From: Christopher Fynn (cfynn@gmx.net)
Date: Wed Jan 05 2005 - 22:24:53 CST
Hi Phillipe
Gregor Verhufen's old Tibetan fonts did *not* use an "underlying
Wylie" representation for Tibetan data or glyphs. They use a
"font-hack" encoding where Tbetan glyps are mapped to Windows ANSI
code-page characters - but there is no real relationship between the
characters they are mapped to and Wylie translitteration. The fonts did
come with Keyman macros which converted Wylie translitteration typed on
the keyboard to the characters the glyphs were mapped to in the font.
The typed in Wylie is converted to other characters by Keyman - it is
these characters which are stored not Wylie. I'm sure if you check with
Gregor he will confirm this. I can even provide you with maps of the
characters used in his font which I made for someone writing a converter
from this encoding to Unicode. (Actually there are several encodings as
the glyph mapping was not consistent in all his fonts).
The Nitartha system is similar - it uses a Wylie like input method and
a bunch of VBA macros wrapped up in a WLL library to convert the input
to the precomposed Tibetan glyphs in 8-bit fonts. However in the case of
the Nitartha system the glyphs for one type face are spread over a
number of different fonts so there is a lot of font switching involved.
Hope this clarifies things.
regards
- Chris Fynn
Philipp Reichmuth wrote:
> Kenneth Whistler schrieb:
>
>>>> I wouldn't rule this out entirely. For example, I know one attempt
>>>> to implement a Tibetan font where the underlying representation was
>>>> Latin (Wylie), and the Tibetan glyphs were generated from the Latin
>>>> transliteration using OpenType rules
>>
>>
>> I presume Philipp Reichmuth was talking about:
>> http://www.nitartha.org/wylieandconverter.html
>
>
> Actually I wasn't; I was referring to an in-house project at our
> institute for Central Asian Studies, a derivative of the fonts from
> http://home.t-online.de/home/0228359452-0001/jamyang.htm. Newer
> versions use "proper" Unicode Tibetan codepoints instead, for obvious
> interoperability reasons.
> The underlying plain text is perfectly Unicode conformant; after all,
> it's only Latin, and the standard isn't concerned about glyphs and their
> meaning. I guess this is "nonconformant" only in that it somehow defeats
> the purpose of Unicode.
>
> Philipp
>
>
>
This archive was generated by hypermail 2.1.5 : Wed Jan 05 2005 - 22:27:10 CST