT, since that's what i'm doing basicly, instead of mucking about
with explicit environment hashes and interpreters.

something to iron out: it's not a new language, it's a cross-compiler:
you want to define functionality accessible in one name space using
functionality accessible in another name space.