From: verdy_p (
Date: Wed Aug 12 2009 - 12:57:14 CDT

  • Next message: Christine Snow: "Japanese font that includes Italics"

    "Asmus Freytag" wrote:
    > The two glyphs for 'a' or for 'g' are distinct only in IPA, but in that
    > context, the distinction is important enough to require distinct
    > characters, because accidentally substituting one for the other would
    > destroy the meaning of the IPA text.

    This is true for the two glyphs for 'a' (the other one is small letter alpha, both are used distinctly in IPA to
    denote distinct sounds, so the standard Latin letter 'a' should not be rendered with a single eye with some styles,
    notably those slightly italicized, where it could be easily confused with the letter alpha). I can bet that, for
    this case, a variation selector for the Latin letter 'a' can be used to specify to fonts and renderers that they
    should not render it with a single eye.

    But wrong for the separate glyph for 'g' (IPA currently only needs and uses one, ***preferably*** the single-eyed
    form, but nothing will be broken if you still use a standard Latin small letter g that can be rendered freely with a
    single or double eye).

    I've not seen any contrasting pair in IPA to denote distinct sounds in IPA between the single- and double- eyed
    versions of 'g'. The actual distinctions use other letters (including for example the Greek small letters gamma,
    chi, rho...), or additional diacritics (e.g. for lengthening, or strengthening, or rhotacizing or lateralizing the
    base sound). This will remain true as long as IPA dose not need a distinction with another glyph for the preferably
    double-eyed 'g' that will then need to be encoded separately.

    In that case, the variation selector will not be an option, because there will effectively be a contrasting pair,
    and the existing single-eyed small letter g encoded for IPA does not warranty a distinction with one of the two
    possible glyphs for the standard letter 'g'. And for this reason, it will be best to encode the doule-eyed 'g'
    distinctly for IPA. At that time, there will still remain a lot of legacy uses of the legacy (with dual glyphs)
    Latin small letter 'g' used previously without any variation selector, and that will possibly be rendered using
    either glyphs : a variation selector will not help solve the ambiguity that could occur with lots of legacy fonts,
    but there will be a clear indication of which fonts actually support the new IPA symbol, and which don't, so it will
    be possible to identify systems that are still not upgraded with a reliable IPA font to support this distinction.

    This archive was generated by hypermail 2.1.5 : Wed Aug 12 2009 - 13:00:23 CDT