I think this note is specifically for the implementers of the srfi, suggesting that further down the road to R7RS-large we'll expect to have hash sets/bags in srfi-113 API, correct?  For those who want to write portable code using srfi-113, they still need to pass both ordering predicate and hash functions, since srfi-113 itself doesn't give the guarantee nor the way to probe it, and this note may just be confusing.

How about clarifying that the latter part of the note is for implementers, and state the context (that we're hoping some future srfis/standards will adopt guaranteed acceptance of hashable comparator)?