Re: Properties

From: Ken Whistler (kenw@sybase.com)
Date: Tue May 17 2011 - 17:10:06 CDT

  • Next message: Andrew West: "Re: Properties"

    On 5/17/2011 1:04 PM, fantasai wrote:
    > I'm looking for this information in order to complete this appendix:
    > http://dev.w3.org/csswg/css3-writing-modes/#script-orientations
    >
    Well, that is part of the conceptual problem we are having here, then.
    That draft appendix refers to the concept of "vertical scripts in
    Unicode 6.0"
    as if that concept were actually defined normatively in Unicode 6.0,
    and the problem is merely acquiring the precise listing somehow.
    But Unicode 6.0 (or any earlier version) does *not* define "vertical
    scripts".

    This is handled better in the css3 text by the beginning of Section 5,
    which divides scripts into three types, defined right there:

    horizontal-only
        Scripts that have horizontal, but not vertical, native orientation.
        Includes: Latin, Arabic, Hebrew, Devanagari
    vertical-only
        Scripts that have vertical, but not horizontal, native orientation.
        Includes: Mongolian, Manchu
    bi-orientational
        Scripts that have both vertical and horizontal native orientation.
        Includes: Han, Hangul, Japanese Kana

    I think the horizontal-only definition is fine.

    The vertical-only definition is also o.k., but the exemplars are not.
    "Manchu" is
    represented (in Unicode) with the Mongolian script -- it is not a
    separate script,
    but rather a separate language, written with the Mongolian script with
    some extensions
    not needed for the Mongolian language. Instead, add Phags-Pa, which is
    the other script in Unicode with "vertical-only" orientation as defined
    here.

    That leaves the bi-orientational scripts, which essentially consist of
    Han and
    all the other siniform scripts or scripts which developed in the sphere of
    influence of Chinese graphology and typography. They all got printed in
    "Chinese books" using Chinese typographical conventions, which started
    out with vertical orientation, and which have, in the recent era,
    switched over
    to left-to-right horizontal layout.

    If you want to have a concept of "vertical script", which apparently means
    "any script which has a vertical native orientation" (i.e., the union of
    vertical-only and bi-orientational), that can be a CSS3 definition, I
    suppose.

    Appendix B should then have *two* tables. The vertical-only table would
    include only Mongolian and Phags-Pa, both of which would have a rotate
    transform. The bi-orientational table would include Han and all the
    CJK-influenced
    scripts, all of which would have a translate transform.

    The script value "Hrkt" should be omitted. There are no characters
    with that Script property value in Unicode 6.0 (nor have there ever been,
    in any prior version of the standard).

    And I don't see what the point of including Egyptian hieroglyphics in
    this table is. Yes, Egyptian hieroglyphics are laid out both in horizontal
    (right-to-left or left-to-right) and in vertical directions. But in
    either case,
    Unicode plain text is not enough. You need a specialized markup system
    on top
    of the plain text to define glyph stacking and grouping, and *then*
    you have the units which can be laid out. In terms that might make sense
    for CSS, Unicode does not define grapheme clusters for Egyptian
    hieroglyphs; and if the layout for CSS is defined in terms of grapheme
    clusters, then there is a hole in the definition. It would be a little like
    trying to use CSS to lay out mathematical formulae or music using
    Unicode characters as plain text.

    For vertical layout of Egyptian hieroglyphics, you would have a further
    problem that the glyph groupings then have a kind of tate-chu-yoko
    behavior, and within those runs, the direction can either be
    all right-to-left or all left-to-right, depending in part on the placement
    of the vertical line of the text (on the right or left edge of a door frame,
    for example). I just don't see the advantage of trying to deal with that
    complexity in CSS, given that people still won't be able to lay out
    hieroglyphics properly. But if it is included in this table, note that
    Cuneiform also has numerous instances of vertical layout.

    --Ken



    This archive was generated by hypermail 2.1.5 : Tue May 17 2011 - 17:15:00 CDT