DOC TYPE: Meeting Minutes

TITLE: Unconfirmed Minutes of WG 2 meeting 35, London, U.K.; 1998-09-21--25

SOURCE: V.S. Umamaheswaran, Recording Secretary, and Mike Ksar, Convener

PROJECT: JTC 1.02.18 – ISO/IEC 10646

ACTION ID: ACT

DUE DATE: 1999-02-15

DISTRIBUTION: SC 2/WG 2 members and Liaison organizations

MEDIUM: Paper

NO. OF PAGES: 64 (including cover sheet)

The convener Mr. Mike Ksar opened the meeting at 9:10h. He welcomed the delegates to the London meeting. There are about 150 documents to go through during the four days of the meeting. We have a full agenda. Delegates were requested to cooperate to be able to finish the business in the four days. Delegates are to give any documents for distribution to the convener directly. Delegations who did not advise the SC 2 secretariat are reminded to send the confirmation of attendance to the secretariat in advance of future WG 2 meetings. He introduced Mr. Bruce Paterson representing the host, the British Standards Institution (BSI).

The following 41 delegates representing 14 national bodies and 3 liaison organizations, including 1 observer, attended the meeting.

The convener went trough the proposed agenda. There are too many agenda items and about 150 documents to cover during the four days. All items marked FYI - we will deal with in a quick manner. Any outstanding questions on these can be picked up later in the week. Several documents are being copied. If new documents are proposed and are not copied in time when a specific agenda item is discussed, we will need to postpone the discussion on that document for later in this meeting or to a future meeting.

One copy of samples of code tables for the next edition of 10646-1 is available with Dr. Asmus Freytag (prepared by AFII). Document N1810 has only four sample pages. The full table includes Hangul but does not include Han. Delegates can contact Dr. Asmus Freytag to review and feedback on the sample code tables.

Ad hoc groups on Myanmar script (Burmese); Mongolian; and Annex N, are planned. Others were formed during the discussion of different agenda items as needed.

The modified agenda was accepted and is reflected in the table of contents of this document you are reading. Some rearrangement of the topics has been done while preparing these minutes. All the discussions have been captured under the appropriate subject titles.

Dr. V.S. Umamaheswaran introduced document N1703 containing the minutes of WG 2 meeting 34.

The minutes were adopted with the following changes (some of the changes were pointed to the convener and the recording secretary prior to the meeting, a few during the meeting, and the rest were found by the recording secretary). Editorial corrections were received from Messrs. Bruce Paterson, Johan van Wingen, Michael Everson, and Ken Whistler.

A planning meeting of Cultural Adaptability Workshop (CAW) of JTC 1 is scheduled for the first week of October in Philadelphia. JTC 1 ad hoc meeting on Cultural Adaptability is scheduled for December 98 addressing the work of JTC 1 SC 2, SC 22 WG 20 and SC 35 (just created). Mr. Mike Ksar will represent WG 2. The SC 2 chair - Professor Shibano, and the WG3 convener - Mr. Evangelos Melagrakis, may also attend.

Action item: Mr. Mike Ksar to contact ITTF (Keith Brannon) about the Error in the Amendment numbers published on the web site.

All of these documents have been published. There was no discussion.

SC 2 program of work is available from the web site of SC 2. http:\\dkuug.dk\JTC1\SC2 (check the web site).

Technical Corrigendum No. 2 (TCOR 2) has been published. There are some typographical errors in TCOR.2, which the editor will correct with ITTF.

Mr. Takayuki Sato: Japan has found a problem with the project on 10646 part 2. It was sub-divided for extension of the repertoire. Japan agrees with the extension of the repertoire. The scope does not include provisions for non-graphic characters. Japan feels therefore that there should be a NP ballot instead of a sub-division of work. The NP should include an explanation of the nature of the non-graphic characters. There is no definition of non-graphic characters (see document N1717) in the current definitions of 10646.

Action item: Messrs. Michel Suignard, Bruce Paterson and Takayuki Sato -- to address the concerns raised by Japan in document N1819 and come up with suggested clarification / explanatory text.

