From: Ngwe Tun (ngwestar@gmail.com)
Date: Wed Jul 05 2006 - 01:00:02 CDT
Thanks N.Ganesan for your reminding to Paul,
Burmese,It is only and one unique circle shaped handwriting.
they didn't put in uniscribe level.
let me know status of burmese shaping engine (Paul said that "Of all of the
script in the world that I have done shaping engine support there are none
that I would consider to be "extreme"." Burmese is extreme case? isn't it?
Have you change new encoding model for burmese?
And we implemented some fonts sample in
http://www.myanmarnlp.net.mm/opentype. Have you tried in your shaping
modules?
Rgds
Ngwe Tun
On 7/4/06, N. Ganesan <naa.ganesan@gmail.com> wrote:
>
> Thanks Paul for the reply, and esp. the picture attachment.
> Very beautiful.
>
> Hopw Adobe, Apple and Microsoft give Indic language support, and
> eventually beautiful handwriting style fonts for India,
> (and Burmese, Thai, Khmer or Balinese).
>
> N. Ganesan
>
>
> On 7/3/06, Paul Nelson (ATC) <paulnel@winse.microsoft.com> wrote:
> > The encoding looks just like regular text. Typographic forms are never
> > encoded. I attached an image of two fonts that have contextual shaping
> > through smart font technology. The encoded text is "This is a simple
> > sample. subsystem sub sp".
> >
> > The concept of splitting display from encoding is very important. An
> > example from India is the Traditional Malayalam font style and the
> > Reformed Malayalam style. Both writing styles have the same encoding for
> > the text, but a font will shape the conjuncts according to the rules for
> > the writing style. This means lexical tools, text to speech, or speech
> > to text can function in a deterministic manner based on a uniform
> > backing store, but a font maker can control the style of the font.
> >
> > The same might also be done for Devanagari script where one region
> > prefers more conjunct and one region prefers overt halant forms. The
> > backing store is the same, but the font maker provides control to the
> > user to decide what forms to use.
> >
> > Regards,
> >
> > Paul
> >
> > -----Original Message-----
> > From: N. Ganesan [mailto:naa.ganesan@gmail.com]
> > Sent: Tuesday, July 04, 2006 1:16 AM
> > To: Paul Nelson (ATC)
> > Cc: Sinnathurai Srivas; Sukhjinder Sidhu; Indic Discussion List
> > Subject: Re: [indic] Re: provisional named sequences
> >
> > On 7/1/06, Paul Nelson (ATC) <paulnel@winse.microsoft.com> wrote:
> > > Of all of the script in the world that I have done shaping engine
> > > support there are none that I would consider to be "extreme". There
> > > are a number of ways that scripts work, but they are normally pretty
> > logical if one learns the rules.
> > >
> > > People consider Latin script as "simple". However, some of the most
> > >complex fonts I have seen lately are Latin script fonts that memic
> > calligraphic handwriting.
> > >I guess it is all a matter of perspective.
> > >
> >
> > Are these calligraphic handwriting Latin letters getting encoded?
> > Like Fractur letters (of the German script) already in unicode.
> >
> > N. Ganesan
> >
> >
> >
> >
>
>
This archive was generated by hypermail 2.1.5 : Wed Jul 05 2006 - 01:18:25 CDT