L2/05-270


Title:  WG2 Consent Docket (Sophia Antipolis)
Date:   September 21, 2005
Author: Ken Whistler


To assist the UTC in finalizing all the approvals for
the Unicode 5.0 repertoire (and future versions), I have, 
as usual, collected together in this document all of the 
currently unresolved discrepancies between approvals of 
record by the UTC for character additions and the latest 
status of WG2 decisions regarding character additions for 
ISO/IEC 10646.

The relevant documents from the most recent WG2 meeting,
in Sophia Antipolis, France are:

WG2 N 2954 Resolutions of WG 2 meeting 47 (= L2/05-271)

WG2 N 2991 Summary of repertoire for FDAM 2 of ISO/IEC 10646 (= L2/05-272)

WG2 N 2993 Summary of repertoire of PDAM 3 of ISO/IEC 10646 (= L2/05-273)

==============================================================

Part I: Discrepancies relevant to Amd 2 and Unicode 5.0

This part is concerned with discrepancies related to Amd 2,
which will be issued soon for an FDAM ballot and which should
define the repertoire for Unicode 5.0. At this point, no
further technical change to Amd 2 is feasible, and the UTC
should simply approve all of these changes and/or additions,
to avoid any chance of a desynchronization of the standards.


A. Uralicist character additions

At the WG2 meeting, Finland brought in a request for the
addition of 6 Uralicist characters for Amd 2. The UTC saw
these at the August meeting (L2/05-189, = WG2 N 2958), but
took no action, as everyone misinterpreted the "progress
report" as simply an FYI. At WG2 it became apparent that
the report was actually an urgent request for the addition
of 6 characters. That request was restated, much more
clearly, in WG2 N 2989 (= L2/05-261). That document
represented the consensus of an ad hoc group at the meeting,
which concluded that all 6 characters were justified.

The 6 characters were accepted into FDAM 2. The UTC needs 
to formally approve them as well.

1DFE COMBINING LEFT ARROWHEAD ABOVE
1DFF COMBINING RIGHT ARROWHEAD AND DOWN ARROWHEAD BELOW
27CA VERTICAL BAR WITH HORIZONTAL STROKE
2C77 LATIN SMALL LETTER TAILLESS PHI
A720 MODIFIER LETTER STRESS AND HIGH TONE
A721 MODIFIER LETTER STRESS AND LOW TONE


B. Horizontal mathematical bracket

The Irish NB ballot comments asked for a bottom version of the
horizontal tortoise shell bracket, to match the already
approved U+23E0 TOP TORTOISE SHELL BRACKET. The UTC had
considered this as a symmetry issue, but hasn't approved
the addition. The original proposal (L2/04-329) hadn't asked
for it and noted: "MathType also supports the top tortoise
shell bracket, but no bottom bracket."

Asmus has indicated that in consultation with the math experts,
it is clear that these brackets would be used consistently,
with top forms for annotations above numerators in stacked expressions
and bottom forms below denominators -- so there really is no reason
not to have them in symmetric pairs.

In any case, WG2 approved the bottom tortoise shell bracket, and
the UTC needs to formally approve it, as well as a code point
change for another character that resulted.

23E1 BOTTOM TORTOISE SHELL BRACKET

And to keep the horizontal brackets in a contiguous range, the
following code point change:

23E1 --> 23E7 ELECTRICAL INTERSECTION


C. Code points for already approved Latin characters

There was a screwup in code point allocation at the May UTC
meeting for the addition of two Latin characters. At the
February UTC meeting, 6 Latin characters for Uighur were
accepted in the range U+2C65..U+2C6A. (102-C10, L2/05-029)
At the May UTC meeting, 2 lowercase Latin characters with
stroke (for Sencoten) were accepted in the range
U+2C65..U+2C66. On May 25, I notified Lisa and Rick of the
code point conflict, while updating the pipeline, and
suggested a correction to the open code points U+2C6B..U+2C6C.
The pipeline has reflected that state of affairs since that
date.

However, corrected minutes for the May UTC meeting have
been in limbo, and when Asmus prepared preliminary (FDAM 2) ballot
documents for the WG2 meeting, he resolved the code point
conflict differently, by leaving the two characters from
the May meeting alone, and instead moving the 6 Uighur
characters down by two code points.

In the interest of minimizing the possibility for further
errors in document preparation, WG2 chose to accept all
8 characters at the code points shown in the preliminary
ballot documents. The UTC needs now to formally accept these
modified code points, so that everything is back in synch.
For absolute clarity, the entire list of 8 is given here.
What is in question is not the names or identities of the
8 characters, but merely an affirmation of the code points
assigned by WG2, which will be in the FDAM 2.

