ISO

International Organization for Standardization

Organisation Internationale de Normalisation

ISO/IEC JTC 1/SC 2/WG 2

Universal Multiple-Octet Coded Character Set (UCS)

ISO/IEC JTC 1/SC 2/WG 2 N1303

Date: 1996-01-26

Title: Unconfirmed Minutes of WG 2 Meeting # 29 in Tokyo, Japan;
1995-11-06--10

Source: V.S. UMAmaheswaran, Meeting Secretary, Mike Ksar, Convener

Action: WG 2 members and Liaison organizations

Distribution: ISO/IEC JTC 1/SC 2/WG 2 members and Liaisons

1. Opening and roll call

1.1 Opening

Input Document: N 1245R Call for WG2 Meeting 29 in Tokyo; Ksar, Convener; 1995-08-28

The 29th meeting of WG2 was held in Room 6-66, at Kikai Shinko Kaikan Building, 3-5-8 Shiba-Kouen 3, Minato Ku, Tokyo 105, Japan, from 1995-11-06 (Monday) to 1995-11-10 (Friday). The meeting was called to order by Mr. Mike Ksar at 09:30h on Monday. He introduced Mr. Takayuki Sato who was representing the host, Japanese Industrial Standards Committee (JISC). Information Processing Society of Japan / Information Technology Standards Commission of Japan (IPSJ/ITSCJ) provided the secretarial support for the meeting. Mr. Takayuki Sato welcomed the delegates on behalf of JISC to Tokyo, provided some information about the meeting location and surroundings, and explained the logistics of the meeting. Delegates were requested to return requested information for the Tuesday evening reception at Park Hall restaurant starting 18:00h.

The objective of this meeting was to continue the WG2 program of work and adopt resolutions on repertoire content and allocation of code space. The target for completion of the agenda was set for Thursday PM, with Friday AM devoted to adopting the resolutions.

1.2 Roll call

Input Document: 1251 Updated WG2 Distribution List, post WG2 meeting 28 in Helsinki; Mike Ksar, Convener; 1995-06- 27

Twenty two (22) delegates representing eleven (11) member bodies, one (1) liaison organization, including two observers (interpreters) from the Mongolian embassy in Tokyo, were present. The latest WG 2 mailing list (document N 1251) was circulated and attendees were requested to verify and update its contents. Mr. Mike Ksar welcomed the presence of the Mongolian delegate, and now that Mongolia is a P member of SC 2 we expect their continued participation in the future. The attendance list, sorted by the country name of the national bodies, is given in the following table.

Messrs. Mike Ksar, Takayuki Sato, UMAmaheswaran and Arnold Winkler were appointed to the drafting committee.

NAMENATIONAL BODYAFFILIATION
Alain LaBontéCanadaGouvernement du Québec
V.S. UMAmaheswaranCanada; Meeting SecretaryIBM Canada
Mao Yong GangChinaChinese Electronics Standardization Institute
Zhang ZhoucaiChina, IRG RapporteurCCID
Sven ThygesenDenmarkKommunedata
Grigorios StathsGreeceUniversity of Athens
Panagiotis DelliosGreeceConsultant
Kohji ShibanoJapanTokyo International University
Takayuki K. SatoJapanHewlett-Packard Japan Ltd.
Tateo KoikeJapanHitachi Ltd.
Kyongsok KimKoreaPusan National University
Chilkhaasuren OchirbatvnMongoliaMongolian National Institute for Standardization and Metrology
L. Nasan BuyanObserver - MongoliaMongolian Embassy, Tokyo
Ts. Bold BaatarObserver - MongoliaMongolian Embassy, Tokyo
Teow-Hin NgairSingaporeNational University of Singapore
Sten G. LindbergSwedenIBM Svenska AB
Bruce PatersonUK; EditorConsultant
Asmus FreytagUnicode ConsortiumAsmus Inc.
Arnold WinklerUSAUnisys
Michael KungUSASilicon Graphics
Michel SuignardUSAMicrosoft
Mike KsarUSA, ConvenerHewlett-Packard Company

1.3 Brief announcements

Mr. Mike Ksar informed the delegates of the following:

A seminar on standards aspects of Global Information Interchange is scheduled for 24--26 January 1996, in Geneva. It is organized by ISO, IEC and ITU. WG 2 members should be interested in making sure that 10646 is one of the key standards adopted by this seminar for GII. Registration forms are available from the convener, and are to be returned to Mr. Keith Brannon, ITTF.

The www ISO Home Page under subject of SC 2 has an entry stating that Amendments 1 to 4 have passed SC 2 ballot, as part of advance notification on 10646.

The technical corrigendum no. 1 (on Æ) has been sent to ITTF by the secretariat. The final text will be distributed for WG 2 for information.

The editorial corrigenda text was sent to ITTF for information - so that it can be sent to CEN as additional information during their adoption of 10646. A decision has not been made as to when to publish this corrigendum text.

2. Approval of the agenda

Input Document: N 1245R Call for WG2 Meeting 29 in Tokyo; Ksar, Convener; 1995-08-28

The draft agenda was reviewed and several new items were added to the agenda. Some additional documents that were identified for discussion later during the meeting were also accepted. The following is the modified agenda that was adopted for the meeting:

1. Opening and roll call: N1251

2. Approval of the agenda: N 1245R

3. Approval of the minutes of meeting 28 in Helsinki, Finland: N 1253

4. Review action items from previous meeting: N 1253

5. Status of actions requested from JTC1 and ITTF: N1264

6. Status of documents sent to SC2

6.1 pDAM-5 results on Korean Hangul: N 1249

6.2 pDAM-6 on Tibetan: N 1255

6.3 pDAM-7 on Additional characters: N 1256

6.4 pDAM-8 on Informative annex on CJK ideographs: N 1257

6.5 Corrigendum No. 1: N 1223R

7. Disposition of comments on pDAM-5 - Korean Hangul: N 1265

8. Repertoire issues:

8.1 APL: N 1087, N 1129

8.2 Runic: N 1210, N 1222, N 1229, N 1230, N 1239 N 1262

8.3 Byzantine music symbols: N 1208

8.4 Mongolian script: N 1226, N 1248

8.5 Uyghur, Kazakh and Kirghiz in Arabic script: N 1225, N 1246

8.6 Proposal by Iran to add two characters to Arabic script: N 1247

8.7 Ogham: N 1103R

8.8 New symbols.

8.8.1 Keyboards: N 1138, N 1235, N 1267

8.8.2 Control characters: N 1174

8.8.3 Electrotechnical symbols: N 985, N 1146

8.9 Translation of character names: N 1219, N 1259,N 1260

8.10 Unique identifiers: N 1231, N1266, N 1272, N 1277

8.11 Proposal by TCA to add 210 Kang Xi radicals: N 1182

8.12 Missing Pinyin characters: N 1282

8.13 Ideographic structure symbols: N 1284

8.14 Braille: N 1279

9. Defect Reports

9.1 Review status of editorial defect reports: N 1207, N 1193

9.2 Review status of non-editorial defect reports: N 1194

10. IRG status and reports: N 1261, N1280,

11. Defect reports status: N 1213, N 1214, N 1228

12. Liaison reports

12.1 Unicode Consortium: N1276

12.2 AFII

12.3 SC 22/WG 3 & SC22/WG 20

12.4 ITU

13. Other business

13.1 Second edition of 10646
13.2 Status of work: N1241
13.3 Issues list: N 1278 JP
13.4 Information on GBK: N 1281 China
13.5 Future meetings

14. Closing

14.1 Approval of resolutions

14.2 Adjournment

3. Approval of the minutes of meeting 28 in Helsinki, Finland

Input Document:

N 1253 Draft minutes of meeting 28 in Helsinki; Uma, Ksar; 1995-08-31

Dr. V.S. UMAmaheswaran presented the highlights of meeting 28 in Helsinki, referring to the different sections of document N 1253. Some corrections had been communicated to the convener and the secretary before the meeting. The minutes were accepted with the following corrections:

Change 'Mr. Mike Everson' to 'Mr. Michael Everson' wherever it occurs in the document N 1253.

Change 'Mr. Sven Thuygesen' to 'Mr. Sven Thygesen' wherever it occurs in the document N 1253.

Section 1.1: Add after the first sentence under 1.1 "... delegates":
The venue of the meeting was Hotel Aurora in Helsinki, Finland.

Section 4.1, Table N1203-12.2, Action Item AI 26.3-b, Status column;
change '... N 1216' to ' N 1257'. N 1216 is the wrong reference.

Section 4.1, Table N1203-12.3, Action Item AI-27-3-d, Status column:
change '... N 1137).' to '... N 1131).'.

Section 4.1, Table N1203-12.3, Action Item AI-27-7, middle column describing the action item, 1st line:
Change 'to prepare N 1999 ... N 1998..' to 'to prepare N 1199 ... N 1198 ..'.

Section 6.4.7, Discussion, Add a new item 3:
Mr. Michael Everson: Ireland does not accept this unification unequivocally, this is not what was agreed. What has happened is that a proposal involving this solution (unification of CAPITAL SEIMHIU and SMALL SEIMHIU with COMBINING DOT ABOVE) has been suggested by Dr. Asmus Freytag. Ireland has agreed to withdraw our proposal for the two SEIMHIU characters pending further study. We explicitly reserve the right to revisit this committee with further proposals regarding these characters should Dr. Freytag's solution prove inadequate for our needs.

Section 12.2, Action Item AI-26-12 should be removed from outstanding action items list - it is Completed.

Section 12.3, Action Item AI-27-4 should be removed from outstanding action items list - it is Completed.

4. Review of action items from meeting 28

Input Documents:

N 1253 Draft minutes of meeting 28 in Helsinki; Uma, Ksar; 1995-08-31

Dr. V.S. UMAmaheswaran reviewed the status of outstanding action items, referring to the list under section 12, in document N 1253, including the corrections noted as items 7 and 8 in section 3 above. Several items had been completed, and the others were either in progress or still outstanding. Some items that had not progressed were removed from the list. The following tables show the status of the various action items. (The notation AI-mm-n-k is used for the action items, where, AI stands for Action Item, mm is the WG 2 meeting number when it was first assigned, and n is the sequence number of a group of action items from that meeting, and optionally, 'k' represents the specific action item -a, b, c, d etc. where there is more than one.)

Action Items of meeting 28

5. Status of actions requested from JTC1 and ITTF

Input Document: N 1264 Letter from SC2 to JTC1 & ITTF on WG2 requests; Jan van den Beld; 1995-09-13

Mr. Mike Ksar introduced document N1264, which is a letter from SC 2 secretariat to JTC 1 on status of action items in SC 2. It was pointed out that the date of the next SC2 plenary in the letter was in error - the 6th SC2 plenary meeting should be during the week of 19th August 96. WG 2 meets 12-16 August 1996. Some points from the letter were discussed.

The SC 2 secretary has forwarded the text of Technical Corrigendum No. 1 on Æ to ITTF (Mr. Keith Brannon). Looks like ITTF is processing this as another ballot in JTC 1. It was pointed out that there was no need for another JTC 1 ballot. The Annex numbering is done by ITTF in the sequence in which they are received. Annex Q (of Technical Corrigendum 1) was changed to P by ITTF. There is also an error in the proof copy of the document. Mr. Mike Ksar will follow this up with SC 2 secretariat and with ITTF.

