From: Karl Pentzlin (email@example.com)
Date: Fri May 04 2007 - 18:16:24 CST
Am Samstag, 5. Mai 2007 um 00:56 schrieb Peter Constable:
PC> Karl Pentzlin [wrote answering John Hudson: ]
>> The @ is a symbol without any case properties. Thus, you compare
>> apples and oranges here [case forms of letter ß].
PC> Declaring this by fiat to be so doesn't make it so. If in some community of users
PC> a typeform distinction for @ is introduced with different forms for lowercase and
PC> uppercase contexts, then it is entirely possible that what you deem to be a
PC> caseless symbol may become for that community a bicameral character pair.
I referred to the symbol "@" as it is encoded in Unicode now, as a
symbol (general category Po: Punctuation, other) with no case properties.
PC> In fact, I can point you to instances in which @ has been used as a lowercase
PC> word-forming letter with an uppercase pair. (Of course, this is among the worst
PC> orthographic innovations imaginable.)
Then, if you want to use such letters in Unicode, these letters are
not the "Punctuation, other" characters U+0040 COMMERCIAL AT.
Instead, a new Latin letter pair should be considered (U+xxxx/U+yyyy
LATIN LETTER CAPITAL/SMALL LETTER COMMERCIAL AT or named otherwise).
This resembles the introduction of U+0241/U+0242 as cased pair in
contrast to the explicitly caseless letter U+0294.
At least in that case, the cased letters were considered different
from the similar looking (to the uppercase letter) existing letter.
Some SIL fonts (e.g. Charis SIL) contain such letters in their PUA,
differently encoded from the similar looking (to the lowercase letter)
U+F247 LATIN SMALL LETTER AT
U+F248 LATIN CAPITAL LETTER AT, see:
(In the SIL PUA documentation:
these characters are marked as "Characters rejected for addition to Unicode".)
- Karl Pentzlin
This archive was generated by hypermail 2.1.5 : Fri May 04 2007 - 18:19:15 CST