From: Ernest Cline (ernestcline@mindspring.com)
Date: Fri Mar 05 2004 - 18:20:40 EST
> [Original Message]
> From: Michael Everson <everson@evertype.com>
>
> I have a proposal for the i.t.a. in the works; I would not unify it
> with this character, which has a specific use in American
> English lexicography.
But that specific use is the same as the use of the reversed t, h
ligature that the i.t.a. uses and the proposed U+0246 (LATIN
SMALL LIGATURE ITALIC TH) from N2656. While all three
have different glyphs, they all represent a voiced th in English,
and it is extremely unlikely that any document would use more
than one such form. The distinction between the glyphs would
seem to me to be more appropriately encoded by font than by
character.
Even if one admits that there is sufficient distinction between
the i.t.a. voiced th and the other two glyphs to merit a distinct
code point for the i.t.a. character, (an admission that further
reflection convinces me is not unwarranted) the same
cannot be said for LATIN SMALL LETTER TH WITH
STRIKETHROUGH and the proposed LATIN SMALL
LIGATURE ITALIC TH of N2656. Both are the result of a
particular dictionary's attempt to distinguish between unvoiced
and voiced th in a manner that is readable as plain text if the
font variation (strikethrough or italic ligature) is ignored.
Furthermore, in yet another dictionary (Webster's New
Twentieth Century Dictionary, Unabridged, 2nd ed.[World
Publishing Company,1962]) A plain th ligature with the
connection formed by the extension of the horizontal bar
of the t is used to represent the voiced th sound. Making
a consistent font distinction within a dictionary to indicate
the voiced th is part of American English lexicography,
but the distinction used is not consistent between
dictionaries. As such I feel that encoding one character
for the lexicographic voiced th and using a choice of
font to determine the exact glyph would be preferable
to encoding separate codepoints for what is in my
opinion the same character.
This archive was generated by hypermail 2.1.5 : Fri Mar 05 2004 - 18:49:17 EST