From: David Starner (prosfilaes@gmail.com)
Date: Tue Jan 13 2009 - 19:59:49 CST
On Tue, Jan 13, 2009 at 8:10 PM, Michael D'Errico <mike-list@pobox.com> wrote:
> Michael Everson wrote:
>>>
>>> In a discussion regarding the possibility of assigning 26 code points to
>>> be used in pairs to encode country flags, such as <FLAG C, FLAG A> to
>>> specify the Canadian (CA) flag, Michael Everson wrote:
>>>
>>>> Not even MILDLY tempting as an encoding model.
>>>
>>> It took a while to figure out how we could be in such disagreement, but
>>> I think I finally did.
>>
>> I don't.
>
> Well, then, it would help if you'd state your reasoning so one does not
> need to guess.
I'm not Everson, but I think I know. Characters like the combining
Hangul are troublesome; they add extra complexity to display, can be
an annoyence in editing and causes fusses when something accidentally
breaks them apart. In exchange, for Hangul, it maps cleanly to the
entry method and to the underlying structure of the script.
Furthermore, had precedent and influence gone another way, a few rows
of combining Hangul could have avoided 11,000 characters of
precomposed Hangul--and I understand the 11,000 precomposed characters
is not complete and that archaic Korean can demand the use of the
combining Hangul.
On the other hand, your combining flag characters don't save all that
many code points; 676 points just isn't that much outside the BMP.
They map to an arbitrary naming system, instead of the logical
structure of the characters; in fact, there really is no clean logical
structure to the characters. And precomposed characters could do
everything that your combining flag characters can do. If a full set
of flags is encoded--I find unlikely, as I'd put my preference and
slight odds towards no flags being encoded, at least in the near
future--676 code points will simply be provided, at the savings of a
bit of complexity.
This archive was generated by hypermail 2.1.5 : Tue Jan 13 2009 - 20:01:36 CST