The letter to NBs warning that their status will be dropped from P to O if the ballot responses are not sent in, seems to be drawing the attention of at least some NBs. Germany and France have voted on the Technical Corrigendum No. 1. Singapore only joined recently. However, they will be participating as P member.

Both Japan and Canada sent in - unsolicited - nominations for candidates for the next SC 2 chair. Mr. Mike Ksar requested Japan and Canada to get together and decide who wants to be the Chair towards assisting the secretariat with the decision. However, this was NOT resolved at this meeting.

Mr. Mike Ksar also pointed out that ECMA may not consider continuing as secretariat for SC 2 if there are no approved projects for ECMA TC1 (technical committee for coded character sets). An ad hoc of ECMA TC1 was also held during the week to discuss possible projects for ECMA TC1. The resources of ECMA that have been made available to support SC2 - such as their role as the Registration Authority for ISO-IR, and their publication resources for ECMA standards that have been used to assist in publishing the 7 and 8-bit standards -- have proven to be valuable, and can be useful in WG 2's work also. Delegates were requested to consider potential work that could be made into an acceptable program of work for ECMA TC1. (Note: The notes taken during the discussion on ECMA-TC1 role were lost due to a system crash and files being irrecoverably lost, in spite of saving after every set of changes. All attempts to recover the lost files were unsuccessful -- meeting secretary)

6. Status of documents sent to SC2

6.1 pDAM-5 results on Korean Hangul

Input Document: N 1249 Table of replies and comments on pDAM-5 (Hangul); SC2 N 2623; 1995-09-08

The pDAM-5 passed SC 2 ballot. Document N 1249 contains the table of replies and comments for the SC 2 ballot on pDAM-5 on Korean Hangul syllables. Of the 23 active P members, 13 P members were in favour (some with comments), 3 disapproved with comments and 2 abstained. Mr. Sven Thygesen informed the group that Denmark's positive vote has not been included in the tally. Of the 10 other P members, 3 approved, 2 disapproved with comments. None of the 10 O-members responded. Dr. Teow-Hin Ngair informed the group that Singapore should have been listed as an 'active P member' instead of 'other P member'. Document N 1265 was prepared by the project editor as the proposed disposition of comments, and was discussed under agenda item 7 (see below).

Action Items:

The Danish member body is to follow up with the SC 2 secretariat about the missing entry for Denmark's ballot in document N 1249.

Singapore member body is to follow up with the SC 2 secretariat about the status of Singapore as an' active P-member' instead of an 'other P-member'.

6.2 pDAM-6 on Tibetan

Input Document: N 1255 pDAM-6 - Tibetan Character Collection SC2 N 2627; 1995-09-06

pDAM-6 is under ballot in SC 2. NBs are encouraged to return their ballots on time. The deadline is 1995-12-20. There was no discussion.

6.3 pDAM-7 on additional characters

Input Document: N 1256 pDAM-7 - 33 Additional Characters (Hebrew, Long S, DONG); SC2 N 2628; 1995-09-06

pDAM-7 is under ballot in SC 2. NBs are encouraged to return their ballots on time. The deadline is 1995-12-20. There was no discussion.

6.4 pDAM-8 on informative annex on CJK ideographs

Input Document: N 1257 pDAM-8 - New informative annex on CJK Ideographs SC2 N 2626; 1995-09-06

pDAM-8 is under ballot in SC 2. NBs are encouraged to return their ballots on time. The deadline is 1995-12-20. The following comments were made on pDAM-8.

Dr. Asmus Freytag: The Unicode consortium is reviewing the document. They have no objection to the information that is in the document. However, the level of carefulness in preparing the document is not up to the standard level. The number of changes we need to bring it up to par is quite large. The document should have had a chance to be reviewed as a working draft in WG 2. The Unicode consortium will be encouraging the US national body to vote conditional NO. In the past the editor was given explicit instructions to process the document for further processing as a pDAM.

Mr. Bruce Paterson: The text that was distributed for pDAM ballot was indeed distributed prior to the San Francisco meeting (Meeting 26, October 1994). There were only some typographical errors - and the document (N 1059 from IRG) was discussed at that meeting. It is open to any national body to propose further refinements or corrections as part of their ballot response to pDAM-8.

Mr. Mike Ksar: The San Francisco meeting resolution (M26.3) was instruction to the editor to further process the document N 1059. The IRG editor did the editing and the principal editor accepted the document without any further checking. pDAM-8 was received at Helsinki and the editor was given the go ahead for its further processing.
A message that the document has been further updated by IRG was received by the convener. A document should not be changed while under ballot. Changes from the version that is being balloted should be clearly marked.

Mr. Takayuki Sato: Japan is proposing to ballot conditional NO - with a proposed updated text.

Mr. Arnold Winkler: A change to a pDAM has to be justified with some contributions in support of the change.

Mr. Zhang Zhoucai: China is interested in participating with Japan in updating the document once the pDAM-8 ballot period ends and all the comments are in.

Action Item: National bodies are encouraged to ensure that they review pDAM-6, pDAM-7 and pDAM-8 and return their ballot responses in time, including any contributions towards improving the documents.

6.5 Corrigendum No. 1

Input Document: N 1223R 10646-1 Corrigendum No. 1 (First Draft); Bruce Paterson, project editor; 1995-06-15

Document N 1223R contains marked up text of 10646 showing most of the editorial corrigenda that have been accepted by WG 2 in response to various defect reports that have been processed so far by WG 2. It includes the Technical Corrigendum on Æ but not all the amendments and some editorial corrigenda that are pending clarification. It will be used as the basis for the next edition of the standard.

Discussion:

Dr. Asmus Freytag: Refer Section 6.4 - Discussion on Long Names in document N 1223R, page 5. Would like to change some of the wordings: " ... denotes the customary meaning "... to "...a customary meaning". In Annex D, ZWJ and ZWNJ - are not controls in the sense that they activate a function. It is a request for a device. Would like to get it aligned with the wordings in the Unicode standard.

Mr. Bruce Paterson: The wordings from Annex K will be used and ensure the section 6.4 is consistent with Annex K. The words will be changed to ....'its customary meaning', to align with Annex K wordings. The changes shown in Annex D are purely editorial -- 'should' is changed to 'are'. The instructions given to the editor were followed in the current text.

Action Item: NBs and Unicode Consortium are invited to review document N 1223R and propose improvements if needed.

7. Disposition of comments on pDAM-5 - Korean Hangul

Input Documents N 1249 Table of replies and comments on pDAM-5 (Hangul); SC2 N 2623; 1995-09-08 N 1265 Proposed Disposition of Comments on pDAM 5 (Hangul); Bruce Paterson; 1995-09-26 Output Document: N 1285 Hangul Syllable Character Name Generation Algorithm; Unicode Consortium; 1995-11-08

Mr. Bruce Paterson introduced the document N 1265, referencing document N 1249. Comments have been received from Canada, China, Ireland, Japan, Singapore and USA. Some points could be resolved only by WG 2. In other cases the editor has made some recommendations.

Discussion:

General Comments:

Dr. Asmus Freytag: The pDAM responses in document N 1249 is ordered by NBs alphabetically. Some comments are shared by several NBs. Perhaps in the future, a comment that is shared by other NBs could be marked on the first occurrence of that comment. Is this the last ballot? Can we make technical changes to a CD?

Mr. Bruce Paterson: The secretariat does not do such a compilation and cross referencing. The editor may entertain such cross referencing while preparing the disposition of comments. Changes to a CD that has been balloted should not negate any of the positive votes received. The disposition of comments should be towards accommodating the comments accompanying negative ballots and thereby reversing them. If WG 2 decides that the accepted changes do not require another SC 2 ballot, then the pDAM progresses to DAM ballot at JTC 1 level. The opportunity for NBs who have voted positive to change to negative based on acceptance of 'disposition of comments at this meeting' still exists during the DAM ballot at JTC1 level.

Comments from Canada

Mr. Bruce Paterson: Comments from USA and Singapore are similar to those from Canada, and can be accommodated by the proposed disposition to Canadian comments. An existing implementation is in conformance with the current edition. An explicit means of distinguishing between pre and post pDAM-5 version still exists. In order to address the comments regarding conformance, a few wording changes have been proposed and depending on whether the final publication will be as an Amendment or as the second edition, the words used will change.
UK is very happy with the proposal from Canada. Is there anyone in this meeting who objects to adopting Canadian proposal?

Mr. Mike Ksar: Is this a technical change or an editorial change? Answer: The proposed change is to apply the holiday period only to conformance statements and free up the vacated positions for immediate reassignments.

Dr. Asmus Freytag: The Unicode Consortium is in favour of adopting the Canadian comment.

Professor Kyongsok Kim: Korea is willing to go along with whatever WG 2 decides.

Mr. Zhang Zhoucai: China abstains on this item at this meeting.

Mr. Takayuki Sato: Are there any restrictions on the use of vacated positions?

Mr. Mike Ksar: The vacated positions will be 'reserved for future use'.

Disposition: WG 2 accepted the comment from Canada, US and Singapore and accept the proposed modification to pDAM-5 by the editor. Japanese comment G- 2, and Chinese comment number 3, are also accommodated by this disposition.

Comments from China:

Comments 1 and 2:

Dr. Asmus Freytag: Chinese comment s do not indicate any alternate solution. We do not collect usage information on other characters.

Mr. Zhang Zhoucai: Chinese comments 1 and 2, are not in line with the principles of pDAM-5 itself.

Note: Chinese comment 3 is accommodated by the proposed disposition for the Canadian comment (see above).

Disposition: The Chinese first two comments cannot be accommodated.

Comments from Japan:

Comments E-1, E-2 and E-3:

Mr. Bruce Paterson: Recommended that WG 2 accepts all editorial comments E-1, E-2 and E-3.

Disposition: Accept E-1, E-2, and E-3.

Comment G-1:

Mr. Bruce Paterson: G-1 - there was no subdivision of work in WG 2. The editor has sent a subdivision of work proposal while the pDAM-5 was under ballot in SC 2. The subdivision of work was not questioned at the SC 2 plenary in Helsinki either.

Disposition: WG 2 notes the objection. In the future, SC 2 secretariat should be advised by WG 2 that pDAMs must have a sub-division information attached to the pDAM ballot.

Comments G-3, G4 and G-5:

Mr. Bruce Paterson: G-3 and G-4 cannot be accommodated - they go against the principles of the pDAM-5. (G-5 - see discussion on names under Irish comments below).

Disposition: G3 and G4 are not accepted.

Comment I-1:

Mr. Bruce Paterson: Comment I-1 draws attention to the potential overall implication of not looking at all the requests at the same time.

Disposition: WG 2 - is aware of the implication. The current practice is to look at each request as they come along, and depending on the maturity of the proposal proceed with the encoding in the standard.

Comments T-1, T-2 and T-3:

Mr. Bruce Paterson: T-1: The note was put in to inform the NBs at the time of balloting by the editor. Japan's suggestion is to delete the note from the DAM.
T-2: Different block names were proposed in the pDAM when both the blocks could be present for some time. If we accepted the Canadian et al comment, the existing block can be used. The disposition of comments proposes to revert to the old names.

T-3: See discussion on names under Irish comments.

Disposition: Accept T-1 and T-2 and their proposed dispositions.

