Re: Dylan Constants

[Bob Kerns <rwk@crl.dec.com>]

    Date: Wed, 21 Oct 1992 08:57:30 EDT
    From: duncan@stowe.bdsw.com (Rich Duncan)
    >The Dylan people at Apple have indicated that this will be fixed.  I don't
    >know if they've yet settled on a choice of syntax, but I believe the
    >current plan is that *all* the "define" forms will produce constants, and
    >some new form will be created for defining non-constant variables.
    
    Probably a good choice since that will make DEFINE semantically closer
    to #DEFINE in C and %REPLACE in PL/I.

Eh?

I'm not sure what you think #define does, or what you think
DEFINE does, or what bit of the semantics you're comparing,
but in my ontology this doesn't compute.

You can re-#define #define's; they're not constant at all.
They perform semi-syntactic argument substitution, preserving
semantics not one whit, and often doing violence to the syntax.

It seems that instead this makes it closer to 'enum'.

I would hope that Dylan never does anything even vaguely resembling
the non-semantics of #define!