>> * Why do we need a new condition type, disjoint from other Scheme
>> values, with 5 new operations, some of which are quite non-trivial?
>>[...]
> Shriram Krishnamurthi <xxxxxx@cs.rice.edu> responded:
>> All of this is also true of continuations.

I agree with Per and Dave: exceptions are nice but conditions
can be dropped / moved to another SRFI. The best abstraction
to recognize particular kinds of values is a predicate; specifying
exact representation, no matter how "normal" it is, just reveals
unnecessary implementation details.

-- Sergei