Re: Problem with Arial Unicode MS font for BOLD/ITALICS in PDF

From: Philippe Verdy (verdy_p@wanadoo.fr)
Date: Sat Jun 21 2003 - 09:21:11 EDT

  • Next message: Michael Everson: "2003-06-21"

    From: "Michael Everson" <everson@evertype.com>
    > At 16:45 -0700 2003-06-20, Richard Cook wrote:
    > >Of course, in pop e-print, nearly everything that can be done to a
    > >character is done ... including Bold-Ital-Outline-Shadow ...
    >
    > Hey, there's no reason only Latin typography should be filled with vulgarism...

    If you just look at Japanese, you'll see that most Hiragana andKatakana text
    use both narrow (half-width) and square (full-width) style within the same font
    but with distinct characters.
    In Latin text, it is most often rendered with the same characters but with
    different fonts.

    Some typographic effects have different usage patterns between Latin/Greek/Cyrillic/
    Hebrew and Asian text. Italic is one of these, and in HTML its use has been
    deprecated in favor of a more language neutral "emphasized" style (which can be
    right-slanted with Latin/Greek/Cyrillic, but can be a distinct font or typographic
    effect applied to some base font).

    In strict typographic terms, the "italic" style modifier is just a convenience to
    select actually distinct fonts. It's true that Arabic is, in its natural form, already
    right-slanted and would not support being more slanted than it is now.
    Slanting artificially a font which contains characters that are already slanted
    such as Arabic, or should not be slanted like Han is quite disastrous for the
    reader.

    Other typographic effects are much more language neutral such as
    font scaling, and varying the inter-letter spacing (but with interesting problems
    with Arabic and Devanagari where most characters should be kept ligated,
    by changing the length of the horizontal ligating stroke), or adjustment
    of margins, usage of borders around paragraphs, and distinct shading of
    backgrounds.

    When I look at most Asian web pages, Bold and Italic or other typographic
    effects used to emphasize some text is often rendered by using distinct
    colors or gray scales. It's interesting to see the colorful patchwork commonly
    found in Asian web pages, even the most serious ones that consist mostly
    of annotated text...

    Also, the Outline and Shadow styles are much more often used in Asian
    compositions than with Latin, notably for titling. And the range of point
    sizes used on the same page is much more important in Asian pages
    than in Latin/Cyrillic/Greek/Hebrew) pages.

    So any attempt to force mapping a italic or bold style on any text seems
    disastrous. It seems more intelligent to try mapping some conceptual
    styles like "emphasized" or "quoted" or "footnote" within stylesheets
    where actual fonts can be selected according to a language context.

    But this goes far beyond what Unicode can do (stylesheets are better
    studied by the W3C style/CSS working groups).



    This archive was generated by hypermail 2.1.5 : Sat Jun 21 2003 - 09:58:43 EDT