From: Michael Everson (everson@evertype.com)
Date: Wed Mar 22 2006 - 03:33:11 CST
At 18:23 -0800 2006-03-21, Mark Davis wrote:
>Michael Everson has pointed out that a proposal to change the
>representation of certain Myanmar characters has been submitted to
>WG2. That proposal did not reach the UTC in time for its last
>meeting.
The proposal wasn't written then. My mission to Yangon took place the
week after the UTC. I was in Chiang Mai during the UTC working on
Lanna.
>Any prospective change to the Myanmar encoding that would affect the
>stability of the Unicode Standard must be in accord with the Unicode
>Stability Policies, and even if in accord with those policies, must
>be very carefully weighed in terms of its potential costs and
>benefits.
The proposal requests some disunifications and some character additions.
>Doing so requires sufficient time to make a very careful examination
>of the issues -- not precipitous action.
Ireland, the UK, and the Myanmar organizations who wrote the proposal
do not propose precipitous action, but propose that the 7 characters
be added to the current ballot as a matter of implementation urgency.
The current encoding causes problems for representation (processing
and rendering) which have been pointed out by the Myanmars and by SIL
for some time now. Unicode-compliant implementations in Myanmar are
all "experimental", with different groups trying different hacks to
make it work. The new proposal addresses the problem by removing the
need for hacks by simplifying and regularizing text representation
for Burmese and for five official minority languages. The urgency
request is to have 7 characters needed for Burmese to be added to
PDAM3; 65 other characters for minority languages are proposed for
PDAM4.
The fact that no one in Myanmar has released a Unicode implementation
is fortunate really... there really isn't all that much *Unicode*
legacy data out there (almost all Burmese text is still represented
in a plethora of 8-bit solutions). The Myanmars want to fast-track
the 7 characters because they could fix their experimental
implementations quickly and they could release Unicode
implementations soon. If we do not meet this market need with the
fast-track, then there is a danger that there *could* be a lot more
legacy data that would need transcoding. Now is an excellent time to
fix the long-existing problems with Myanmar encoding.
-- Michael Everson * http://www.evertype.com
This archive was generated by hypermail 2.1.5 : Wed Mar 22 2006 - 03:42:09 CST