Comment T-4:

Mr. Bruce Paterson: A relationship table between old and new code positions for the moved syllables is requested in comment T-4. At the Geneva meeting, Korea had promised a tool -- not a conversion table.

Professor Kyongsok Kim: Was not aware of the promise of a table. There is no problem in providing the conversion tables.

Dr. Asmus Freytag: A member of Korean national committee has delivered conversion table to the Unicode Consortium, and it could be made available to interested parties. As to the normativeness of the relationship the difficulty of converting from hard to soft versions should be pointed out. An authoritative statement relating the old code positions and new code positions for the moved characters -- for example, by using / freezing the same names for the characters - will serve the purpose.

Dr. Teow-Hin Ngair: Singapore is willing to assist in the verification of the contents of the tables.

Mr. Bruce Paterson: If the same names are kept for the old and the new positions, one could derive the relationship. The physical form may take the form of about 10 pages. The editor can include the tables in the DAM before it goes out.

Disposition: Accept the comment. Korea will provide the ftp site for the conversion tables -- without any Copyright notices.

Action Items:

Korean NB to set up an ftp site for the conversion table and inform others on how to access the information.

The editor, with assistance from Korean NB, is to prepare the table of correspondence between the old and new code positions and include in the text for DAM-5 ballot at JTC 1 level.

Comment T-5:

Mr. Bruce Paterson: Comment T-5 is in support of a tool to migrate from the old to the new, as was mentioned by the Korean delegation in Geneva.

Mr. Takayuki Sato: If a tool is provided for example via an ftp -- there is no 'assurance / guarantee' that the tool will deliver the results of the conversion / migration of the code positions.

Dr. Asmus Freytag and Mr. Michel Suignard: Where is it minuted as to the Korean delegation volunteering tools to assist? The appropriate section from document N 1171 was pointed out.

Mr. Bruce Paterson: Expectation of a tool blessed by WG 2 is not realistic. We can document the migration of the code positions, but not the tools related items as part of the standard.

Disposition: Comment T-5 is not accepted.

Comment from Ireland:

Mr. Bruce Paterson: Ireland has proposed a more harmonized, systematic romanized names for characters in the pDAM-5. Japanese comments G-5 and T-3 are also related to the names of characters.

Professor Kyongsok Kim: The current version of the standard follows the previous tentative agreement between South and North Korea. The older names were the tentative ones. The final South Korean version should be used. Have not checked if there is enough space. Will impact many characters.

Mr. Takayuki Sato: Japan's comment T-3 is also on the same subject. The proposal is to keep the names as they are -- but add an annotation to relate to their use in other standards. For example: HANGUL SYLLABLE GA (ka), to accommodate North Korea. The number of additional characters for annotation is not too many. The current set of names is picked based on a failed CD on transliteration of names.

Dr. Asmus Freytag: Unicode consortium is not in favour of changing the names from that in pDAM-5. We should not be adding annotations on characters as much as possible.

Dr. UMAmaheswaran: If the names in pDAM-5 are not the correct one - we have the opportunity to change these.

Dr. Asmus Freytag: We could change the naming of all the characters to something based on the principles used in identifying the elements that compose the names for these characters, instead of using the transliteration of names. We can also put annotations in Annex P for existing set of Hangul syllables. For example, trigraph 'END' versus trigraph 'E', 'N', 'D'.

Professor Kyongsok Kim: Character names spelling out the syllables does not explain how to transliterate. The suggestion would require greater understanding of the construction.

Mr. Mike Ksar: Looks like the Japanese comment T-3 or the Irish proposal will not be accommodated -- unless the Korean NB can think about artificial constructed names based on the three components and come up with a resolution.

Professor Kyongsok Kim: Publication of revised national standard corresponding to pDAM-5 is targeted before end of this year. The revised Korean standard has been drafted. The code tables are not in the current draft.

Mr. Zhang Zhoucai: The latest Korean standard document (from Professor Kim) being published still has one glyph in the Korean column of unified ideograph as a serious defect. The Korean NB has not fixed it nor reported to WG 2.

Professor Kohji Shibano: What does the new Korean standard relate to? Are the character names frozen? Can we request Korea to decide on the names soon? The DAM has to freeze the names - and once frozen, the names being normative, cannot change in 10646. if the current pDAM-5 is different from Korean published standard it will cause problems. The names in 10646 should be kept the same as those in current standard. WG 2 should request Korean NB to stop publishing the national standard if it is going to have names different from 10646. One of the rationales for the name change was due to the overlap date. If the overlap date is removed, should we change the names of existing characters?

Dr. Asmus Freytag: Is Korea going to react soon to the name change proposal? Is the new suggestion to keep the same name as in 1993?

Mr. Mike Ksar: Japan, Unicode and Korea are requested to work on the naming issue and come to an agreement. We need a Yes or No from Korea to any new proposal - it may take up to a month before Korean NB feedback will be available.

Professor Kyongsok Kim: Are we talking about the naming structure or the transliteration? The changes in current pDAM-5 were to avoid the duplication of names. In general, there were always two different proposals for names -- one based on structure and the other one based on transliteration. Cannot answer which one to keep immediately. Transliteration is the current decision in the NB. They may not be able to go to and fro.

Dr. Asmus Freytag: If our alternative is having to wait for a long time to come to a decision, then we cannot accommodate the Japanese or Irish comment.

Dr. UMAmaheswaran: The proposal seems to be - now that the question of duplication during the holiday period has been addressed by removing the deadline -- the names need not be changed from the old.

Mr. Bruce Paterson: WG 2's principle states that we will not change the code position and the name. In pDAM-5 we have broken both principles. If we can avoid changing the names, keeping the hex position change, we would have broken only half the principle. UK's main concern when considering the pDAM-5 ballot was the code position at the time of taking a ballot. UK will not have an objection to changing the names back to the old ones.

Professor Kyongsok Kim: Korea cannot accommodate the request to keep the old names.

Dr. Asmus Freytag: Question to Japan - if the names are listed algorithmically and we point to two transliteration names, then Korea need not change what they have. Would that be acceptable? The proposal is to point to an algorithmic description instead of naming all characters in order to avoid annotation etc.

Mr. Mike Ksar: A request to Korean NB - please do not proceed with Korean National Standard which is based on the pDAM-5, till there is agreement on names in DAM-5, and the other proposed disposition of comments on the pDAM ballot.

Professor Kyongsok Kim: KIAA is in a rush to make the Standard as soon as possible. The possibility of changes is very remote. The Korean NB position is that it is OK for putting annotations like:
a. HANGUL SYLLABLE GA (ka) or
b. HANGUL SYLLABLE GA (kiyeok a)
These are names used in pDAM-5 with annotation. The existing names from the 1993 standard are not kept. The alternative is to keep old name and add annotation to the name:
c. HANGUL SYLLABLE KIEYOK A (ga).
The tentative agreement to use 'KIYEOK A' between North and South Korea does not have a basis anymore.
As to whether names should be presented in the form of tables or algorithm, the Korean NB's preference is to have tables. If WG 2 decides to choose the algorithmic way of naming the code positions, Korean NB will have no objection. The names in the tables are for use by ISO Publications (for the benefit of non-Koreans). Suspect that the name tables will be the simpler rather than going through the computations involved in the algorithms. Korea does not have any objection to publishing as algorithm either.

Mr. Takayuki Sato: Japan's first choice would be 'a' above. No objection to 'b'.

Dr. Asmus Freytag: A simple algorithm could be used. On behalf of Unicode consortium, because of the systematic relationship between code positions and the components, an algorithmic way of naming may suffice. It may be more acceptable to use the algorithmic way to make the changes.

Professor Kyongsok Kim: A simple algorithm may not be simple to understand.

Mr. Sven Thygesen: I cannot believe that NBs will vote NO on the basis of the tables versus algorithms.

Dr. UMAmaheswaran: Does the algorithm produce the same character string as if the string is currently written out? The answer was yes.

Mr. Bruce Paterson: The names are typically meant for computer programmers for creating data bases etc. Does the algorithm also generate the annotations used? The answer was yes. Coming back to the Ireland question - the request is to put the new names as additional information in Annex P. If we can satisfy the Ireland comment also, we may be able to change their negative ballot.

Mr. Mike Ksar: In an algorithm description, we could put a few examples to show how the NAME STRING can be generated. Will Korea have an objection to use the Algorithms for generating the name in ISO tables also?

Professor Kim: WG 2 can discuss whether to reduce the ISO publication page count, by using the algorithm versus the table etc. Korea does not have any position on this.

Dr. UMAmaheswaran: Why don't we produce the DAM with tables alone and go ahead with the ballot? Prior to the final publication of the Amendment, we can have another decision point.

Mr. Bruce Paterson: We need to see the algorithm in a paper, and we need to take a ballot on it.

Disposition/Action Items:

The US NB and Unicode consortium are invited to document the algorithm for table generation mechanism for advanced information to WG 2. If the incorporation of this algorithm, after study by the national bodies, is considered suitable as a replacement for the name table, WG2 will have the opportunity to declare it as editorial or not.
On the last day of the meeting, Dr. Asmus Freytag tabled document N 1285, containing an algorithm for naming Hangul Syllables in pDAM-5. NBs are to review it for suitability for using in naming the Hangul syllables in pDAM-5.

The editor, with assistance from Professor Kyongsok Kim (Korea), Dr. Asmus Freytag (the Unicode Consortium) to decide whether we should add annotations to Annex P, to resolve the comments from Ireland; to relate the old Jamo names with the new Jamo names; and to update the document containing the disposition of comments and modified pDAM-5 text (N 1265) accordingly.

Korean NB is invited to take note of the following comment from Mr. Zhang Zhoucai: "The latest Korean standard document (from Professor Kim) being published still has one glyph in the Korean column of unified ideograph as a serious defect. The Korean NB has not fixed it nor reported to WG 2".

Relevant Resolution:

M29.2 (pDAM-5 - Korean Hangul syllables): Unanimous

With reference to documents document N 1249 and document N 1265, WG 2 instructs its editor to prepare an updated disposition of comments reflecting the discussion at meeting no.29. The editor is further instructed to prepare the text of Draft Amendment No. 5, and forward it to SC 2 secretariat for further processing.

WG2 expects the decision on the final presentation of names of Hangul Syllables as a table of names or using an algorithm would be an editorial matter.

8. Repertoire issues

8.1 APL

Input Documents N 1087 Suggested additions to repertoire of APL; SC22/WG3; 1994-10-07 N 1129 Editorial Improvements to Character Names of APL Functional Symbols; Bruce Paterson; 1994-11-24

Mr. Mike Ksar: The request from APL standards committee was addressed at the San Francisco meeting in 1994. The convener had an action item for contacting the APL committee to get a Proposal Summary Form filled. All the needed information had already been provided by the requester, though not in the form of a proposal summary form. With reference to document N 1087, items 1 and 2 were already dealt with as defect resolutions, the dispositions can be found in editorial corrigendum document N 1223R. Item 3 was a request for a new character - "APL Function Symbol Quad".

Discussion:

Dr. Asmus Freytag: Should we unify it with another character? Mr. Ken Whistler (a member of the Unicode technical committee) had already gone through this exercise.

