L2/05-252 Source: Ken Whistler Date: 2005-09-21 Subject: WG2 Meeting Summary (Sophia Antipolis) I haven't seen any report go around yet, and Asmus has bad connectivity this week, so I thought I would give a brief summary here for everyone regarding the conclusions from last week's WG2 meeting in Sophia Antipolis, France. If anybody has any questions, I can provide further information. --Ken 1. General As usual, the meeting had sizable delegations from Japan, China, and ROK. Also attending were TCA (for Taiwan), Finland, Ireland, Germany, and Canada. It was a little surprising that there were not more European national bodies represented, including nobody from France. 2. AMD 2 The disposition of comments for FPDAM 2 for 10646 was completed, and the amendment now progresses to an FDAM ballot. This clears the way for Unicode 5.0, as the repertoire is now fixed. The discussion of AMD 2 was rather contentious during the meeting, because of remaining disagreements about a few characters for N'Ko (Canada abstained on progression of the amendment over this issue) and because of a late request by Finland for 6 characters needed to complete the Uralic Phonetic Alphabet set needed for an ongoing publication project. The disagreement over the precedent being set by Finland's request for 6 character additions at such a late stage in the ballotting (and without supporting ballot comments on the FPDAM itself) was handled by writing a much more explicit set of guidelines into the WG2 Principles & Procedures document regarding what would or would not be appropriate to consider for late additions. And the Finnish request *was* then accomodated by adding the 6 characters in question to the amendment. It is notable that there was no disagreement about the characters themselves, which all seemed well-justified -- only about the procedural issues in adding them to an amendment so late. The positions endorsed by the UTC regarding AMD 2 were accomodated. The details can be found in the WG2 resolutions document, WG2 N2954. The summary, however, is that there were no character removals from the amendment. There were 35 characters *added* beyond what were in the FPDAM document. Of the additions, 9 were related to case-folding stability and had already been endorsed by the UTC. The dotted square was also added based on U.S. ballot comments. Many of the others added have also been considered and approved by the UTC already, but had not officially been requested on U.S. ballot comments from L2. For those few characters not already considered by the UTC, including the 6 additions for Uralic, I will be preparing the usual WG2 Consent Docket document for the next UTC, so that the UTC can bring its approvals for Unicode 5.0 back into synch with the forthcoming text of the FDAM. 3. AMD 3 In addition to the work on Amendment 2, WG2 agreed to issue a ballot for a subdivision of work and a PDAM ballot for Amendment 3. The scripts approved for the PDAM (so far) are: Lepcha, Vai, Saurashtra, and Ol Chiki. In addition, miscellaneous small collections of characters for Malayalam, Devanagari (Sindhi), and Greek epigraphy were included, plus a couple of symbols. Most have already been considered and approved by the UTC. Details on differences in approval status will be included in the WG2 Consent Docket. Unlike the work on Amendment 2, there was very little controversy about any of the content for Amendment 3, in large part because there were no procedural issues with issuing a PDAM for this content, and the 4 script proposals were quite mature and had evidence of community support. 4. Hangul One additional issue came up which was rather controversial at the meeting. It turns out there is a mismatch in language between the Unicode Standard (Section 3.12) and 10646 (Clause 26.1) regarding the definition of what Unicode calls a "standard Korean syllable block" and 10646 calls a "complete syllable block". The problem springs in large part from a lack of clear definitions in 10646, but the controversy arose because of differences of opinion regarding how old Hangul syllables should be represented in normalized text. To move the process along, Asmus (for the UTC) and I (for the U.S. NB) collaborated during the meeting to draft and submit a defect report regarding the current text in 10646. Michel has drafted some clarificatory text that could be added to the next amendment for 10646, but there will have to be a further cycle of discussion about this at the next WG2 meeting in April 2006 before any resolution of the issue. 5. Future Meetings Meeting 48 had already been set for April 24-28 in Mountain View in the U.S., hosted by the Unicode Consortium at the Microsoft site. Meeting 49 has now been set for September 25-28 in Tokyo, hosted by the Japan NB. Meeting 50 is tentatively scheduled for Spring 2007 in Europe, but no host has volunteered yet.