> Now, there _is_ possibly one way of doing explicit names that would be
> fast enough to be a contender -- where the names are chosen uniquely,
> using a global counter, so no capture check is required.  I can
> imagine that that would be fast.

This is pretty much how the Pict front end works.  We started out with a
standard deBruijn representation, but later switched to a scheme where,
during parsing, we choose a globally unique spelling for each bound name
(they're not *that* ugly).  We have to do a little work later to make
sure that uniqueness is maintained, but it seems faster on the whole.

       B