Re: Representative glyphs for combining kannada signs

From: Philippe Verdy (verdy_p@wanadoo.fr)
Date: Wed Mar 22 2006 - 04:00:29 CST

  • Next message: Antoine Leca: "Re: Representative glyphs for combining kannada signs"

    From: "Mike Ayers" <mayers@celequest.com>
    > Philippe Verdy wrote:
    >> Who is that third party (regarding support of MS products by MS itself)
    > > that would write a later version of Uniscribe?
    > While it is not common to use the term "third party" to refer to an
    > internal supplier, it is sometimes appropriate, as in this case where we
    > are discussing the use of one (internal) product (Uniscribe) by two
    > different (internal) products (Notepad and Charmap). Here, "product" is
    > as in "product of a build process", not "product for sale".

    That's internal cooking at Microsoft. Users get a single licence covering Uniscribe, Notepad and Charmap as part of a single product, Windows. The whole has to be supported as one product, and the interoperablity of these tools or libraries that are part of the same product needs to be effective. Anyway, I eventually see another internal party involved here, that may not be Microsoft but one of its suppliers: the foundry that makes the AUMS font for Microsoft (mostly glyph designs).

    But it's up to the "Microsoft Typography" team to verify that it works well in its own Office product for which it is built, and Windows as the target platform that Microsoft also supports. And I think that Microsoft is directly involved itself in the design and verification of the supplementary OpenType shaping feature tables embedded in its licenced font. Personnally, I think that the Microsoft Typography team decision to include partial support (glyphs only) in AUMS for some Indic scripts without the needed OT tables was really bad.

    MS should have better included only the glyphs for the scripts it supports, possibly adding the unsupported scripts in a distinct and non-supported font, but that users could easily replace by a third party font without dropping the additional and useful additions of AUMS for Latin/IPA/Greek/Coptic/Cyrillic, Armenian, Hebrew, Arabic, Georgian, Hangul and symbols (anyway, I still wonder if Arial Unicode MS is the appropriate font to put symbols, there are better places for that in "Wingdings").

    And I do think that the supported Indic scripts (Devanagari, Bengali, Gurmukhi, Gudjrati, Tamil, Telugu, Kannara, Malayalam, and even Thai) should have been included in a separate font, possibly through internal font linking (like in Tahoma), if the sans-serif designs are compatible (for example, I don't see the interest of including a large set of Han ideographs in AUMS, when there's certainly a better way with the GB18030 distinct font).

    The current inclusion of incomplete Oriya, Lao and Tibetan in AUMS without the needed features was highly questionable at the origin (Microsoft Typography did know that it would not work properly with any coming Uniscribe engine that would sometime have to support those scripts), and separate these pre-alpha script support should have been taken out of a production release of AUMS (using instead an internal font linking, with some custom OpenType table that specifies alternate fonts to lookup by Uniscribe, where those internally linked fonts would be considered only as a last option after user options).

    Note that OpenType font format supports multiple designs in the same font file, but each with their own font name, separate glyph tables, and separate OpenType feature tables (so internal font linking would have referenced a font in the same OpenType font file).



    This archive was generated by hypermail 2.1.5 : Wed Mar 22 2006 - 04:20:26 CST