[R6RS] multiple-value binding

[dyb at cs.indiana.edu]

> > Right.  That's what I meant by "gumming up let".
>
> If we're thinking about un-gumming LET, shouldn't we consider moving
> out named LET into a separate form?

We were thinking about not gumming it up further, but un-gumming it is
worth considering too.

> (Kent---in Bloomington, you
> mentioned to me the primary reason why it's sitting in LET is that
> someone didn't want to give up the name RECUR.  Is this
> characterization correct?)

That was the reason given by the most strong opponent, if not the only
opponent, when a proposal was made to use the name recur for the construct
instead of overloading let.

> I'm sorry if this is too big a can of worms.  I sure often write
> programs with named LET, but when I need to explain them to people
> with little prior Scheme knowledge, I usually wish I hadn't.

I've become used to name let over the past 20+ years, so I'm not sure
whether I'd vote for a change at this point.  I'm interested in hearing
what other people think, however.

Kent