Mr. Bruce Peterson: There are other characters White Square, White Rectangle, etc. which look like APL Functional Symbol Quad. Characters in code positions 25A1, 2610, 25AD, 25AF - are all candidates for being used as the equivalent. In Annex P (Technical Corrigendum No. 1) an annotation can be made to identify the use of the preferred one as the APL Function Symbol Quad.

Action Item: The convener is to send a liaison statement to APL subcommittee informing of them of the possibility of unification of the requested character with one of the existing characters 25A1, 2610, 25AD or 25AF, and seek their preference. Once their selection is known, an entry in the annex on additional information on characters will be made in the standard, to reflect its use as APL Function Symbol Quad.

8.2 Runic

Input Documents N 1210 Proposal Concerning Inclusion of the Runic Characters; The Swedish Member Body SIS; 1995-04-28 N 1222 Names and ordering of the Runic characters - comments on document N 1210; Michael Everson; 1995-05-20 N 1229 Response to Michael Everson comments (N 1230) on Runic; Sweden National Body; 1995-06-16 N 1230 Feedback on Runic; Michael Everson; 1995-06-21 N 1239 Icelandic position on Runic; STRi, Olafsson; 1995-06-23 N 1262 Consensus Name & Ordering for the Futhark (Runic); Michael Everson, expert contribution; 1995-09-19

The documents were tabled for information. They will be discussed at the next meeting.

Action Item: National bodies are to review the documents and provide further feedback.

8.3 Byzantine music symbols

Input Documents N 1208 List of Greek (Byzantine) musical notation system for inclusion in 10646; ELOT, The Hellenic Standards Body; 1995-03-30

An attachment to document N 1208 dated 1995-11-07 was tabled at this meeting. Professor Gregorios Staths presented the document. A piece of Christmas time music that was written using the Byzantine music notation, was played. Mr. Panagiotis Dellios gave further elaboration on the points mentioned in document N 1208 attachment. The user community is about 230 million people. The notation is not only for Greek Orthodox Church's religious music but also could be a platform for supporting traditional musical elements from other religious and non-religious areas. Encoding of this notation will assist creating electronic archives of the music for multimedia applications. If more information is needed Greece will be pleased to supply them.

Discussion:

Mr. Bruce Paterson: On the principle of inclusion of music notation - lot of music we hear everyday is notated on 'staff' (5 horizontal lines), and shapes positioned on the staff. The Western musical notation is therefore not completely suitable for encoding in 10646 - and requires a different data type - for example, a 'music data type'. The proposed Byzantine musical notation is more like an alphabet - unlike the western musical system. It can be embedded along with other text encoded in 10646 -- I do not see why we cannot accommodate the request for inclusion. I would recommend we accept the repertoire presented.
The names chosen in the proposal have the first word for each as 'music'; some have 'Byzantino' etc. as qualifier. I would like to see the qualifier in the first word itself - for example:' Byzantine Music'. Also, every name has the word 'sign' at the end, similar to the three music signs we currently have (reference Table 42, Row 26, positions 6D, 6E and 6F). Some consistency in naming of the characters seems to be in order.

Mr. Sven Thygesen: Couple of meetings ago, we had a discussion on Hebrew Cantillation marks. Is there a relation between them and Byzantine symbols? If they are similar, we could use some of the principles we used for encoding the Cantillation marks to the Byzantine notation also.

Mr. Arnold Winkler: The US NB has not studied the proposal in great detail. We have accepted that the repertoire is a candidate for inclusion. However, the feeling is that due to lower frequency of use, we are not sure about where it should be coded. We will be discussing this further.

Mr. Mike Ksar: In SC 29 activities, the musical coding is done, but what is proposed here is an alphabet for musical notation. Suggested that WG 2 accepts the repertoire of 220 characters to support the Byzantine Musical system. There seems to be some disagreement on its categorization as 'A', and on their inclusion in the BMP or elsewhere.

Mr. Michael Kung: Personal opinion - these characters do not seem to be candidates for the BMP. Need some more work about whether combination technology can be used or not.

Dr. Asmus Freytag: If we look at the way other characters are standardized -- we could avoid the word 'sign' in the character names. Also the preference is not to use hyphen in the names.
From the Unicode Consortium point of view - the experts have considered the earlier document, had a discussion in their Boston meeting. The conclusion was - we do not think that these symbols are completely executable like the Western music. There is sufficient similarity to characters that they could be considered for coding. If they are to be coded, because of their relationship to other alphabets and other factors, the preference would be to encode them outside the BMP. Codes outside the BMP are no less characters than those in the BMP. The consideration is based on the relative frequency of use vis a vis other repertoires. There are other considerable collections of characters that may be more appropriate to include in the BMP.
Not sure that we are ready to accept the repertoire. People have focused on the overall question of whether in principle the Byzantine notation system qualifies for inclusion. The repertoire still needs to be examined in greater detail. There are some character shapes - such as the one in the block A8 to AB, D8 to DB - for which we would like to see if composition technology has been investigated or not; also I would like to see some background rationale documented for reference and to answer questions from the experts in Unicode Consortium.

Mr. Panagiotis Dellios: With a different Greek extended set, we represented each sound separately. We are working with technology that currently does not support combinations. On page 16 of attachment examples of combined forms are shown.

Mr. Bruce Paterson: C-8 of proposal summary form shows there is use of composite sequences with reference to ELOT 1326, that would have been sent to WG 2. Hopefully it will contain more information.

Professor Gregorios Staths: These characters can combine with other characters. The proposed characters cannot be split any further.

Mr. Bruce Paterson: Show those characters that are capable of combining and being combined as additional information.

Mr. Michel Suignard: Would like to see a sample of the FONTS used in the personal computers. The tables shown in the proposal seem to be hand drawn. Not sure if we can accept the repertoire as presented without further clarification on several points.

Mr. Mike Ksar: Request to prepare the code charts using Computer Fonts rather than hand written.

Dr. Asmus Freytag: In the same light, I cannot distinguish between some of the similar looking characters. We need some explanation.

Mr. Takayuki Sato: As a matter of procedure, if we decide to put some characters outside the BMP, we need to create an NP. If Greek has a great urgency, then the only place we can put the characters is the BMP.

Professor Kohji Shibano: If we need to create a new part of the standard, we may not need an NP, but a subdivision of the project alone may be sufficient.

Mr. Mike Ksar: We have already an architecture to access the supplementary planes. This may be the first one to go outside the BMP.

Mr. Bruce Paterson: We now have Part 1. For supplementary planes, we need another part - Part 2. One can access the characters in the supplementary planes now using UCS-4, one does not need DAM-1 to access the supplementary planes, though it may be easier using DAM-1. The work of creating the text for Part 2 will be after we accept the NP or subdivision of work.

Mr. Takayuki Sato: Because of the additional delay that is involved in processing the NP or subdivision for a new part, is the requirement from Greece urgent - and if so, how can we accommodate their request?

Mr. Mike Ksar: The consensus seems to be that we accept the Byzantine musical notations set as a candidate for inclusion - in principle. The proposal needs more refinement in terms of the repertoire, the questions regarding the use of composition or not etc.

Mr. Sven Thygesen: If there is an association of scholars - may be a statement of support as to the completeness of the repertoire from them will be useful in the evaluation.

Dr. UMAmaheswaran: Who are the other interested parties - who can feed back information to Greece? For example, Bulgaria. Who in WG 2 is interested?.

Dr. Asmus Freytag: As to the urgency of the request, there are still a number of unanswered questions that have to be answered, and WG 2 will not be able to accept the proposal till these are satisfactorily answered. The Unicode Consortium is interested in receiving information from Greece - ahead of time, to enable their experts to have meaningful discussion and feedback. If discussion has to take place in Copenhagen meeting we need the input from Greece in time.

Disposition: WG 2 appreciates the presentation by the Greek delegates, giving us more details on the earlier submission. Request Greece to consider the feedback given at this meeting. WG 2 may have to get ready for potential next part of UCS, defining supplementary planes.

Action Item: Greece is invited to refine their contribution on Byzantine music notation, addressing the various comments received at this meeting, and distribute it in time for NBs and liaison organizations to review and be prepared for discussion at the Copenhagen meeting WG 2-M30 in April 1996.

Relevant Resolution:

M29.8 (Greek Byzantine musical notation): Unanimous

With reference to document N 1208 and its attachments from Greece, at the request of the Greek national body, WG 2 appreciates the information on Greek Byzantine Musical Notation presented at this meeting, and invites the Greek national body to create a revised proposal based on feedback from interested parties for consideration at meeting no. 30.

8.4 Mongolian script

Input Documents N 1226 Mongolian character set proposal; Mongolian National Institute for Standardization and Metrology (MNISM), Mongolia, Ochirbatvn CHILKHAASUREN; 1995-06-12 N 1248 Proposal to add Mongolian by the Mongolian National Institute for Standardization and Metrology; MNISM - Mongolia; 1995-08-28

N 1268 Strategy for Coding Mongolian Script in UCS Hugh Ross, UK, expert contribution; 1995-10-11

N 1273 Feedback on Mongolian Script,, Unicode Consortium, Joe Becker, 1995-11-06

Output Documents N 1286 Ad hoc report on Mongolian Script in Tokyo; China, Mongolia, Unicode and U.K.; 1995-11-08 Presentation:

Mr. Chilkhaasuren Ochirbatvn from MNISM introduced (with the assistance of a translator from the Mongolian Embassy in Tokyo) the document N 1248. Modified document N 1248 page 1 was distributed. He expressed his appreciation for being invited to this meeting and facilitating inclusion of the Mongolian script. Mongolia submitted its requirement in March 1993 to the ISO/IEC. This proposal has been discussed two times before and has been postponed each time. The initiator of this proposal is MNISM. Because at least 5 P members are to vote on the subject, we would like to identify who are the P members interested on this subject. US, UK, South Korea, PRC, Unicode, Germany and Mongolia are some known interested parties. Request the meeting to identify the participants on this topic. Because there are no rules for making another WG inside a WG, the coordination has to be done by the convener of this WG. The Mongolian proposal has included only the classical form of the Mongolian script. The ancient variants of Mongolian are not used. We may need to add more characters to include other characters used for Mongolian. Document N 1248 contains a code proposal. He invited comments from the convener. He also had some comments on comments from China. He expressed his hope this meeting will make clear decision to make it easier to come back with a final proposal to WG 2.

Discussion:

Mr. Mike Ksar: We would need a technical requirement for Mongolian to be identified. There are several interested parties -- several of them are P members attending this meeting. We have some input from China and Unicode consortium indicating they are interested on Mongolian. We also have input from UK expert. Other P members like Germany, South Korea have not expressed their interest on Mongolian to this WG. We need to hear about the linguistic requirements etc. on Mongolian from the Mongolian experts

Bruce Paterson: Document N 1268, an expert's (Hugh McG Ross) contribution, attempts to raise some questions and provides some answers. May be it is easier to discuss in a small group of people interested in Mongolian.

An ad hoc group consisting of Messrs. Chilkhaasuren Ochirbatvn (Mongolia), Bruce Paterson (UK), Asmus Freytag (Unicode Consortium) and Mao Yong Gang (China) discussed the various documents on the subject and presented their report - document N 1286.

