No no no! For the love of $(DIETY), don't call them openstache and closestache, but rather leftstache and rightstache.
While there isn't a difference in left-to-right language, those of us unfortunate enough to have to support right-to-left languages need a nontrivial algorithm to decide whether openstache is actually a leftstache or a rightstache (because the grpahic form of the character is that of a leftstache, independent of directionality, rather than openstache, which mirrors based on directionality)
There is (AFAIK) one programming language that is RTL -- a variation on Basic which has been dead for about 15 years now.
The open-paren / left-paren problem is there even in e.g. Word, explictly _because_ it's considered open-paren rather than left-paren; When you type 'alif' (arabic 'a' equiv) 'ba' (arabic 'b' equiv) shift-9 (left paren), you get ')' 'ba' 'alif' . but if you type 'a' 'b' shift-9, you get 'ab('. if you insert a different directionality character immediately in fromt of the open-paren, the paren will be flip.
The rules for flipping parens are specified in the Unicode standard, and are extremely nontrivial and nonintuitive (both for implementing and for using -- it's often hard to get the kind of character you want!).
And it all exists because some idiot thinking "abstraction! It is an open-or-close that the person means, not the left-or-right!" was sitting in the committees making the decisions; There _was_ dissenting opinion, giving exactly the example I gave above, but abstraction was deemed way more important than usability.
While there isn't a difference in left-to-right language, those of us unfortunate enough to have to support right-to-left languages need a nontrivial algorithm to decide whether openstache is actually a leftstache or a rightstache (because the grpahic form of the character is that of a leftstache, independent of directionality, rather than openstache, which mirrors based on directionality)