Thu Mar 31 00:21:45 EDT 2011

.f files and early binding

You know, I went through all this work to be able to use units and not
have to rely on context-sensitive "include", but I have to say that in
the light of incremental upload and late binding, that approach is
quite valid.  I mean, it's sane: you can load the same code multiple
times with different bindings without fear of changing anything in the

However.  Big caveat.  You can't redefine things in Forth code
implemented in racket modules, because it uses a different binding

It would be great to be able to unify those two approaches, i.e. using
some form of lexical nesting to in the racket modules to get to the
same obscuring behaviour.

It looks like it's only n@f+ and n@a+ that need an additional wait-ack