Mr. Mao Yong Gang presented the ad hoc group's report - see document N 1286 for details. The relevant inputs were considered by the ad hoc. Coding principles from the contributions by Messrs. Joe Becker and Hugh Ross have been accepted in principle. The improved contribution from Mongolia will explain the principles and address the various comments received. China will hold an open meeting on Mongolian - December 5, 1995, and Mongolia will host the second meeting tentatively set for February 10, 1996. The meetings are open to all. The Unicode Consortium has offered to invite the relevant experts on Mongolian, to UTC meetings, towards improving the proposals.

Mongolia received feedback on document N 1248. This document will be updated based on input from China, UK, and Unicode consortium. A revised proposal will be tabled for consideration at the next meeting in Copenhagen. Mongolia confirmed they did not have any problems with the ad hoc report in document N 1286.

Mr. Takayuki Sato: What was the major issue for Mongolian? See document N 1286.

Mr. Mike Ksar: The proposal is not perfect - there are several issues - 6 from McGregor Ross, and others from Unicode consortium.

Relevant Resolution:

M29.7 (Mongolian): Unanimous

With reference to documents document N 1226 from TC 46 and document N 1248 from Mongolia, WG 2

appreciates the work of the ad hoc group on Mongolian script created at this meeting

accepts its report in document N 1286 which includes the principles of encoding and the beginning of a time able for further development, and,

invites the member bodies of Mongolia, China, Ireland, UK, the Unicode Consortium, and other interested parties, to create a joint proposal on the Mongolian script based on further discussion.

8.5 Uyghur, Kazakh and Kirghiz in Arabic script

Input Documents N 1225 Proposal for Uyghur, Kazakh and Kirghiz in Arabic Script; China; 1995-06-26 N 1246 Comments on Chinese proposal for the Uyghur, Kazakh, and Kirghiz languages - document N 1225; Khaled Sherif, Egypt; 1995-08-28

Document N 1246 is being studied in China by the Uyghur experts.

The Unicode Consortium's comment is -- no presentation forms should be added to 10646, and therefore, there is no action to code from WG2.

Action Item: China is invited to take these comments as feedback to the appropriate experts.

8.6 Proposal by Iran to add two characters to Arabic script

Input Documents N 1247 Proposal to add two characters in ARABIC block; Iran - ISIRI; 1995-05-28 N 1275 Proposed Disposition of Iranian Proposal (N 1247) on Arabic Script, Unicode Consortium, Joe Becker, 1995-11-06

Discussion:

Mr. Mike Ksar: The two requested characters are already there in the standard. Refer to document N 1275 (feedback from Unicode consortium). There is no reason to add these as new characters - perhaps to include this Annex P. The contribution in document N 1275 can be considered as response to Iran.

Dr. Asmus Freytag: Some additional information addressing the use such as that proposed by Iran can be included in Annex P.

Relevant Resolution:

M29.3 (Iranian request): Unanimous

With reference to documents document N 1247 and document N 1275, WG2 instructs:

its convener to communicate to the Iranian national body that:
Iranian PSP (pseudo space) can be unified with
200C ZERO WIDTH NON-JOINER
Iranian PCN (pseudo connection) can be unified with
200D ZERO WIDTH JOINER
along with the explanatory text in document N 1275.

its editor to prepare an editorial corrigendum capturing the explanatory text on the usage of ZWNJ and ZWJ provided in document N 1275.

8.7 Ogham

Input Documents N 1103R Proposal for coding Ogham script; Ireland Michael Everson; 1994-10-11

There was no discussion at this meeting.

8.8 New symbols

8.8.1 Keyboards

Input Documents N 1138 Proposal to add new characters (Keyboard related) to 10646; Liaison statement from SC18/WG9 to SC2/WG2 Alain LaBonté; 1995-01-30 N 1235 Proposed correction to attachment of document N 1138; La Bonté; 1995-06-26 N 1267 Liaison statement from SC2/WG2 to SC18/WG9; Alain LaBonté, liaison, SC 2 to SC18/WG9; 1995-09-11 N 1258 Repertoire additions for 10646-1 Cumulative List No. 2-; Bruce Paterson, editor; 1995-09-05

Document N 1138 contains the original request from SC 18 WG 9. Document N 1235 contains suggestions to correct some entries in document N 1138. Document N 1267 is a liaison statement from SC2 WG2 to SC 18 WG 9. Document N 1235 replaces document N 1138.

Discussion:

Mr. Bruce Paterson: When is it appropriate to propose changes to names of the symbols? UK would like to change the names to be more consistent with others. Would like to be able to change and they are not frozen.

Dr. Asmus Freytag: As much as possible we should have the names as close to fixing as possible. A pool of documents containing an accepted set of symbols should be maintained. Such documents should contain only the symbols that have been accepted in principle. If a proposal has been only partially accepted then only that part that is accepted should be included. There is also a question of fonts of the right kind. The next edition of Unicode standard will be using set of tools and is expecting true type fonts. The same set of tools can be used for publishing 10646 also.

Mr. Bruce Paterson: Document N 1258 R contains the beginning of a Pool of Documents to keep track of. It contains the list of papers that contain characters or symbols that have been accepted in principle to be included in the standard.

Mr. Mike Ksar: Would like to have glyphs using true type fonts.

Mr. Michel Suignard: A product called Fontographer does the conversion from Bit Maps or EPS files etc. to True Type.

Mr. Alain LaBonté: Agreed to provide the symbols for Keyboards in true type fonts.

Disposition: WG 2 will continue to accumulate accepted symbols for encoding, with names as close as possible to the final. Suggestions for consistent naming is still possible prior to the encoding decision in WG 2.

Action Item: Mr. Alain LaBonté is invited to provide the keyboard symbols in document N 1258R in true type fonts to the editor.

8.8.2 Control characters

Input Documents N 1174 Query on the representation of Control Characters; Michael Everson, Ireland, expert contribution; 1995-03-14 There was no discussion at this meeting.

8.8.3 Electrotechnical symbols

Input Documents N 985 Electrotechnical Symbols for First Addendum; Hugh McG Ross UK.; 1994-03-28 N 1146 Extra Electrotechnical Symbols - proposal summary form for N 985; Hugh McG Ross; U.K.; individual contribution

Dr. Asmus Freytag: Document N985 has to be reviewed. At least one symbol Greek Epsilon Symbol - can be unified.

Action Item: Dr. Asmus Freytag is invited to review document N 985 and provide feedback on which characters can be unified, which can be accepted etc.

8.9 Translation of character names

Input Documents N 1219 Comment on translation of character name of ISO/IEC 10646-1; Japan - TKS; 1995-05-03 N 1259 Translation of character names - proposal for a WG2 policy; Bruce Paterson, personal contribution; 1995-09-11 N 1260 Concerns on change of character names of 10646; Japan - TKS; 1995-09-11

The subject of whether or not character names in 10646 can be translated in non-English language versions of 10646 has been discussed in the previous two meetings. The subject matter was also discussed at the last SC 2 meeting and the following resolution was adopted by SC 2, with an action item on WG 2.

SC2 RESOLUTION M5.4 (Translatability of Character Names in ISO/IEC 10646-1: 1993)

UNANIMOUS

SC 2 resolves with respect to SC2/N 2605:

a. SC 2/WG 2 Resolution M27.21 is canceled

b. WG 2 is instructed to prepare a proposal on the properties of character names with reference to their translatability and their potential use as identifiers, taking account of, among other aspects, the current practice in other SC 2 standards, at its next meeting.

c. SC 2 requests ITTF to postpone the publication of any French translation of ISO/IEC 10646-1: 1993 that includes translation of character names, until SC 2 has agreed to the proposal prepared by WG 2.

Mr. Mike Ksar: The different points to discuss are:

Response to SC 2 resolution

Note that resolution M27.1 of WG 2 (Geneva) meeting is canceled

Answer whether 10646 character names are translatable or not

If they are translatable, ITTF hold on translated text should be released

What appropriate text is to be included in 10646?.

Based on the discussion on unique identifiers (see agenda item 8.10), we have accepted the need for unique identifiers independent of languages, and have an agreement in principle on a proposal. We may be able to suggest to SC 2 that a national body is free to translate the names if they so choose.

Discussion:

Mr. Alain LaBonté: It would be OK to have such a statement minuted.

Mr. Bruce Paterson: There is no clarification in the text of 10646 as to whether the names are translatable or not. From the point of view of UK, a clear statement must be made in 10646 to prevent such questions from being brought up again to WG 2. Document N1223 proposes a simple sentence or note to prevent this problem.

Mr. Panagiotis Dellios: One could say there is one main version, and other versions may be 'less important' etc. We provided all the names in Latin Script for 10646.

Mr. Bruce Paterson: ISO directives cover all these. We will be wasting our time discussing this.

Mr. Takayuki Sato: We said for some reason before, that the English version is the only one. Now we are saying that we would like to revisit.

Mr. Mike Ksar: We have agreement in principle on unique identifiers; I think the question on whether the characters are translatable or not seems to be answered as YES.

Mr. Sven Thygesen: There is a concern that the meaning of the names are not changed during translation.

Dr. Asmus Freytag: What should the naming guidelines be for the translated versions? Will they have accented characters or not?. Should WG 2 have any guidelines on other Language versions of 10646 - on naming guidelines? If ISO publishes a French Language version of the standard what is the equivalent of Annex K?

Mr. Sten Lindberg: In Sweden we do translate the names of characters; A Overcircle, etc. we do translate as appropriate in the language.

Mr. Alain LaBonté: The ISO translation guidelines are clear on the translation of standards.

Mr. Bruce Paterson: Annex K should have stated that these naming guidelines are for English language versions of the standard. There are other standing guidelines for translating names of characters. The 10646(E) version is the collection of unique English names for use in English language versions of the standard. Document N 1223 addresses several of these points. Volunteered to submit a proposed clarification text for inclusion in the standard.

Mr. Takayuki Sato: Explained the points raised in document N 1260 by Japan. The discussions on changing existing names in 10646 was raised as a 'time waster' and suggestions are proposed to rectify the situation. There was general agreement to the proposals and they were accepted. The second bullet - proposed text will be included in Annex K. The first bullet is covered by a reaffirmation by WG 2 of Clause 12 of the standard.

Disposition: Prepare a resolution 'because we have unique identifiers the names are translatable in different language versions'. Mr. Bruce Paterson will prepare clarification text regarding translatability, and addressing points raised in document N 1260 by Japan. WG 2 reaffirms Clause 12 in the standard. See relevant resolution M 29.5 below.

Relevant Resolution:

M29.5 (Translatability of names): Greece and Denmark Abstain

With reference to resolution M5.4 from the SC2 plenary in Helsinki, and documents N 1219, N 1231, N 1259, and N 1260, in light of the acceptance of resolution M29.4 on Unique Identifiers (see section 8.10),

WG 2 expresses its opinion to SC 2 that the names in 10646 may be translated in other language versions of the standard, and requests SC 2 to inform ITTF and SC 2 member bodies accordingly.

WG 2 instructs its editor to prepare editorial corrigendum reflecting the proposals for clarification text in the standard, for consideration at meeting no. 30.

8.10 Unique identifiers

Input Documents N 1231 Uniqueness of names; Canadian National Body; 1995-06-22 N 1266 Unique identifiers for characters in 10646; Liaison SC22 to SC2; 1995-09-22 N 1272 NP on UCS Character Identification; Canada; 1995-10-26 N 1277 Comments on SC22 document N 1968, "Unique Identifiers for Character Names"; Ed Hart; 1995-11-02 SC 2 N 2616 Resolutions of SC2 meeting # 5, SC2 Meeting 5, Helsinki, Finland: 1995-06-29--30; 1995-06-30

Presentation:

Mr. Mike Ksar: We discussed document N 1231 in Helsinki. Canadian body has proposed an NP in document N1272.

