> Consider Kent's implementation of cond in the reference
> implementation using the common extension:
>
>   (define-syntax (cond x)
>     (syntax-case x ()
>       [(_ c1 c2 ...)
 > ...
>                 [_ (syntax-error x)]))]))]))
>
> With define-syntax-case (at least if I understand you correctly)
> it becomes:
>
> (define-syntax-case cond ()
>   (x (syntax-case x ()
>       [(_ c1 c2 ...)
 >        ...

You could do that.  But since syntax-error is not in the specification,
I'm not sure it is relevant.  If you leave out the syntax-error call,
you get a worthwhile simplification:

(define-syntax-case cond ()
   ((_ c1 c2 ...) ...))

Note also that if you *do* want syntax-error, it might be better to
report error location more specifically that the entire cond.

> Adding an extra binding form also adds to the cognitive load
> for those learning the macro system. For the experienced
> macro writer it hardly matters which is used.

I don't know.  Scheme API design has this bad tradition of
designing APIs for maximum verbosity of the common case.  call/cc is
another example, given that 90% of the time the argument is a
lambda expression, which makes code that can be hard to understand
even more difficult.
--
	--Per Bothner
xxxxxx@bothner.com   http://per.bothner.com/