Re: Public Review Issue 232 Proposed Update UAX #9, Unicode Bidirectional Algorithm (Copy of email sent to the list; also posted by me to unicode feedback/public review issue-- but this has not yet posted there)

    These sorts of "absorption" rules are discussed in great detail in
    Geoffrey Nunberg's "The Linguistics of Punctuation", which I highly
    recommend for anyone interested in this and related issues.

And sometimes there is no absorption but simply a hard constraint
against semantic cooccurrence [sic about "oo", which is really the
standard]. For me, the biggest annoyances are:

  * lack of ?: and of the less important !: (saying that ?: is merged
    into : is possible, but the reader wouldn't be able to tell)
  * the hard constraint against commas on the inside of a parenthesis:
    (, and ,)

What I just wrote in my other email
     "[...] but (as most people here will know), there it has a
different function."
is actually a punctuation mistake (there is descriptively no room to
maneuver here): with the parenthetical phrase, there is a strong need
for a comma before "there" (though there's a bit of wiggle room wrt
whether it's truly obligatory), but without the parenthetical, no comma
would be used (to the point where I'd assume the writer's grammar is
poor). We have a weird situation here: If the parentheses are assumed to
absorb commas,
(1) the absorption is invisible to the reader (he won't be able to tell
this apart from the comma-less case, because there are no clues elsewhere),
(2) one would naturally assume they do so equally on both sides (though
I suppose a sophist might try to argue this one),
(3) and thus "but (as [...] know) there" would prevent a
non-parenthetical reading with only a comma on the right (that separates
a subordinate clause from a subsequent main clause).
If the parentheses are assumed to not absorb commas, the "but, there
[...]" subreading would be incorrectly punctuated; again, a sophist
might argue the comma before "there" (in "but (as [...] know), there")
is less necessary because of the visual grouping by the parentheses.
There is simply no way to punctuate this with the intended semantics.
Very annoying. We'd need ",)".

But everyone is familiar with the much more common case of one wanting
to write "(, " (and space absorption doesn't work here) or ",)" in lists
with a parenthetical element.

Stephan
Received on Sat Feb 02 2013 - 17:27:14 CST