Mr. Alain LaBonté presented document N 1272. There is currently no notion of identifiers in 10646. The requirements, leveraging on SC 22/WG 20 (see document N 1266), is formulated as a proposal for NP. The requirements for a unique identifiers is in section 3 of document N 1272. The Annex proposes a simple scheme for an identifier - including the ability to handle the rare situation when a code position is reallocated in UCS.

Discussion:

Mr. Mike Ksar: Would like to alert WG 2 that WG 2's charter is such that we do not have the responsibility to meet all the identified requirements.

Dr. UMAmaheswaran: Document N 1272 from Canada summarizes the requirements known to WG 2 so far. A proposed solution is in the Annex to document N 1272. The proposal is to create a normative annex to 10646.

Dr. Asmus Freytag: The proposed solution needs more work.

Mr. Arnold Winkler: All of the stated items are within the scope of SC 2/ WG 2. Mr. Ed Hart's contribution in document N 1277 summarizes the requirements and proposed solutions.

Mr. Sven Thygesen: Has the annex been discussed at the SC 22 /WG 20 level? The answer was No.

Dr. Asmus Freytag: Document N 1277 also summarizes all the known requirements. Clarification not to move characters by WG 2 is needed. We can take a look at the requirements and can examine the proposed solution and see if it is acceptable.

Mr. Panagiotis Dellios: According to the Rome agreement, there is a sensitivity to the language versions of the standard. It costs a lot to have translated versions of the standard. It is difficult to say yes or no at this moment.

Mr. Mike Ksar: It is up to the national body to determine whether an ISO standard is to be translated into a particular language.

Mr. Alain LaBonté: In 10646 as it is published there is no notion of identifiers. In several other ISO standards the names have been translated to different language versions including names of characters. The proposal is for an annex or a paragraph in 10646, including the solution in Mr. Ed Hart's paper - document N 1277.

Mr. Michel Suignard: We should not reopen the discussion of names being used as identifiers. A mechanism to identify version numbers is not desirable.

Dr. Asmus Freytag: Because it is after the pDAM-5 the identifier will not apply to pre-pDAM-5.

Mr. Bruce Paterson: Should we include version of the standard or not? We do have a principle not to move code positions. However, if we do move on a special basis, we do need a version identifiers. If we declare that the identifiers are post pDAM-5 only, then the 8 character ids will do. However, if we refer to pre-pDAM 5, then a special prefix will be needed. I would like to see the definition in main front part of the standard than in a normative annex.

Dr. UMAmaheswaran: If we reallocate code position assignment to characters the unique identifier cannot be maintained, even though the standard claims that code positions shall not be reallocated. The first character added is the version identifier as a prefix to the 8 character sequence.

Dr. Asmus Freytag: Are we able to work off the proposal in document N 1277 or do we have to go through further examination in a future meeting.

Mr. Bruce Paterson: Let us take the proposed solution in document N 1277 as the basis and see if we can make progress.

Mr. Sven Thygesen: Would like to check if the proposed solution can meet all the requirements from SC 22/WG 20. Any proposed solution should be examined in greater detail. Concerned about whether the solution is adequate. Also, has the discussion on mnemonics for character gone away? The mnemonics are not part of this standard.

Professor Kohji Shibano: We may have to cope with possibly three versions - pre-PDAM-5, post-pDAM-5 and when they are allocated new characters.

Mr. Arnold Winkler: Mnemonic discussion has taken place.

Dr. Asmus Freytag: We do have the option of working with a fixed version working off of a set of identifiers. Do we need four characters or eight characters?

Mr. Alain LaBonté: The mnemonics usage is based on using the xxxxxxxx (UCS-4 code position) as the anchor point.

Mr. Mike Ksar: Had discussion with Mr. Keld Simonsen - the xxxxxxxxx may be used as an anchor point for mnemonics.

Mr. Mike Ksar: I want to ensure that everyone understands the issues; the proposed solutions are in the papers. Want to ensure that the solution proposed by Canada, is NOT a requirement on any national body to translate the 10646 standard into any other language.

Mr. Panagiotis Dellios: There is an agreement between EU and NAFTA that the names shall not be changed.

Mr. Takayuki Sato: There seems to be an agreement on need for short identifiers.

Mr. Mike Ksar: We have to decide whether to use 4 or 8 characters, and on the version numbers.

Mr. Bruce Paterson: If the first four characters are all zeroes, they may be omitted. We do not discuss the version numbers at all until we have a problem.

An ad hoc group (consisting of Messrs. Arnold Winkler, Takayuki Sato, Teow-Hin Ngair, Michael Kung, Asmus Freytag and UMAmaheswaran) was charged with proposing a text for consideration by the meeting based on the discussion so far, and based on document N 1277. A draft text was produced by the ad hoc group, and was presented and further refined by Dr. UMAmaheswaran. Summarizing the key aspects:

Unique Language-Independent Machine Identifiers, with 10646 code positions as the anchor points

Fixed Length with Two options -- 4 or 8 hex digits

Optional prefixes + / - to distinguish between 4 and 8

Optional prefixes T / U to distinguish between pre-AM-5 and post-AM-5 allocation of code positions to some Korean Hangul syllables.

The text is to be one of the main clauses in the body of the standard.

The proposed text was discussed and it was decided that the exact wording of the text will be left to the editor. It will capture the agreement in principle on the components of the identifier. See relevant resolution M 29.4 below:

Relevant Resolution:

M29.4 (Unique identifiers): Denmark Abstains

With reference to documents N 1231, N 1266, N 1271, N 1272, and N 1277, on unique identifiers, WG2 accepts the requirement for unique identifiers and instructs its editor to prepare draft text for a pDAM on Identifiers capturing the principles in the following paragraph, and reflecting the subsequent discussion on it at its meeting no.29:

"A language-independent identifier for a character is a sequence of eight hexadecimal digits representing the hexadecimal value of its UCS-4 code position. Optionally, if the leading four hexadecimal digits are all zeroes, the four leading zeroes may be omitted. To distinguish between the four- and eight-digit forms of the identifier, an optional prefix character, a '+' (PLUS SIGN) for the four-digit form, or a '-' (MINUS SIGN) for the eight digit form, may be used. In addition to a '+' or '-', a further prefix letter 'U' (CAPITAL LATIN LETTER U) may be used. For code position assignments prior to the Amendment No. 5 (Korean Hangul Syllables) version, the prefix letter 'T' (CAPITAL LATIN LETTER T) shall be used instead of 'U'. These language-independent identifiers are not case-sensitive."

8.11 Proposal by TCA to add 210 Kang Xi radicals

Input Documents N 1182 Proposed to add 210 Kang Xi Radicals, TCA; TCA; 1995-03-23

See discussion under IRG report (Section 10, Item E, point 'b'), on feedback from IRG on the subject.

Action item: NBs and the Unicode Consortium are requested to review document N 1182 - proposal on radicals from TCA.

8.12 Missing Pinyin characters

Input Document N 1282 Two Pinyin Letters with Tone Marks missing; China; 1995-11-07

Presentation:

Document N 1282 was introduced by Mr. Zhang Zhoucai. It contains a request for including two missing Latin characters m-acute and n-grave that are needed to complete the Pinyin set of fully composed Latin characters. Note that these characters are erroneously called m-grave and n-acute in document N 1282 and were corrected during the meeting. All the other characters needed for Pinyin are unified with existing Latin characters. They are used quite frequently - exemplified by several Pinyin dictionaries.

Discussion:

Mr. Alain LaBonté: I would support the request. Both lower and upper cases should be included. Also, the missing n-grave is used in current technology elsewhere.

Dr. Asmus Freytag: The question as to why they cannot use - combination method - must be answered. Are these implemented in existing systems? Do we use only the lower case in implementations?

Mr. Zhang Zhoucai: All other characters needed by Pinyin set are already included. It is coded in GBK, exists in WIN 95/PRC in extended DOS.

Mr. Michel Suignard: Pre-composition is a requirement for ease of implementation. Both cases should be supported in order not to break previous implementations.

Mr. Arnold Winkler: We could perhaps process this as a 'defect report' due to incompleteness of repertoire to support Pinyin.

Disposition: China, Canada and US, are invited to submit a defect report on the subject. WG 2 accepts that the missing characters can be considered as a defect report. A proposed resolution to the defect report would be to include them in 10646 BMP as:

01F6 M UPPERCASE ACUTE

01F7 M LOWERCASE ACUTE

01F8 N UPPERCASE GRAVE

01F9 N LOWERCASE GRAVE

Relevant Resolution:

M29.10 (Missing Pinyin characters): Unanimous

With reference to document N 1282, WG 2 accepts the missing characters as a defect, and invites the member bodies of China, Canada and the US, to prepare a defect report for consideration by WG 2 at meeting no.30.

8.13 Ideographic structure symbols

Input Document N 1284 Ideographic Structure Symbols (Repertoire Addition Request); China; 1995-11-07

Presentation:

Mr. Zhang Zhoucai introduced document N 1284. It is a request for adding 13 ideographic structure symbols to the BMP. GBK has encoded 13 of these already. They are also in Beta version of Windows 95 at the system level.

Discussion:

Mr. Mike Ksar: Is this structure applicable to Ideographs in general or only in China? Was this discussed in IRG?

Mr. Zhang Zhoucai: IRG could not determine if it is within the scope of IRG. Also, for compatibility with national standards, we need it in 10646.

Dr. Asmus Freytag: We cannot simply take the code positions in GBK and use it directly without some review. It is a good feature that we are reducing the number from 3000+ using these symbols. I would like to encourage China to give additional information to enable other interested parties to make informed decisions.

Mr. Arnold Winkler: Could these be also used for Japanese?

Dr. Asmus Freytag: They could be used for some Japanese ideographs, but not for all. Would be better if they could be used for all 10646 users, not necessarily just for China.

Professor Kohji Shibano: The method of combination is not the same.

Mr. Takayuki Sato: Some Japanese characters can be used by any combination method. If the question was -- If it will satisfy a Japanese customer? -- the answer will perhaps be NO.

Mr. Zhang Zhoucai: China's request is to use the proposed symbols as an assistance for 3000+ only, and not for all ideographs. China can present this to UTC and discuss the results in Copenhagen.

(See also discussion under Section 10 - IRG report.)

Action Item: China is invited to discuss the proposal with other interested parties, including the Unicode consortium, and present additional information at the next WG 2 meeting.

8.14 Braille

Input Documents N 1279 Braille letters (Repertoire addition request); Japanese national body; 1995-10-27

Presentation:

Professor Kohji Shibano: The proposal from Japan was to use neutral names for the shapes used for Braille. A set of 512 shapes are proposed to be included. The proposal came from the Japanese Braille committee. The names came from the Japanese coding committee. The migration from 6 dots to 8 dots has happened.

Discussion:

Mr. Takayuki Sato: All NBs were invited to comment on whether the shape or the meaning to be coded. This is a response from Japan to that invitation.

