[racket] Again on bindings visibility in eval

>
> P.S.  I'm enjoying this discussion.  I feel like I'm finally getting a
> handle on what kind of how "syntax" is represented and manipulated in
> Racket.  It still seems like an awful lot of work, though, so I would
> like to see more examples of why it is worth the

I wrote a package the other day to simulate Python's chained
comparisons syntax; it's on PLaneT and shows both the use of the
syntax objects in preserving source locations, as well as using the
hygiene to bind the values between the binops to avoid duplicate
evaluation.  Syntax objects are a bit rich because they're doing a lot
of work.