Re: Back to Coptic (was: Demystifying the Politburo)

Date: Fri Jul 08 2005 - 04:56:48 CDT

  • Next message: Erkki Kolehmainen: "Re: Arabic encoding model (alas, static!)"

    ----- Original Message -----
    From: "John Hudson" <>

    > Doug Ewell wrote:
    > > I thought Michael had answered the question: Greek CHI RHO (U+2627) and
    > > Coptic KHI RO (U+2CE9) are encoded separately because they belong to
    > > different scripts, sort of like Latin P and Greek Π and Cyrillic П.
    > > U+2627 is commonly identified as a symbol made up of two Greek letters,
    > > although it is not encoded in the Greek block.

    Well, to me it is no more Greek than Latin or Coptic whatever its origin. It is a
    universal symbol used in many scripts (also Latin), it should not be decomposed
    and analyzed if it shares the same meaning, the same appearance, the same name,
    the same typographical behaviour and the same origin (a KHI and a RO, after all
    these Coptic letters are also Greek in ancestry). I find this disunification abusive.

    Analogies rarely completely work, but should we find as many NUMERO symbols
    U+2116 as there are scripts using this symbol (Latin, Cyrillic, Greek, etc.)?

    > This is my understanding also: once the decision had been made to
    > disunify Greek and Coptic, it made sense that this disunification
    > would also apply to Chi Rho / Khi Ro.

    I hear you. And I'm willing to accept this is the consensus although I disagree
    with (or do not understand, in meeker terms) the reasoning behind the
    disunification of this kind of universal symbol.
    > Regarding the other Coptic question, I'm inclined to think that
    > U+2CEA might be presented either with or without the bar, and that
    > this is a glyph design decision. Personally, I would include both
    > forms in a font, but would probably make the form with the bar the
    > default glyph simply because it seems much more common. The issue
    > is a little clouded by the separate encoding of the abbreviation
    > bar, but there is at least a partial parallel in the European
    > ordinals, which may be written with or without a bar beneath them.

    I would tend to agree with you: change the reference glyph as it
    may not be unencoded. In its current form in the chart it is just a
    ligature like fi or ffi in Latin and should thus most probably not have
    been included. I suspect it was included because the proposers did not
    know or did not want to depend on newer technologies like OpenType
    which could easily compose this abbreviation in all its forms (ligated or
    not, flattened or not, with or without abbreviation bar) given the basic
    Coptic letters and the abbreviation bar.

    Note that Mr Everson did say the opposite: one has to manually add
    the bar over this ligature to obtain the Lord abbreviation and thus it is
    not a matter of variant glyphs (one with, the other without the bar).
    Although, if one has to add the bar, it don't see what the U+2ECA
    ligature could mean by itself but again Lord! I have never seen the
    CS ligated in any other context.

    Ashraf Sadek

    This archive was generated by hypermail 2.1.5 : Fri Jul 08 2005 - 04:57:29 CDT