Re: Combining across markup?

From: Peter Kirk (
Date: Wed Aug 11 2004 - 05:43:21 CDT

  • Next message: saqqara: "Re: Combining across markup?"

    On 11/08/2004 10:35, Michael Everson wrote:

    > At 10:19 +0100 2004-08-11, saqqara wrote:
    >> This is one example where colour support in fonts would be useful. A
    >> useful
    >> addition to the OpenType specifications if any readers here have
    >> influence
    >> on such matters.
    > Out of scope. You can use markup to make characters red.

    Not if the glyph is multi-colour, which is common with certain
    representations of Egyptian hieroglyphs. (More to the point I can
    imagine people wanting to put coloured icons, corporate logos etc into
    the PUA of fonts so that they can be used within text - far more likely
    to drive changes to OpenType etc than anything ancient Egyptian.) Nor if
    the colouring is needed to make a plain text distinction, which is
    necessary in at least one script which was discussed here within the
    last year - although I think we agreed that that particular script did
    not justify encoding.

    > ...
    > The answer to the original question about black a with red macron is:
    > You cheat, just like they did in the good old days of lead type. For
    > that matter, just as they did when they had to put down one pen and
    > pick up another.

    You can't cheat within Unicode. This is at least potentially a necessary
    distinction, although whether it is plain text or not is debatable.
    While I don't have an example to hand (although I have seen this kind of
    thing in biblical Hebrew and Greek texts), I can easily imagine a
    scholarly edition of a text which uses one colour for text, including
    base characters and some combining marks, which are in the actual
    manuscript and another colour for marks which have been supplied by the
    editor. Not even Chris' very sensible suggestion

    > <span style="color: rgb(0,0,0); mark-color: rgb(255,0,0);">Some Arabic
    > Text</span>

    can deal with that situation properly. But does that mean that this kind
    of text is to be ruled forever outside the scope of Unicode. I'm not
    saying it should be plain text. But Unicode should be able to support
    markup schemes which do allow such things.

    Peter Kirk (personal) (work)

    This archive was generated by hypermail 2.1.5 : Wed Aug 11 2004 - 05:43:56 CDT