Re: MS Windows and Unicode 4.0 ?

From: Peter Kirk (peterkirk@qaya.org)
Date: Wed Dec 03 2003 - 15:51:08 EST

  • Next message: Peter Kirk: "Re: MS Windows and Unicode 4.0 ?"

    On 03/12/2003 12:44, John Hudson wrote:

    > At 04:30 AM 12/3/2003, Peter Kirk wrote:
    >
    >> An adequate proposal for a complex script should surely include a
    >> proper account of the script behaviour and sample glyphs of
    >> presentation forms. And so such a proposal should include all that is
    >> needed for a developer, and is available some time before the new
    >> script is officially standardised.
    >
    >
    > I disagree. What you describe may be desirable, but in no way is it
    > necessary. What is important to document in a proposal is what is
    > necessary to *encode* text, not to display it. Remember that a lot of
    > work was done on encoding complex scripts in Unicode before there were
    > adequate font and shaping engine technologies in place to implement
    > the character/glyph model as envisaged. Also, for some complex
    > scripts, especially Arabic, how do you define what is 'needed for a
    > developer' independent of the particular script style, individual
    > typeface design and specific rendering technology? What is needed for
    > Tom Milo to render Arabic using his technology is quite different from
    > what is needed to render the same text in the same style in a typical
    > OpenType implementation.
    >
    >
    Understood. But in fact TUS contains some quite detailed rules for
    Arabic shaping. I am only asking that there should be similar rules for
    other scripts.

    As for different script styles, there may of course be variants but each
    script has a typical variant, and any major deviations from that should
    be considered a separate script. I am not suggesting normative rules,
    just implementation recommendations.

    -- 
    Peter Kirk
    peter@qaya.org (personal)
    peterkirk@qaya.org (work)
    http://www.qaya.org/
    


    This archive was generated by hypermail 2.1.5 : Wed Dec 03 2003 - 16:44:08 EST