Mr. Mike Ksar: Document N1822 (SC 2 N3144) is a contribution sent to JTC 1 from the Netherlands national body on the functioning of SC 2 (also JTC 1 N5499). It is for information to WG 2. I would encourage delegates to review and send their comments to SC 2.

An ad hoc group met and prepared a draft response from WG 2 to SC 2, in document N1879 – see the next section for more details.

Mr. Erkki Kolehmainen presented the ad hoc group's draft report. The report addresses comments aimed at WG 2's work. It points out the universal nature of UCS. There may be rational limitations -- for example, no pre-composed characters. The goal of universal nature (of 10646) is widely accepted by the industry. It is also clear that any national body can have only a limited exposure on several scripts. However, several national bodies participate due to other expertise. The Netherlands has suggested that learned societies should use the Fast Track process - however, this path is not available to these societies. Even if we overcome the fast track process area, the standard becomes highly fragmented instead of a single focal point for assigning code positions in a coordinated manner.

1. Mr. Johan van Wingen: I cannot represent the Netherlands position at this meeting. The national JTC 1 committee took the decision. Some editorial corrections were pointed out. The wording of one of the paragraphs was corrected.
2. Dr. Umamaheswaran: An attachment, which shows the disposition of comments to negative ballot from the Netherlands on Amendment18 should be included.

Disposition: Revised document N1879 was accepted to be forwarded as WG 2 input to SC 2.

Action item: Mr. Mike Ksar to take the revised document N1879 along with the disposition of the Netherlands' negative ballot comments on Amendment 18 and forward to SC 2 secretariat.

Document N1844 contains comments for applications for registration from US to SC 2. It is for information to WG 2. There was no discussion.

TR15285 has been published by ITTF. The referenced documents are for information to WG 2 members. There was no discussion.

Mr. Bruce Paterson: There were 20 approval votes - 6 of these had attached comments, and no disapproval votes. The FPDAM passed unanimously. Each comment was taken and discussed towards a disposition.

1. Dr. Asmus Freytag: The way we are planning to print the standard 10646 (in synch with publishing Unicode 3.0) is to print the main part with resources from AFII and the Ideographic parts with help from IRG. For 10646 proper, we can do with the Camera Ready Copy for those parts of the standard by coordinating with IRG. Unicode 3.0 is not using the 5 col. format and cannot readily use the IRG formats. Unicode 3.0 requires an outline font to do the formatting needed. I have spoken with Mr. Zhang Zhoucai to work with a local font vendor to get a free license for printing Unicode 3.0. It is not acceptable to be able to print the 10646 and NOT the corresponding parts of Unicode 3.0. Also we should have matching fonts to existing sections on Ideographs. The requirement for Unicode is not the same as that of WG 2. We need a single font for a single column for Unicode, whereas five columns are needed for WG 2. Some of the fonts used in printing the 5 columns format are not available even to IRG. If we place an action item on China to provide at least one font for ALL the characters in Extension A then it would be adequate.
2. Mr. Mike Ksar: Can IRG provide not only the CRC, but also a means of generating electronically the same material, and also the fonts needed for Unicode 3.0. The electronic version of the document is needed for ITTF plans for electronic publication of the document.
3. Mr. Zhang Zhoucai: To whom are we providing the fonts -- to WG 2 or to Unicode?
4. Mr. Mike Ksar: The fonts are to be provided to AFII who is assisting in the publication. Fonts for the first edition of 10646 were provided to AFII.
5. Mr. Zhang Zhoucai: We have to discuss this within IRG with the vendors to get the appropriate license. Only China and TCA have provided TrueType fonts. Japan, Vietnam and Korea have provided only Bit Maps. The software that prepares the five-column format mixes the two types of fonts.
6. Dr. Asmus Freytag: I do not have the tools that mix the different types of fonts for printing the single document. The requirement is to get a free license to the Unicode consortium containing a single column for a single font containing all the characters.
7. Mr. Takayuki Sato: IRG should discuss this in particular for WG 2. Unicode and WG 2 requirements are different. The electronic version that is needed by ITTF is a different matter. The first version of the standard was only in hardcopy. The quality of the fonts used by AFII in electronic form is very poor. This needs to be discussed again in IRG with the vendors.

