Re: Character identities

From: Doug Ewell (
Date: Thu Oct 24 2002 - 01:43:19 EDT

  • Next message: Kent Karlsson: "RE: Character identities"

    David Starner <starner at okstate dot edu> wrote:

    > Likewise, ä is printed as a with e above in old texts.* Would it be
    > acceptable to make a font with a a^e glyph for ä? It's not even
    > changing the meaning of the character in any way.

    Indeed, that is exactly what Sütterlin fonts do. (Then again, Sütterlin
    fonts assign the long-s glyph to U+0073 and make you type $ to get a
    round s, so they may not be the best example.)

    Stefan Persson <alsjebegrijptwatikbedoel at yahoo dot se> replied:

    > Unicode defines "a^e" as U+0061 U+0364 (though it's exactly the same
    > character as "ä"). Why?

    They're not exactly the same, except in this particular German example.

    Combining superscript e was encoded along with combining superscript a,
    i, o, u, c, d, h, m, r, t, v, and x, none of which evolved into a "real"
    diacritical mark the way e did. Combining e had non-German uses as
    well, as in early modern English "Yͤ" (which did not become "Ÿ").

    As for the diaeresis, its use in French, English ("coöperate"), and
    other languages often has no relationship to the letter e. Indeed, in
    the sequence "güe" in Spanish, the diaeresis serves as a sort of anti-e,
    ensuring the separate pronunciation of the u when the e would otherwise
    prevent it!

    Historically speaking, I and J were once equivalent, and U and V were
    once equivalent, but they are all encoded today.

    -Doug Ewell
     Fullerton, California

    This archive was generated by hypermail 2.1.5 : Thu Oct 24 2002 - 02:34:02 EDT