Hi all
as a rule cybercafes tend to be responsive to users. If enough users ask for
Unicode 5.1 support, and there are enough Unicode 5.1 sites, then cybercafes
will support it.
And the reality is that proper Burmese support in applications will be built
on Unicode 5.1, wether its OpenOffice, Firefox, Thunderbird, Microsoft
applications, etc
Burmese locale support is Unicode 5.1 based. Line breaking and collation
routines that ake their way into major applications will be Uniocde 5/1
based, etc.
The question is how you facilitate the uptake rather than encouraging people
not to migrate.
2008/7/26 Michael Everson <everson(a)evertype.com>
>
> >1. As mentioned in Ko Ngwe Tun's previous
> >discussion, we use internet cafe computers and
> >have less option for installing new fonts or
> >browser based addon.
>
>
> IT experts in Myanmar should certainly be
> encouraging internet cafés to migrate to Unicode
> 5.1. And if there are Facebook or Flicker sites
> using non-Unicode encodings, those should also be
> encouraged to migrate, as Htoo and I have been
> trying to do with the Wikipedia.
>
I'd agree, but there is one key issue nobody seems to be addressing. Most
users are using a pre-Vista version of Windows, and currently I am unaware
of any legal software solutions for rendering Myanmar Unicode 5.1 on
pre-Vista Windows, other than a handful of specialist SIL applications
running Graphite, and some very old experiemnts in Firefox and OpenOffice
For uptake, it needs to be easy and free or very very cheap.
> >2. Myanmar Unicode has history of breaking
> >previous version and luck of migration support.
> >It is rather strange that there is NO convertor
> >available for myanmar1 (Unicode 4.0) to myanmar2
> >(Unicode 4.1) or myanmar2 to myanmar3 (Unicode
> >5.1).
>
> It isn't Unicode's place to make such a
> converter. The owners of myanmar1, myanmar2, and
> myanmar3 should have a care for the problem. But
> there *are* converters available for text
> conversion.
>
agree, and from memory a nnumber of convertors exist
> >3. Unicode font are not redistributation (yes
>
Some fonts are re distributable, but Uniscribe isn't.
>5. Unicode font input methods are low quality,
> >extra cost or branding. Zawgyi uers has long
> >being enjoy phonetic input.
>
This is a furphy.
There is absolutely no reason why Unicode 5.1 can't be phonetic. Although I
question why it should be phoentic or more properly psuedo-phonetic.
Likewise a keyboard layout could be either based on a logical or visual
keyboard layout.
This is just an implementation issue, the reality is that keyboard layout
developers choose to implement keyboards certain ways, using certain
technologies.
>
> Zawgyi users are forced to encode their language
> texts in a non-standard way. This will only hurt
> them, and Myanmar, in the long run, as it does in
> the short run.
>
From the point of view of keyboard layout design principles the Zawgyi
keyboard layout is sub-optimal with many design problems. Even if you decide
to implement a phonetic keyboard for Unicode 5.1 you would not base it on
the Zawgyi layout. The Zawgyi layout uses teh same approach as legacy 8-bit
fonts essentally. The Zawgyi layout indicates that the font doesn't support
complex rendering, and the layout wasn't implemented for a "smart" input
system.
> >Myanmar Wikipedia should not take the burdon. At
> >bottom line, Wkipedia is providing knowledge,
> >not font. Also people don't think Wikipedia is
> >good enough to install a new font.
>
In part I'd agree, Myanmar Wikipedia should not be a burden. Although I
don't think that avoiding font downlaods solves teh problem.
At some point people will start migrating, Optimising the CSS will allow
people to use any Uniocde 5.1 font they have is a necessary approach.
Hopefully over tiem they determien which font suits tehir needs best. I'd
rather ot tie it to a single font, whether its one particular uniocde 5.1
font or Zawgyi.
--
Andrew Cunningham
Vicnet Research and Development Coordinator
State Library of Victoria
Australia
andrewc(a)vicnet.net.au
lang.support(a)gmail.com