On Tue, Dec 13, 2016 at 9:59 AM, Marc Nieper-Wißkirchen <xxxxxx@nieper-wisskirchen.de> wrote:

One way to cure the defect could be to demand that set-comparator and bag-comparator provide comparison procedures at least when the implementation of sets and bags is based on the comparison function of its elements (and not only on the equality predicate and the hash function).

My inclination is to leave this as a quality of implementation issue.

-- 
 John Cowan          http://vrici.lojban.org/~cowan        xxxxxx@ccil.org
It's the old, old story.  Droid meets droid.  Droid becomes chameleon.
Droid loses chameleon, chameleon becomes blob, droid gets blob back
again.  It's a classic tale.  --Kryten, Red Dwarf