Mr. Bruce Paterson: Here is an example, where a shape can have multiple meanings depending on the language. The shape is the criterion to be used. There are two rows with some holes proposed. Do we need to distinguish between basic, Upper etc. The first word could be 'Braille', the next one could be 'Symbol' and the third word could be just a number. There are already blocks of other characters which are only the shapes. We should not assume a particular model of only text - as it is traditionally known. There can be entirely Braille specific applications on their own, with bridges to the conventional text world. In those cases where the Braille form is only a presentation form of languages should be narrow

Mr. Michel Suignard: Based on previous answers - we have encoded on the basis of shapes. Unfortunately, we cannot do any processing based on the shapes alone. The only place these are used can be for presentation purposes or inputting purposes -- translated into another language depending on context.

Dr. Asmus Freytag: Would you not loose the meaning if the Braille was linked to a language. For applications searching based on language etc. this may not work if we decide to code on shapes. In the proposed tables, there are quite a number of positions which are left unused. The names could be just based on the hex value rather than possibly meaningless names. There may be a number of other shapes in the standard because of other reasons. We need consensus from the WG 2 members. There can be discussion on content. All NBs and WG2 members had enough advance notice for comments. One consequence of using this proposal - we have customary operations that could be performed on characters. For example, case conversions, classification of letters etc. Even in the Latin alphabet they can be language based. This model may break those text based character processing logic etc.

Mr. Alain LaBonté: Though it is based on shapes, compared to U umlaut or U trema etc. for example, a language tagging is additionally needed. Braille is a language on its own. French Braille, English Braille etc. are present. The shapes are the only those that count. In the same way as other text character based languages, the Braille has its own properties and is a language by itself.

Professor Kyongsok Kim: I do not see much problem in adding these characters to UCS. I am not convinced they are simply presentation forms for different languages.

Mr. Mike Ksar: The discussion seems to be going in the direction that -- some say they are presentation forms, some say they are not.

Dr. Asmus Freytag: Answer to Q 7 on presentation forms (in the proposal summary form) is NO - and the response for 'if YES reference' - does not explain much.

Mr. Bruce Paterson: Suggest removing the answer for the question 'if YES reference'.

Action Items:

The convener should forward document N 1279 with a cover letter (to be prepared with assistance of Dr. Asmus Freytag) to TC 173 / SC 4 in ISO, as a liaison document, and request feedback and confirmation that it meets their requirements. TC 173/SC 4 is developing 8 bit codes for Braille.

Japanese NB is invited to update document N 1279 based on the discussion at this meeting, for discussion at the Copenhagen meeting.

9. Defect reports

9.1 Review status of editorial defect reports

Input Documents 1193 Disposition of Editorial Defect Reports and instructions to the editor; Defect Reports ad hoc committee (Bettels, Everson, Sato, van Wingen, Winkler); 1995-04-06 1207 Editorial Corrigenda - Cumulative List - Issue 2 (update of document N 1131 R); Bruce Paterson, WG2 Project co-editor; 1995-04-26

See discussion under section 6.5 on Corrigendum 1 text in document N 1223R.

9.2 Review status of non-editorial defect reports

Input Document 1194 Disposition of Non-Editorial Defect Reports for ISO/IEC 10646-1, Geneva, April 3-7, 1995; J. Bettels, M. Everson, T.K. Sato, J. van Wingen, A. Winkler; April 6, 1995

See discussion under section 6.5 on Corrigendum 1 text in document N 1223R.

10. IRG status and reports

Input Documents N 1261 Conclusion List for IRG # 5; Zhang Zhoucai, IRG Rapporteur; 1995-08-22 N 1280 Additional Report on IRG Activities; IRG Rapporteur, Zhang Zhoucai; 1995-11-04

Presentation:

Mr. Zhang Zhoucai, the IRG Rapporteur, presented document N 1261. Three separate sections were identified: Prioritization of Vertical Extension repertoire, IRG Mission and on Ideographic Components.

A. Prioritization: of Vertical Extension

The IRG was requested to prioritize the initial collection of ideographs proposed for Vertical Extension. HCS (Han Character Set) - was the original proposal from China containing about 29000 characters in 1990. It was split into HCS-A and HCS -B. HCS-A is identical to UniHan in 10646-1993. HCS-B was merged with additional input from Japan, TCA, Korea initially and more recently with input from Vietnam and Singapore - about 20000 (?) characters were included in Draft 1. Draft 2 has been prepared with 3 levels of prioritization identified in document N 1261. Level I - 7084, Level 2 - < 400 and Level 3 < 4000. Japan had consolidated their requirement to less than 1000 prior to Seoul IRG meeting. China, Korea and Singapore also have reduced their requirements. All the characters selected for Level 1 are Category A. The drafts have been distributed in IRG.

IRG needs further instructions from WG 2 as to what should be included in the next draft of Vertical Extension. Draft 3 from IRG will reflect the accepted levels from WG 2 and will be proposed as repertoire to be added to 10646.

Discussion:

Mr. Takayuki Sato: Clarification - the total repertoire in Draft 3 will be same as in Draft 2. The only difference would be prioritization by IRG into levels 1, 2 and 3. Each IRG member country decided on what they can live with as the minimum that should be kept in Level 1 priority - the criterion was their own.

Mr. Mike Ksar: Looks like, there was no single criterion in IRG. What is the process used by IRG - for example, did Japan agree to include China's request for xxxx number of ideographs in Level 1 etc.

Mr. Zhang Zhoucai: The IRG member countries came up with their own individual prioritization, the resultant set were discussed in IRG and a compromise list was agreed to by Consensus within IRG -- the individual participants (countries) in IRG are satisfied with the resultant levels. IRG had a unanimous resolution on the subject of categorization.

Mr. Mao Yong Gang: China used its own criterion to arrive at its own prioritization. Each country had its own process of arriving at the prioritization.

Mr. Michael Kung: The CORE - 2147 ideographs - is unified Han. The only exception was some characters from the set of Chu Nom characters from Vietnam (category B1) are included in the core.

Dr. Asmus Freytag: It seems the process used by IRG is somewhat different from the process used in WG 2. We have not used the process that is used in IRG for other characters that are being added. Do we expect a different prioritization method?

Dr. UMAmaheswaran: Requested that IRG add a paragraph to one of the status reports describing the procedure used by IRG to arrive at the prioritization.

Mr. Sven Thygesen: Was the procedure documented in document N 1252 used? If the procedure in IRG was relying on the judgment of individual member countries, then the question to each IRG member country would be if they have followed the procedure in document N 1252.

Mr. Zhang Zhoucai: Yes - for example, only Category A characters were included.

Mr. Michel Suignard: Document N 1261 - when was it distributed? It was done mid-September 1995, to UTC.

Dr. Asmus Freytag: Not able to give a detailed feedback on the item at this time.

Mr. Michael Kung: The UTC has not had enough time to review the document from IRG and be able to feedback at this meeting of WG 2. It is a synchronization problem of the meetings.

Mr. Takayuki Sato: The countries participating in IRG have met and prioritized based on input from IRG members. It is up to WG 2 to look at the numbers presented by IRG and see if we accept the presented numbers or not. If we do not agree, we need to give additional instructions to IRG.

Mr. Zhang Zhoucai: The IRG meeting minutes have been distributed via E mail to the Unicode IRG representative Mr. Lee Collins. WG2 asked IRG to prioritize without any criteria for prioritization. We have done the work per these instructions.

Dr. Teow-Hin Ngair: Mr. Zhang has clearly indicated how the numbers were arrived at. It will be documented. Singapore is happy with the process used by IRG. We should look at how we can further progress.

Dr. UMAmaheswaran: Suggested that IRG be instructed to produce the document 'draft 3.' including Level 1, 2 and 3, and send it to NBs and liaison organization for WG 2 feedback.

Dr. Asmus Freytag: I support the proposal.

Mr. Michel Suignard: Endorse the statement that NBs have to give feedback. If it was reviewed by NBs ahead of time, then the delegates would be in a better position.

Dr. Teow-Hin Ngair: Is it the proposal that the entire document be sent to all NBs. (Yes). If so, IRG will need some time to prepare a document that has been proposed - due to additional work needed due to the possible unification with Chu Nom.

Mr. Zhang Zhoucai: Would like to see some more direction from WG 2 -- such as give a certain area for coding and we can fit the Level 1 into that zone and document it.

Dr. Asmus Freytag: For the member bodies who are not participating in IRG, it may be interesting to get additional information on the selection of characters. The IRG process should include the process WG 2 uses for evaluation of new requests. For information, UTC is meeting 7-8 December 1995. Documents for consideration by UTC have to be distributed; they should please be given to UTC in time.

Mr. Zhang Zhoucai: The proposal summary form used with the Vertical Extension needs to be updated.

Disposition: IRG is requested prepare an updated VE document, containing the entire repertoire, as well as the proposed Levels of Prioritization, indicating the current three proposed levels of prioritization, along with some rationale and summary explanation attached to it. National bodies are to review the updated document and feedback to WG 2 in time for the next WG 2 meeting. See relevant resolution M29.9 at the end of this section.

B. IRG Mission:

The scope has not changed from before. The program of work includes:

preparation of repertoires for Internal / Horizontal supplement, Vertical Extension, and components of ideographs, document the rules for unification for horizontal supplementation (pDAM-8), and provide the final code table when requested by WG 2. It was emphasized that providing the final code table is only upon request from WG 2.

Mr. Mike Ksar: Suggested amending the Scope to read: "To prepare proposed repertoires and related text for ..."

Disposition: Accept IRG's scope and program of work in document N 1261, with the Scope reworded as suggested. See relevant resolution M29.9 at the end of this section.

C. Components of Ideographs:

Presentation:

Mr. Zhang Zhoucai: In addition to C1 and C2 of document N 1261, document N 1280 was also referenced. Each ideograph can be regarded as a combination of components. All IRG members agreed to include components in the BMP of the standard. These should be separate from Vertical Extension repertoire. Some components are already included in CJK area in 10646-1: 1993. Additional components should be a separate set. IRG is in the process of collecting all the input on the topic. A merged additional components has been attached to the status report in document N 1280.

Discussion:

Professor Kohji Shibano: What is the definition of Component? What is the definition of Common Usage?

Mr. Michael Kung: What is the definition of Component? Is there a list of fully composed list or methods for composition related document?

Mr. Zhang Zhoucai: These are taken from dictionaries, from Education ministry documents etc. There are references in IRG documents. Document N 1280 is only a status report. IRG documents can be requested for more information from the IRG Rapporteur.

D. IRG Meeting Schedule:

Meeting No. 6 is scheduled for 1996-2-5--9, San Francisco Bay area; host Apple Computer

Meeting No. 7 is likely in Hong Kong, May 1997

Meeting No. 8 is likely in Singapore - time ??

See relevant resolution M29.9 at the end of this section.

E. Issues - to be addressed by WG 2:

a. Ideographic Structure Symbols:

Mr. Zhang Zhoucai: Requirements concerning ideographic structure symbols - should it be in IRG, WG2 or WG 3? Mr. Shigenabou Kato, Japan, says that these should be coded in 6429 - promised a contribution in 15 days - not received yet. He will be contacting Mr. Willy Bohn, editor of ISO/IEC 6429. The sentiment in IRG is that 'structured symbols' is not the business of IRG.

