Per Bothner wrote:

> Even in this context of pure data (as opposed to procedure calls),
> builtin keyword support has teh advantage that keywords are
> self-evaluating, and they are distinct from normal symbols.

The whole point of quoted data is that they are not evaluated, so I
see no benefit at all to the fact that keywords are self-evaluating.
In a quoted context, I can just as well use symbols, with whatever
syntactic form you want.  Some people might prefer colons:, others
*stars*, and still others <<ducks feet>>.

> I don't know where this is coming from - I see keywords as completely
> orthogonal to objects.

Since others have asked also -- the paper Classes and Mixins (Flatt,
Krishnamurthi, Felleisen, POPL 98) elucidates this in (a little) more
detail on the first page.  The same goes for optional arguments.

'shriram

PS: Since this is really diverging from SRFI-13, I think it would be
    best to take further messages on this to c.l.s.