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

> We found that retaining the circular shape of the dot in the comma-shaped
02BC but increasing the size of the character to 140% of the original makes
a good, legible character.

At the time that these were encoded, there was no distinction made in the
appearance of  \u{2019} and \u{2BC} or of \u{2018} and \u{2BB}. And that is
the policy followed by fonts in general, eg:

… ’ and ʼ or of ‘ and ʻ …

Of course, one can design a special font to make them distinct. But if
there is free variation among users as to what they type, and if a majority
of widespread fonts don't distinguish them, it just causes confusion (and
mismatches) to artificially separate them. It's like the old proposal to
distinguish . as sentence terminator from . as abbreviation terminator from
.. as number grouping separator from . as number decimal separator. Nice
idea in theory, but only in theory.

Mark


On Mon, Oct 8, 2018 at 2:49 PM Michael Everson <[email protected]> wrote:

> On 8 Oct 2018, at 13:29, Mark Davis ☕️ <[email protected]> wrote:
>
> > 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 was a bad idea. I’ve been publishing material in the Hawaiian
> language for some time now (Alice, Looking-Glass, the Wizard of Oz, Harry
> Potter, and more in preparation).
>
> It is NOT possible to read Hawaiian narrative texts (you know, with
> conversational dialogue between characters, using quotation marks and
> nested quotation marks) if the quotation marks and the glottal stop are
> identical. Whatever the origin of the modifier letter apostrophe, if it is
> not distinguished from the quotation marks, the text becomes illegible. We
> found that retaining the circular shape of the dot in the comma-shaped 02BC
> but increasing the size of the character to 140% of the original makes a
> good, legible character.
>
> At some stage (in my non-existent abundant spare time) I should write a
> Unicode Technical Note about this. But the status quo is not acceptable to
> actual readers of the language. (Same goes for Sāmoan and Tongan.)
>
> Michael Everson
> _______________________________________________
> Ietf-languages mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/ietf-languages
>

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