bear <xxxxxx@sonic.net> writes:

> On Mon, 22 Aug 2005, Michael Sperber wrote:
>
> I don't think I believe that "phase violations are always errors".
> Why shouldn't I be able to define a new object type at runtime,
> along with specialized syntax that operates on such objects, and
> thereafter use those syntax forms in the same program?

Phases are not (only) about compile time and run time, they're about
making sure that something is defined when it's used.  You violate
that, you're making an error.  Systems that enforce this (such as
Scheme 48 or PLT Scheme) don't take away any expressiveness---quite on
the contrary, they *add* expressiveness for describing the phases.

--
Cheers =8-} Mike
Friede, Völkerverständigung und überhaupt blabla