Re: Click letters in ISO 639-3 and Registry (long)

Arthur Reutenauer <[email protected]> Mon, 8 Oct 2018 14:45:49 +0200
Newsgroups gmane.ietf.languages
Message-ID <[email protected]>
On Mon, Oct 08, 2018 at 02:29:50PM +0200, Mark Davis ☕️ wrote:
> Revisiting this because I happened on
> =/Kx'au//'ein
> which caused a spreadsheet to hiccough (not prepared for initial ASCII =)..
> It would be far better to use ≠ rather than =/.

  Since ≠ (U+2260) is neither latin-1 nor ASCII, is there any advantage
to refraining from using the real character (U+01C2 ǂ)?

> I would really like to see all the ASCII hacks removed, and replaced with
> either the correct character or an escaped version thereof. BTW, as for
> 02BC, in CLDR we ended up mapping that to 2019, because nobody is
> consistent about the distinction. (The danger of having two identical
> characters, not separated by script.)

  That may be wise; there is definitely a grey area with many ambiguous
cases.  Around the time of that discussion I researched some of the 100+
language names from ISO 639-3 containing apostrophe-like characters, and
it was sometimes really hard to decide between U+02BC and U+2019, in
particular.  (It was however really instructive to investigate all these
language names.)

	Best,

		Arthur

_______________________________________________
Ietf-languages mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ietf-languages