Dear editors: Please post this update to SRFI-105.

Nothing like an announced deadline to get last-minute comments in :-).

This version has a bunch of minor stuff, again, it does *NOT* change the semantics of the specification at all:
      - Fix URL for Paul Graham comment about infix being "natural"
      - Clarify text noting that the mappings like e(...) MUST NOT be applied
        if there's a space between e and the open paired character.
      - Add MUST for recursion left-to-right, to clarify that this is required
      - Make mapping discussion of f{...} a little more consistent
      - Note that datum comments are from SRFI-62, R6R7, and R7RS draft 6
      - Add {...} around example {#1=f(#1#)}
      - Emphasize in rationale that curly-infix allows ANY symbol to be used
        as an infix operator (that's the difference, not that you can use it)
      - Move text about symbols in infix.plt, to make it clearer
      - A few minor style improvements
      - Add text about $nfx$ and $bracket-apply$ based on a
        guile-devel post by Mark H. Weaver on 2012-10-26

--- David A. Wheeler