Re: Joining hamzah

From: John Hudson (john@tiro.ca)
Date: Tue Aug 26 2008 - 21:39:36 CDT

  • Next message: Arno Schmitt: "Re: Joining hamzah"

    Andreas Prilop wrote:

    > Is it okay to write the sequence

    > U+0640 U+0654

    > for a "joining hamzah"? Example:

    > مسـٔلة

    My main concern with this approach is that it makes an encoding
    distinction based upon visual display that results in the same semantic
    character being encoded in two different ways depending on how you want
    it to appear.

    I also think it may be an unreliable mechanism depending on the
    rendering engine and font technology, since it presumes that U+0640 is
    rendered as some kind of glyph. In some script styles and, hence, font
    implementations the tatweel character is dumped during display and only
    represented, if at all, by elongation of the connecting stroke that
    forms part of the preceding (and/or following) glyph. This means that
    there may be situations in which e.g. a word such as

            مسـٔلة

    may be rendered with the glyphs

            مسٔلة

    in which case U+0654 will be rendered above the preceding letter rather
    than between the letters where it belongs if it is intended to represent
    the letter otherwise encoded as U+0621.

    The U+0640 'tatweel' is generally problematic, since it is an encoding
    of a piece of metal rather than a semantic character. It is an artefact
    of a particular technology for typesetting a particular style of Arabic
    type. It shouldn't have been encoded, and I think it should be avoided.

    John Hudson

    -- 
    Tiro Typeworks        www.tiro.com
    Gulf Islands, BC      tiro@tiro.com
    I should be very surprised if any of those who are
    daily trying to imitate hand-cut type by mechanical
    means would be prepared to accept, say, a frying pan
    with so-called hammer marks if the hammer marks came
    out of a mould and the pan were forced on some machine.
                        -- Jan van Krimpen
    


    This archive was generated by hypermail 2.1.5 : Tue Aug 26 2008 - 21:43:13 CDT