Re: dual-language systems increase modularity

[Matthias Felleisen <address@hidden>]

On Tuesday, November 18, 2003, at 12:07 PM, Miles Egan wrote:

> On Tue, 2003-11-18 at 07:56, Matthias Felleisen wrote:
>> Until further notice, types stand. -- Matthias (a reconstructed
>> type-disliker)
>
> I'd love to hear a little more about your conversion.  What was the
> turning point for you?
>
> I think a lot of what I miss in languages like OCaml, for instance,
> could be provided by better support for introspection and wouldn't
> really require loosening of the type system.

Perhaps you misunderstood. I still believe that types have huge 
disadvantages
for various parts of the world. I just have a much higher appreciation 
for what
they do, how powerful they are, and where they fit into the process. I 
understand
them now can easily translate between Haskellians and Smllers when they 
discuss
types and cross paths.

When you say you want introspection, I bet that you really want 
syntactic
abstractions and possibly "fexprs". Taha (Oregon, Yale, Rice) has 
worked on MetaML,
which is basically a typed version of "fexprs". He claims he has 
macros; you
won't like them. Others are also working on staged typed languages, but
they don't have a useful implementation yet.

That's all of what I want to say in this forum. -- Matthias