Action item: Chinese national body to make arrangements to get the necessary fonts for printing the Ideographic code tables for the next edition of 10646-1 and for Unicode 3.0, and forward it to the convener (to forward to AFII).

Disposition: Accept the UK Technical comment.

UK editorial comment: Better wording is suggested – to list the CJK ideographs first and then the Extension A, even though the code position orders for these are reversed.

1. Dr. Asmus Freytag: Any deviation from following the code positions order is possibly an impediment to the user of the standard. Going from code position to character should be the preference.
2. Mr. Bruce Paterson: UK has proposed one way - does not matter whether the Extension A comes first or later.
3. Dr. Ken Whistler: Keep the order of code positions and keep the note -- Extension A first and then the CJK.

Disposition: Accept the UK editorial comments.

Disposition: Editor is to prepare new disposition of comments reflecting the above discussion, and send FPDAM-17 for further processing as FDAM text.

Action item: Japan is to provide relevant information to assist the editor with proper references in Annex L of FPDAM-17 text.

Mr. Bruce Paterson: Document N1780 is the summary of replies on FPDAM 18. It was approved with 19 approval votes, and 1 disapproval vote. According to the JTC 1 procedures, the amendment is approved. We will try to accommodate as many comments as possible as long as it does not negate any of the positive votes. The final text will be sent to JTC 1 for FDAM ballot - for a YES or NO answer. Document N1789 contains the draft disposition of comments -- Ireland, the Netherlands (Negative), UK and USA, had comments.

Characters SQUARE WAVE and MONOSTABLE SYMBOL (came from different sources). They need not be separate characters -- the proposal is to keep the MONOSTABLE SYMBOL and remove the SQUARE WAVE SYMBOL.

The Netherlands does not wish to add the S and T with the Cedilla below. WG 2 had accepted to make the distinction from the S and T with Comma Below. We cannot accommodate the Netherlands comments. 19 national bodies have accepted these.

After off-meeting discussion between Messrs. Asmus Freytag, Johan van Wingen, and Bruce Paterson (possibly others) some text for annotations in Annex P was prepared and discussed.

1. Mr. Bruce Paterson: The following is proposed for Annex P annotation -- for each of the S and T comma below characters:
   "The typographical distinction between this character and the character with the cedilla is a matter of fonts. Both characters are found in some languages. … Both the forms are provided. In 8859 part 2 only a single form is provided -- this form is Cedilla Below."
2. Mr. Johan van Wingen: I would prefer – "in a single language, for example - Romanian and Turkish". There is a gentleman's agreement that if things do not take a turn for the worse, the Netherlands vote can be reversed.
3. Mr. Erkki Kolehmainen: If the Netherlands does not reverse its negative vote based on this wordsmithing then it is a waste of time.
4. Mr. Michael Everson: I do not see the point of referencing 8859-2. I would like the Romanian national body to have a look at the disposition of comments. As a matter of courtesy we should send it to them.
5. Mr. Keld Simonsen: I have serious doubts that it will satisfy what we are trying to do. Denmark had voted on a similar note on 8859-2. Our vote may change to negative based on the new wordings. I need to carefully look into the wordings and would like to get confirmation from Romania.
6. Mr. John Clews: I would like to support what the editor has suggested.
7. Mr. Mike Ksar: The proposed text, in my judgement and in the judgement of several other experts here, should be acceptable to Romania. If Denmark or Ireland wants to change their ballots even if Romania agrees, our attempt to getting consensus is wasted.
8. Mr. Erkki Kolehmainen: If there were no consensus I would like to propose that we stick with the earlier wording for the disposition of comments.
9. Dr. Asmus Freytag: I think the words reflect the drafted words produced by Mr. Johan van Wingen and myself. We have added the example languages and are agreeable to it. The Netherlands should not be saying that it may or may not agree. We have cooperated with the Netherlands to resolve its concerns.

Disposition: Mr. Mike Ksar: I believe that the proposed text for disposition of comments satisfies the Netherlands and their ballot would have been considered to have been reversed. The document will go forward for FDAM processing with the above agreed upon annotations to the four characters in Annex P.