Mr. Mike Ksar: They should be in 10646 and not in 6429.

b. Radicals proposed by TCA:

Mr. Zhang Zhoucai: TCS had proposed 214 radicals, and 300 digits (document N 1182). These are coded in CNS. IRG reviewed these, and the result is that IRG has no objections to WG2 considering them for the standard - based on rule of compatibility criterion. Though they are duplicated coding they are candidates for encoding in the Compatibility Zone.

Professor Kohji Shibano: The rule was for round trip with all the standards prior to first publication of 10646.

Mr. Michael Kung: Symbols and components seem to be related closely to composition.

Dr. Asmus Freytag: The UTC is certainly interested on the topic. Document N 1281R related information is requested at some appropriate time. A composition method should be part of 10646 and not a part of another standard that could be used with the standard. They should not be treated as another set of control characters.

Mr. Mike Ksar: The proposal is to keep the work on composition in WG 2. An ad hoc group consisting of interested parties such as Unicode, IRG (for the ideographic repertoires), other interested NBs or liaison organizations should deal with the topic in WG 2.

Disposition: Structured symbols and ideograph components will be part of the composition related discussion, to be dealt with in WG 2. IRG is welcome to have discussion and prepare input to be discussed at the WG 2 level. See relevant resolution M29.9 below.

Action Item: NBs and Unicode consortium are requested to review document N 1182.

Relevant Resolution:

M29.9 (IRG): Unanimous

With reference to documents N 1261 and N 1280, WG 2:

appreciates the progress made by the IRG

has received and discussed the proposed prioritization of the vertical extension repertoire

invites the IRG to prepare a new document reflecting the complete repertoire, showing clearly the different prioritization levels and send it to the convener for WG 2 distribution and feedback

endorses the mission and the program of work of IRG, documented in N 1261 section M1, with the following reworded scope:
"To prepare proposed repertoire(s) and related text(s) for standards on
ideograph-related matters and submit them to ISO/IEC JTC1/SC2/WG2."

endorses its next scheduled meeting in the San Francisco area, on 1996-02-05--09.

confirms that the subject matter of 'Ideographic Structured Symbols' is to be dealt within
WG 2, in conjunction with the topic of "components of ideographs", with input from IRG.

11. Defect reports status

Input Documents N 1144 Additional Arabic letter and symbols - proposal summary form for N 879 and N 983; Hugh McG Ross; U.K., individual contribution; N 1213 Response to document N 1144 on Arabic Script; Alaa Ghoneim, IBM, Egypt; 1995-04-07 N 1214 Unsatisfied Defect Reports from UK; Hugh Ross and Bruce Paterson, personal contribution; 1995-05-27 N 1228 Defect Report Index; Arnold Winkler; 1995-06-26

Document N 1213 is the response for some of the defects reported in document N 1144. The WG 2 data base (document N 1241) has captured the information. The editor should use document N 1213 -- in preparing a response to defects in document N 1144. See resolution M 29.1.

No new standing documents will be created for defect reports - the data base maintained by Denmark on behalf of WG 2 must be referenced.

Relevant Resolution:

M29.1 (additional information on characters): Unanimous

With reference to document N 1144 from UK, and response in document N 1213, WG 2 instructs its editor to prepare editorial corrigendum text reflecting the following:

Add the following entries in the annex "Additional Information on Characters":

00AB LEFT-POINTING DOUBLE ANGLE QUOTATION MARK
This character could alternatively be used as: Arabic opening quotation mark with a variant glyph
00BB RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK
This character could alternatively be used as: Arabic closing quotation mark with a variant glyph

12. Liaison reports

12.1 Unicode Consortium

Input Documents N 1276 Unicode Consortium Liaison Report; Unicode Consortium; 1995-11-06

Presentation:

Dr. Asmus Freytag introduced the document N 1276, containing the liaison report from the Unicode consortium to WG 2. The Unicode Consortium is in the process of editing the next revision of the Unicode standard. In this process, we have discovered some repertoire items that have been approved in 10646 but not in the Unicode consortium. There is a risk of being out of synch. The Unicode consortium officers met on how to improve the process. Unicode to WG 2 liaison seems to be working. The consortium would like to see in WG 2 procedures, appropriate steps to submit appropriate documents to Unicode to enable it to synchronize its reviews.

While the consortium is aware through the liaison to WG 2, the Unicode consortium is not a liaison to SC 2 and since a large number of documents are not necessarily near approval etc. Unicode consortium is requesting WG2 to inform UTC explicitly about appropriate documents at the appropriate stages of processing for review and feedback.

WG 2 members are welcome to participate in the Unicode consortium's conferences. The September 1995 conference was a success. About 10 percent of the 400 audiences were from outside the US. The number and scope of implementations described is quite large.

The web site: www.unicode.org -- has the status reports, call for future papers etc.

Ethiopic is another item that is being worked on in UTC.

Discussion:

Mr. Mike Ksar: The procedure requested is not different from what is used with other liaison organizations, such as between CEN and JTC1 etc. We should strive to strengthen the relationships between Unicode consortium and WG 2 in the interests of the standards being used by members of Unicode. The documents are distributed by Mr. Mike Ksar as and when they are received by the convener.

The processes that we have in place are adequate to distribute the information. Convener and liaison member to Unicode are to get together and examine ways to strengthen the liaison and harmonization.

Action Item: The convener and Dr. Asmus Freytag will get together and discuss ways to improve the WG 2 to Unicode consortium feedback and synchronization channel.

12.2 AFII

There was no written report. Mr. Mike Ksar informed WG 2 that AFII is willing to participate in the publication of the next edition of 10646.

12.3 SC 22/WG 3 & SC22/WG 20

Input Documents N 1266 Unique identifiers for characters in 10646; Liaison SC22 to SC2; 1995-09-22

This document from SC 22 to SC 2/WG 2 contains a list of requirements an identifier should satisfy. See discussion under 8.10 on Unique Identifiers.

12.4 ITU

There was no liaison report at this meeting from ITU.

13. Other business

13.1 Second edition of 10646

Mr. Bruce Paterson, the editor, will make a proposal for publishing the next edition of the standard. There are several open issues related to Fonts, tools to create / convert etc.

Discussion:

Dr. Asmus Freytag: Do we have a consensus on what goes into the next edition.?

Mr. Mike Ksar: AM-5, the minor editorial corrigenda that have been agreed on, and all the addenda that have been agreed to. Only the pages that need changes may need creation. ITTF may republish the whole thing. We can freeze the candidates for inclusion into the next edition.

Mr. Bruce Paterson: Amendments 1 and 2 will involve changes to the front material as well as new annexes. The defects dispositions will change some part of the front text, and additional annexes, plus a few pages in the code table. The front matter and the end material may have to be reprinted / republished. How much of the intermediate matter - code tables - should be reproduced is the open question. The Unified Han with the 5 columns will be ready at the best in another 18 months for publication, but not earlier. They may not go into the second edition.

13.2 Status of work

Input Documents N 1241 SC2/WG2 Summary of work items; Sven Thygesen; 1995-06-27

Mr. Sven Thygesen will update the document for review at the next meeting, reflecting the resolutions and dispositions to date.

13.3 Issues list - Japan

Input Documents N 1278 WG2 Issues List; Takayuki K. Sato; 1995-10-30

Presentation:

Mr. Takayuki Sato introduced the document. The list is an initial list prepared by Mr. Takayuki Sato. NBs are requested to review the list, take specific disposition actions for consideration at the Copenhagen meeting. The list contains things that WG 2 undertook to do - but have not been followed through.

Discussion:

Dr. Asmus Freytag: In what way is this different from Action Items list which is included in the meeting minutes? We do not seem to have a method of tracking multi-step action items. Currently we have a means of reviewing what gets posted to Action Items list.

Mr. Sven Thygesen: The tools available at our disposal -- the minutes, the list of action items, the cumulative lists of things, the data base in Denmark -- are all there. It would be nice to see how they all interact.

Dr. UMAmaheswaran: The project tracking and status reporting document - database maintained by Denmark on behalf of WG 2 -- may be used for the purpose.

Action Item: All NBs are requested to review the list of issues (document N 1278) prepared by Japan, and comment on whether there are more items to be identified.

13.4 Information on GBK Standard - China

Input Documents N 1281 GBK is being finalized; China; 1995-11-07

Mr. Zhang Zhoucai introduced document N 1281 as an information document. It contains the current plan for an extension (K in the GBK) of an existing GB standard to include more characters. It is a code supported by the industry. It is different from UCS,. The long term direction of China to use 10646 has not changed. The GBK will eventually become included as a part of GB 13000 referenced for transcoding to and from UCS.

13.5 Future meetings.

Input Document:

N 1283 Invitation to host WG2 Meeting 30 in Copenhagen; Denmark; 1995-11-08

The meeting planned to be held in San Francisco Bay area for the week of 29 Jan 1996, was CANCELED (not enough time between meetings). Meeting 30 will take place in Copenhagen, Denmark, from Monday, 1996-04-22 through Friday, 1996-04-26.

Mr. Sven Thygesen welcomed WG 2 to Copenhagen, and introduced the document N 1283 containing information on the meeting 30. Copenhagen has been declared the cultural city of Europe for 1996.

Canada is to confirm the Meeting 31 arrangements (August 12--16, 1996; followed by WG 3 and SC 2 plenary meetings the next week) to the convener by end of November 1995. UK and Cyprus are backup locations for the August 1996 meeting.

Relevant Resolution:

M29.11 (Future Meetings): Unanimous

WG2 confirms the following future meeting schedule:

Meeting planned for January 29 to 2 February 1996 in the San Francisco Bay Area is canceled.

Meeting no. 30: 22 to 26 April 1996, in Denmark (see document N 1283)

Meeting no. 31: 12 to 16 August 1996, in Canada (the week prior to WG 3 and SC 2 meetings) (backup UK, Cyprus)

Meeting no. 32: 20 to 24 January 1997, in Singapore (backup Cyprus)

Meeting no. 33: 23 to 27 June 1997, in Cyprus (no backup)

China is one of the future host countries.

14. Closing

14.1 Approval of resolutions

Output Document:

N 1304 Approved resolutions of meeting 29 in Tokyo, Japan; WG2; 1995-11-10

Of the eleven (11) national bodies who were present during the meeting deliberations, nine (9) were present when resolutions were discussed, in addition to the liaison representative from the Unicode Consortium. The final set of resolutions is documented in document N 1304 (and quoted at the appropriate places in this document).

14.2 Appreciation

The following resolution on appreciation were approved by acclamation:

M29.12 (Appreciation): By Acclamation

WG2 thanks JISC, IPSJ/ITSCJ and its staff for hosting the meeting, providing secretarial and administrative support and especially for its outstanding hospitality.

Thanks were also expressed for the dinner on Tuesday night hosted by ITSCJ.

14.3 Adjournment

The meeting adjourned at 11:50h on Friday, the 10th of November 1995.

15. Action Items

All individuals, national bodies and liaison organizations are requested to take prompt necessary steps to complete the action items identified against them in the following tables to avoid undue delays in WG 2's progression of work.

NOTE: If you have any comments or corrections to be made to the above minutes, please inform the convener and the meeting secretary, well ahead of the next WG 2 meeting. This will assist in speedier review and acceptance of these minutes at the next meeting.

Action Items of meeting 29


This page is maintained by Keld Simonsen (keld@dkuug.dk) using HTML Author. Last modified on 96-02-26.