Re: Ancient Greek (symbols versus letters and duplicate letters)

From: Edward C. D. Hopkins (
Date: Mon Apr 07 2003 - 17:12:33 EDT

  • Next message: Edward C. D. Hopkins: "Re: Ancient Greek (symbols versus letters and duplicate letters)"


    Since Koppa has the same modern meaning as the ancient character (letter
    Koppa and 90), this runs contrary to my understanding of the Unicode
    philosophy, as I understand it. Is there an established Unicode Greek
    sorting algorithm?

    I can see no reason to have lunate Sigma U+03F9 as a separate codepoint in a
    font. Unless convincing information is received, I think I'll include a
    lunate Sigma only as an alternate glyph to the "true" capital Greek Sigma,
    U+03A3. This should preclude sorting and search problems.

    Many thanks,

    Chris Hopkins

    ----- Original Message -----
    From: "Joop Jagers" <>
    > First the koppas. The ancient Greek form is the lollipop koppa. This is an
    > alphabetic character comparable with Latin q. It also has a numeric value,
    > The lightning koppa is used in modern Greek as a numeral only (=90).
    > You're right: C-shaped sigma is *not* a symbol, but a stilistic variation
    of a
    > regular sigma, and so there's no reason to include it in Unicode. Unicode
    > even adds the capital form. It's mainly used when editors don't want to
    > differentiate between final sigma and normal sigma. Those who'd like to
    > c-shaped sigmas should bear in mind that it has a different encoding which
    > searching difficult.

    This archive was generated by hypermail 2.1.5 : Mon Apr 07 2003 - 17:50:02 EDT