Discussion on "Reversed Question Mark":

1. Dr. Ken Whistler: I would like to raise an additional issue -- the position of character 2047 - a character from the graphical representations of control characters 2047 (has the shape of reverse question mark). It has been placed along with a block of punctuation characters and the implication is that it should be moved to control character area.
2. Mr. Michael Everson: Does the US body have a particular view of where it goes.
3. US: In the control pictures area.
4. Dr. Asmus Freytag: We have realized that there is a character having the same shape - the Arabic Question Mark - and this allocation of the character can potentially cause confusion. The implied nature due to a similar looking glyph that could be potentially confused with the Arabic Question Mark --though we did not capture this at the ballot stage, this came to light later.
5. Mr. Bruce Paterson: This comment has not been made earlier and according to the procedures comments from the floor leading to further changes cannot be accommodated.

A defect report was prepared on the character and was discussed during the meeting (another day).

Dr. Asmus Freytag: The defect report proposes that the REVERSED QUESTION MARK should be moved from where we have currently coded at code position 2047 and the suggestion is to move it to 061F.

1. Mr. Bruce Paterson: Per new JTC 1 procedure we do have some freedom and make technical adjustments at the FPDAM ballot resolution time, if we do have consensus.
2. Mr. Michael Everson: I agree we should accept this change.

Disposition: Accept the defect reported in document N1875, and incorporate the proposed change in the disposition of comments for FPDAM 18.

Action item: The convener should contact Romanian national body. The wordings for Annex P annotation will be provided in writing for member bodies to review.

Mr. Bruce Paterson: Document N1841 contains the disposition of comments. There were 20 approval votes and no disapproval votes. There were a few comments -- from Ireland, Israel and the UK.

Procedurally I am not sure if we can accommodate the Irish comments there will be some additional points to consider.

1. Mr. Michael Everson: The ad hoc on Sinhala put together the PDAM text -- the four letters are in the Printed Sinhala standard. The code tables in the standard itself does not show these. Messrs. Glenn Adams and Michael Everson requested that these characters be added in. If there were no serious objections the position was that these four characters should be in the standard. My request is that these be accepted and added to the PDAM.
2. Dr. Asmus Freytag: The Irish representative has verbally given us the rationale for the four characters -- not in the ballot comments. The mechanism could be not to add these characters to the current FPDAM -- but deal with it as a proposal for additional characters to the Sinhala script and deal with them along with other collection of characters to be processed. We need not worry about the font issues in the disposition of comments
3. Mr. John Clews: I would support the proposal.
4. Mr. Michael Everson: The letters with dots should be the ones that go out. 0DF4 - glyph will be changed to more appropriate one.
5. Mr. Hugh Ross: Sri Lanka has not been represented in our committee directly. Sri Lanka has been a member of the Commonwealth -- to that extent the BSI has felt it a responsibility to bring the script forward. We feel glad that this script has been accepted.
6. Mr. Mike Ksar: Disposition of Technical comment 1 from Ireland is to suggest to Ireland to prepare a new proposal for additional characters for Sinhala.

Action item: WG 2 to consider the four characters in the ballot response from Ireland, as request for four more characters to be added to the Sinhala script.

1. Mr. Bruce Paterson: JTC 1 procedure permits small technical adjustments and if there is consensus at SC 2 level changes can be made at the FPDAM ballot resolution stage. There was a general discussion on understandings of the provisions in JTC 1. We do have a little more leeway in this regard than what we thought we had yesterday.
2. Mr. Michael Everson: Would not like the four character proposals to be the test case in case these end up in holding back the FDAM.
3. Mr. Mike Ksar: I would like to have a consensus from the national body representatives on these four characters being added.
4. Mr. Johan van Wingen: If there is a chance to vote on it - it should be OK. I understand that some few characters may be introduced using this process into the FDAM stage.

Disposition: Accept modified disposition of comments for FPDAM-21 on Sinhala - document N1841R. Accept the four additional characters requested by Ireland for inclusion in the DAM text.

Mr. Bruce Paterson: Document N1766 contains the ballot responses. The amendment is approved with 2 disapproval votes - Germany and USA; and 19 approval votes, with comments from Ireland and UK.