2C65 LATIN SMALL LETTER A WITH STROKE
2C66 LATIN SMALL LETTER T WITH DIAGONAL STROKE
2C67 LATIN CAPITAL LETTER H WITH DESCENDER
2C68 LATIN SMALL LETTER H WITH DESCENDER
2C69 LATIN CAPITAL LETTER K WITH DESCENDER
2C6A LATIN SMALL LETTER K WITH DESCENDER
2C6B LATIN CAPITAL LETTER Z WITH DESCENDER
2C6C LATIN SMALL LETTER Z WITH DESCENDER


D. Cuneiform

The U.S., Ireland, and Canada all asked for a coordinated set
of minor updates (name and glyph changes, some removals) to a small number
of Sumero-Akkadian cuneiform characters in the ballot, based
on input from the expert, Steve Tinney. Michael Everson fielded
a few more last-minute fixes for glyphs and names from Tinney,
and those were also incorporated in the revised chart and
names list seen in WG2 N 2991 prepared for the FDAM 2 ballot.

The UTC should give another "just in case" approval for the
revised chart and character name as shown in WG2 N 2991, to
pick up these last revisions.


E. N'Ko Name Changes

WG2 agreed to 4 N'Ko character name changes requested by
Ireland. The UTC needs to accept these name changes. The
list is:

07E8 NKO LETTER JONA JA   (< ... OLD JA)
07E9 NKO LETTER JONA CHA  (< ... OLD CHA)
07EA NKO LETTER JONA RA   (< ... OLD RA)
07F6 NKO SYMBOL OO DENNEN (< ... OO DEENE)


F. Phags-pa Glyphs

Andrew West requested a number of minor glyph changes for
Phags-pa in documents WG2 N 2972 (= L2/05-255) and WG2 N 2979
(= L2/05-257). The UTC took no position on this.

In consultation with China and Ireland at the WG2 meeting,
an acceptable set of glyph changes were agreed upon. The
characters impacted are:

A843..A845, A852, A856..A857, A859, A863..A864,
A867..A868, A870..A871

All of these changes are rather minor, and were acceptable
to China. With the exception of the change A857, which
was a matter of choice between two alternative glyphs,
the rest of the changes amount to replacing a short continuation
bar at the lower right of a glyph with a small serif (or
nothing at all).

The UTC should simply agree to the glyphs as now
shown in WG2 N 2991.

Note: WG2 agreed to fixes for all of the glyph errata that the
UTC had posted, so with agreement on these Phags-pa glyph
changes, the UTC and WG2 should once again be in synch
in agreeing about the glyphs to use in the code charts.

===========================================================

Part II: Discrepancies relevant to Amd 3 and Unicode 5.x

Unlike the discrepancies related to the FDAM for Amd 2, the
other discrepancies listed in this part deal with WG2
approvals for a PDAM ballot for Amd 3. In these cases the UTC has
greater flexibility, as there are two rounds of national
body ballotting to go, and there is leeway to disagree and
change things. The choices now are either to

  1. simply approve what WG2 has done and move on
  2. disapprove (in part or in whole) now and take the
     relevant actions to start preparing positions for
     ballot comments, or
  3. take no action now, and wait until actually confronted
     with the discrepancy later when reviewing the PDAM
     
In most cases I don't advise option 3, unless there is significant
potential for ratholing in discussion of some topic now, as it raises
the risk that we will overlook a discrepancy in the future in
responding to a ballot. (And it is more work for me, Michel,
and Asmus to track things.)


G. Inverted Interrobang

WG2 approved:

2E18 INVERTED INTERROBANG

on the basis of WG2 N 2935 (= L2/05-086). The UTC has discussed
the character in extenso, but not yet approved it.


H. Musical symbol

WG2 approved:

1D129 MUSICAL SYMBOL MULTIPLE MEASURE REST

on the basis of WG2 N 2983 (= L2/05-258). An earlier version of
that document was discussed on the unicore list, and it is my
opinion that the resolution in L2/05-258 is probably the best
way to go on the issues discussed there.


I. Lepcha /ng/

WG2 approved the entire proposal for Lepcha, based on WG2 N 2947
(= L2/05-158). That includes one character not yet approved by 
the UTC:

1C35 LEPCHA CONSONANT SIGN KANG

The UTC needs to review the evidence and decide what to do.


J. Old Chiki

WG2 approved the encoding of the Ol Chiki script at a new block
1C50..1C7F, based on document WG2 N 2984 (= L2/05-243). The
encoding is not problematical -- in fact it is elaborately and
profusely praised by at least some of the Santali-speaking
community who want to use it. (But recall, there was some
fuss awhile ago by a counter-community that disapproves of the
use of the old script at all.)

I would recommend that the UTC simply approve the encoding at
this point.


K. Saurashtra

WG2 approved the encoding of the Saurashtra script at a new block
A880..A8DF, based on document WG2 N 2969 (= L2/05-222). There
has been a lingering disagreement with Peri Bhaskararao
regarding one character in the proposal (U+A8B4 SAURASHTRA
LETTER UPAKSHARA), but at this point my
recommendation would be to approve the entire proposed encoding
to synch up with WG2 and to let the discussion about that one
character continue and resurface if it needs to.