L2/08-204


Title: Quick Report on WG2's Meeting Last Week
Source: Ken Whistler
Date: April 30, 2008


WG2 met last week, April 21 - 25, in Redmond, Washington,
at a meeting hosted by Microsoft.

For those of you on the list who weren't present and who
haven't been actively spelunking through the WG2 document
register looking for clues, I thought I would give
a quick overview of what took place and the outcome in
terms of new amendment progression for 10646.

The meeting had national body participation by Canada,
China, Ireland, Japan, the Republic of Korea, Poland,
U.K., and the USA. It had liaison participation by
SEI - UC Berkeley, the Taipei Computer Association, and
the Unicode Consortium.

China, in particular, brought quite a few experts, because
of their ongoing concerns about pushing forward encodings
for minority and historic scripts of China: Old Turkic,
Nushu, Tangut, and Lisu all got into ballot, and there
was also an expert on Classical Yi who came to work on
that script with other experts.

The high-level overview is that WG2 progressed Amd 5 and
Amd 6, and didn't start any other new amendment.

For a quick visual overview of where things stand for
character repertoire additions, you can
refer to the Unicode Pipeline page, which has now been
updated with the latest results from WG2:

http://www.unicode.org/alloc/Pipeline.html

Principles and Procedures

WG2 agreed to incorporate material into its Principles
and Procedures document based on two UTC/L2 contributions.
The first had to do with writing in more details to
ensure that all stakeholders for newly encoded scripts
get consulted early on in the process. The second was
the set of recommendations about dandas for various
already encoded and yet-to-be-encoded Brahmi-derived
scripts.

Amendment 5

Amd 5 had is disposition of comments for the FPDAM ballot,
and will now be moving on to its (non-technical) FDAM
ballot. Basically, this one is now done.

There were a number of difficult issues to resolve for
Amd 5, but it amounted to wrangling around the edges.
One involved a handful (3) of additional CJK compatibility
characters for mapping to the ARIB standard. Those stayed
in. Then there were some complications about how to
reference Unicode and references to the Ideographic
Variation Database, and some details about editing the
text and glyphs for Korean. A few characters were removed
from Amd 5 for future study.

A number of glyph fixes were approved, and those will be
applied immediately, by being published in Amd 5, rather
than waiting for Amd 6.

Amd 5 now goes to the FDAM ballot with the following
repertoire additions:

  1. More Old Hangul jamo letters, to complete that set
  2. Tai Tham script
  3. Tai Viet script
  4. Avestan script
  5. Egyptian hieroglyphics (Gardiner set and some extensions)
  6. CJK Extension C (4149 more CJK unified ideographs)
  7. A handful of miscellaneous character additions
 
Amendment 6

Amd 6 had its disposition of comments on its PDAM ballot.
If you recall, PDAM 6 had a rather limited set of new
repertoire content, consisting mostly of 3 historic
scripts (Imperial Aramaic, Inscriptional Parthian,
and Inscriptional Pahlavi), and then a scattering of
additional miscellaneous characters. Because there were
a large number of additional proposals ready for ballot,
WG2 decided to consolidate them into Amd 6 and simply
reballot Amd 6 as a PDAM, rather than going through the
extra work of balloting Amd 6 as an FPDAM and putting
all the new stuff into a new amendment to ballot as
a PDAM -- which would just have created additional
work for everybody.

Because the danda recommendations had been picked up
by WG2, it was then possible to move Meitei Mayek
(now to be spelled Meetei Mayek) out of its on hold
status and back to ballot. Dandas were added, and
the script was reordered based on feedback from
Manipur. It is now in Amd 6.

WG2 then added a number of scripts to Amd 6 that had
already been reviewed and approved by the UTC: Javanese,
Samaritan, Lisu, Kaithi, Old South Arabian, and Tangut.

There was a fair amount of controversy about the addition
of Tangut, which mostly revolved around the issue of
whether to hold off until Tangut radicals could all
be identified and encoded, along with the repertoire
of ideographs, and how that might impact the decision
about the order of the large main repertoire (5910
ideographs). In the end, the position of the U.S. and
China to proceed with balloting the set of 5910 ideographs,
and to let the discussion of the potential encoding of
radicals continue and be resolved later, prevailed, and
so Tangut was added to Amd 6.

WG2 also added the Vedic extensions that the UTC has
been reviewing for so long to Amd 6. The one exception
was the prishthamatra e, which was withheld from ballot
for now, pending further input from the Government of India
and further technical discussions about it.

WG2 also added miscellaneous characters that had been
approved by the UTC for Unicode, including the
Rumi numeral symbols, the livre tournois sign, the
decimal exponent symbol, and a handful of others.

WG2 agreed to add the 289 Tamil named sequences to Amd 6.

There were also a number of additions to Amd 6 of sets
of characters that the UTC has not yet approved. These
included 18 characters added to the Myanmar script for
minority language support, and 39 characters added to
the Unified Canadian Aboriginal Syllabics.

WG2 also decided to proceed to add 389 characters for
Nushu (a local women's writing system in China). That
addition was somewhat controversial, as there are still
some questions about the makeup of the repertoire and
mapping issues between sources. So it will need to be
carefully reviewed during the ballot period.

WG2 added the Old Turkic runes. The UTC has reviewed
various iterations of that proposal in the past.

Finally, WG2 agreed to add 186 Japanese TV symbols,
for interoperability with the ARIB standard. This is
also a proposal that the UTC has worked on for several
cycles -- but it was further modified during the WG2
meeting, and will also need careful review during the
ballot for Amd 6.

Amd 6 now goes to a second PDAM ballot with the following
repertoire additions:

  1. Various additions for Vedic texts
  2. Myanmar extensions for minority languages
  3. UCAS extensions for several languages
  4. Meetei Mayek script
  5. Lisu script
  6. Javanese script
  7. Imperial Aramaic script
  8. Old South Arabian script
  9. Inscriptional Parthian script
 10. Inscriptional Pahlavi script
 11. Old Turkic script (runes)
 12. Kaithi script
 13. Tangut script
 14. Nushu script
 15. 186 symbols for interoperability with the Japanese
     ARIB TV standard
 16. Miscellaneous other character additions

Of potential interest: the WG2 resolutions were all
passed unanimously, without even any abstentions. I
consider this to indicate that despite the occasional
sharp disagreements over some details, WG2 is still
proceeding in its work on 10646 with a great deal of
consensus among the national bodies participating --
regarding both the standard itself and what kinds of changes
should be made to it.

--Ken