On Wed, 7 Sep 2005, Per Bothner wrote:

> I think it would be important to also show how SRFI-39-style
> sharing is handled.

Since sharing is not part of the standard yet, I would like to defer
implementation of this.  If an implementor would like to take upon himself the
pain of doing this, I would be glad to add a link to his implementation in
the document ;-)

> It would also be nice if pairs output from
> the expander would be eq? with pairs in the input, when that
> is the case with existing implementations.

I'm pretty sure I understand neither the formulation of, nor the reason for,
this desideratum... :-)

Cheers
Andre