RE: MS Windows and Unicode 4.0 ?

From: Michael Everson (everson@evertype.com)
Date: Wed Dec 03 2003 - 05:53:21 EST

  • Next message: Michael Everson: "Re: MS Windows and Unicode 4.0 ?"

    At 01:44 +0100 2003-12-03, Philippe Verdy wrote:
    >Michael Everson writes:
    >> At 15:14 -0800 2003-12-02, Patrick Andries wrote:
    >>
    >> > > Actually, if you look at the Last Resort Glyphs (at a large enough
    >> >> size) you will see that the block name and range numbers are part of
    >> >> the image. See http://developer.apple.com/fonts/LastResortFont/
    >> >
    >> >I believe the name is in English.
    >>
    >> That's correct. I tried to get Apple to put all the block names in
    >> Irish, of course.... ;-)
    >
    >Using the official Unicode script name in English is not a problem.
    >But a OS vendor could as well choose to translate these names in
    >localized versions of this font if the OS itself is translated.

    At some expense. You are welcome to lobby Apple to commission a
    localized French version of the Last Resort Font.

    >The other good thing could be to include also the script name in a
    >typical language using that script natively (for example the
    >"lastresorthebrew" glyph usezd for the hebrew block should not only
    >display "HEBREW" with ASCII letters on the top, but also the Hebrew
    >term for "HEBREW")...

    You are welcome to draw up a list of all of the "native" names.

    >Such a font seems easy to create automatically by using the basic
    >glyphs of a base font containing the ASCII letters and digits, and
    >a source text file giving the name and range of Unicode code point
    >blocks, as well as a representative character or string.

    You don't know much about drawing fonts, evidently.

    >Hinting the generated LastResort font it is not really necessary,
    >but some automatic hinting could be used in the generated glyphs
    >to hide (at PPM values below 64), the border text that displays
    >the block name and range of hex code points.

    To what end?

    >I just wonder if it would then not be simpler to have a way to
    >define algorithmic fonts implemented as native compiled code in
    >a DLL or shared library for the OS using it, instead of needing
    >it to be installed as a regular TTF font.

    Wonder away.

    >I don't know for example if Windows or MacOSX can reference a
    >DLL using some known and documented COM interface as if it was
    >a font: It would allow, possibly, an easier development for
    >advanced hinting, or kerning or ligature, or glyph processing
    >by native code rather than with tables in TTF/OTF/AAT fonts.

    "Possibly"?

    >Of course the same question comes to Java: can a Java class
    >be used to implement a font for use in Swing or Java2D or AWT?
    >It seems that it is possible, as Java handles fonts selection
    >by using a Font object instanced by the resource loader but
    >that may be subclassed, and instanciated as another font.

    "Possible."

    >As Java is a de-facto common standard for cross-platform
    >compatibility of binary code, this could be a very interesting
    >alternative to TTF encoded fonts which are complex to develop
    >and test. At least, this could be used to develop complex
    >fonts before they are finalized in a OTF/AAT format.

    "Interesting."

    I don't think so.

    -- 
    Michael Everson * * Everson Typography *  * http://www.evertype.com
    


    This archive was generated by hypermail 2.1.5 : Wed Dec 03 2003 - 06:44:41 EST