Re: Arabic letters separated by markup

From: Edward H. Trager (
Date: Tue Jun 14 2005 - 11:41:15 CDT

  • Next message: Mete Kural: "RE: Arabic letters separated by markup"

    On Monday 2005.06.13 15:07:19 -0700, Mete Kural wrote:
    > >It should be possible to implement multi-color obligatory ligatures by
    > >creating 2 or more glyphs for each ligature, possibly with kerning. I
    > >haven't checked to see whether any APIs can kern across runs or change
    > >colors within a run, but that's a separate issue.
    > So pretty much do we have the technology today to individually color both regular joining characters and also parts of ligatures then?
    > >We are talking about rather strange cases here, so the implementors
    > >might not get around to implementing them soon even if the specs were
    > >embellished.
    > Yes some of these cases are rare such as someone changing the font or
    > font size within an Arabic word. I mean who in their right mind would do that :)

    Here is a potentially real-world example: As we all know, enlarged initial drop capitals
    are a common feature of higher-quality western typography. For example, initial
    drop capitals are often used at the beginning of chapters in books. CSS and good modern
    browsers now permit the use of "first letter" styling rules such as:

    div#content h1 + p:first-letter{


     ... which specifies that the first letter of a paragraph following an "H1" header in the
    "content" region of a web page
    be displayed in a large size (25pt) and special color (#988682). The intent here is surely
    to allow web-based "typography" (if you will) to achieve results similar to what is
    traditionally quite common in print media.

    If we examine a sampling of modern print media (books and magazines) from around the
    world, it doesn't take long to realize that the embellishment of the initial letter of
    an introductory paragraph is no longer the exclusive domain of western typography.
    It is being done even for Indic-derived scripts like Thai (see my URL below). As you might imagine, things
    become a little trickier for Indic and Indic-derived scripts like Thai which can have glyphs for
    vowels that precede, follow after, sit on top, or hang below the initial consonant of a word.
    How are such vowels (and tone marks, too) handled in the case of embellishment of an initial
    "letter" in high-quality modern typography ? Are initial letter embellishments ever used
    in joined and cursive scripts like Arabic or, say, Devanagari?

    I think these are very interesting questions which should be very relevant to how CSS "first-letter",
    "first-word" and "first-line" specifications are written and, eventually, implemented in
    browsers. These sorts of typographical embellishments are sufficiently common
    in print media all over the world now that, I would argue, this type of typographical ornamentation
    is much more common in general than the case of changing colors of glyphs in the middle of a word (in arabic
    or some other script) in order to highlight a root or for some other didactic purpose.

    In the case of the CSS "first-letter" pseudo-element, I can tell you that current browsers do not
    come even close to what I would expect for the cases that I have examined.
    In fact, quite some time ago, I wrote a little blog about this which I
    think will be somewhat informative in the context being discussed here, and I encourage
    readers interested in these issues to have a look:

    There are a few limitations to this blog, as follows:

     (1) My personal library has mostly Thai and Chinese books among the non-Western scripts, so
         I didn't have any examples to look at for Arabic and Devanagari (two other scripts I was
         interested in at the time of writing (and still interested in)), and I never
         found the time to do additional research
         by taking a trip to the university library and hunting down examples...

     (2) The chart at the bottom of the blog is really incomplete.

     - Ed Trager
       Kellogg Eye Center
       University of Michigan
       Ann Arbor

    > But I think the color change case is more justified since it does not disturb
    > the shapes and joining of the letters at all, simply change the color from the
    > default black to some other color. Also coloring words in red for instance is
    > a common practice found in many Quran printings.
    > Regards,
    > Mete
    > --
    > Mete Kural
    > Touchtone Corporation
    > 714-755-2810
    > --

    This archive was generated by hypermail 2.1.5 : Tue Jun 14 2005 - 11:24:45 CDT