Re: Public Review Issues Updated

From: Philippe Verdy (
Date: Fri May 07 2004 - 06:50:38 CDT

From: "Anto'nio Martins-Tuva'lkin" <>
> On 2004.04.29, 09:27, D. Starner <> wrote:
> > In the Bureau of American Ethnology reports, early Americanist
> > characters are used, and they are used with casing, including the C
> > with stroke. (...)
> Aren't these just U+20D2 U+0043 and U+20D2 U+0063 ...? I'm supposing
> that U+20D2 can be made "intelligent" enough by the rendering engine
> to adapt to the height of the base letter, just like, say, U+0300 is.

I suppose you mean really U+0043 U+20D2 and U+0063 U+20D2?

Note that U+20D2 is explicitly a combining _long_ vertical line overlay, and
U+20D3 is a _short_ one, defined initially as diacritics for symbols (for
example in mathematical notations). What if mathematics makes distinctions
between short and long line overlays? I don't think that a renderer would mix
the two characters which have their own identity, where the short and long
versions should be differentiated.

Instead such diacritic could be useful to represent the CEDI currency symbol, if
there was not an ambiguity of which combining line overlay to use for it
(slanted or vertical).

Also are these two special diacritics appropriate for general usage on any
letters, when it was already agreed that overlaid diacritics would better be
deprecated in favor of encoding separate "precombined" letters?

This archive was generated by hypermail 2.1.5 : Fri May 07 2004 - 18:45:26 CDT