1. Mr. Alain La Bonté: Germany's position is on the wrong premises. The symbol is an internationally registered symbol with some proportions to be maintained.
2. Dr. Asmus Freytag: The code we are adding is to add a character from another standard. The approved pictorial reality also should be recognized if possible to avoid the confusion from the user community. If we accommodate the German request, the intended community of users who are used to their existing shape can get confused. It is valid comment made by Canada. We should not try to improve the shape. I would like Germany to consider the impact on the user community and request them to withdraw the comment.
3. Mr. Michael Everson: We can reduce the size of the glyph, keeping the proportions as in the keyboard symbol registered in ISO. I personally never liked the glyph.
4. Mr. Bruce Paterson: We could make the accommodation by making the size of the glyph smaller, keep the proportions and closer to the base line.
5. Mr. Michel Suignard: US has made several comments -- to the US the shape will not affect our community of users - if Germany can accept the accommodation we have no objection.

Irish comments were accepted in part -- some names have been changed to accommodate the Irish comments.

US comments have been accommodated in different ways.

UK suggestion to delete nine symbols was not accepted.

Disposition: Accept modified disposition of comments. Refer to document N1816, section on PDAM 22 -- the changes are to the wordings and to the glyph.

Relevant resolution: M35.3 (PDAM-22 on Keyboard Symbols)
WG 2 accepts the disposition of comments in document N1816R, and instructs its project editor to prepare the final text of FPDAM-22 with assistance from the contributing editor, and forward these documents to SC 2 secretariat for further processing with unchanged target dates.

Mr. Bruce Paterson: Document N1872 contains the summary of voting and table of replies. There were 15 approval votes and none is against. Unanimously approved. There were comments with 6 of the positive votes.

Canada – Typographical errors were accepted.

Israel and Sweden - Noted. No action needed.

UK: Editorial - accepted.

US: Accepted all.

Disposition: Editor will prepare a disposition of comments, and process the updated document N1797 as FDAM-23.

Mr. Bruce Paterson: Document N1834 has the results of the ballot. The PDAM is approved with 15 approval votes, and no disapproval.

Israel and Sweden comments - no action required.

1. Mr. Michael Everson: Information from the Department of Education of the Maldives is that the ordering is not correct and further information has not been made available.
2. Mr. Mike Ksar: If there is such official communication the convener should be informed as soon as possible, so that some actions can be taken to progress the information flow. Ordering of the characters is outside the scope of the standard. One could take a position that the ordering should be dealt with in Sort standards faster. At this moment there is no further information to resolve the concerns regarding the ordering.
3. Mr. John Clews: There are two additional contacts in Maldives (whom I met at a conference in India last month) we may be able to reach -- from the Official languages Group in Maldives.
4. Dr. Ken Whistler: One characteristic of Thaana script is that it is based on use of Arabic numbers. It is quite likely that the Ministry of Education want to promote an order that is based on phonetic or other criterion than the numeric based ordering. What we have here is based on published material and the Ministry may be interested in changing that order. We may not be able to accommodate ordering requirement or different views of the alphabet.
5. Mr. Michael Everson: We had other experts from Maldives who had expressed that they were satisfied with the order. If more information is available during the FPDAM stage we may be able to accommodate the comments.
6. Mr. John Clews: The ordering of these characters should be acceptable. I have seen other characters in earlier drafts - what happened to some other characters?
7. Mr. Michael Everson: They were deleted in the PDAM version that was agreed to.
8. Mr. Mike Ksar: Once we process it as an FPDAM it cannot be changed. If we decide to change the order after a FPDAM ballot, then we may have to go to another FPDAM ballot.
9. Dr. Asmus Freytag: This committee seems to be not so much worried about the correct ordering. We may be able to assign the responsibility to some people before finalizing the next FPDAM draft.
10. Mr. Hugh Ross: Mr. John Clews has indicated that there are published versions of the sorted order in the form currently used in the amendment - it should be sufficient to accept the current order.
11. Mr. Mike Ksar: We can prepare a letter to get the Ministry of Education, Maldives to give us more information -- and if it is made available to us, we could instruct the editor and the contributing editor to accommodate the new ordering of the characters before sending out the FPDAM.
12. Mr. John Clews: Contacted Dr. Waldi (?) Maldivian expert - the sorting order as it is proposed is acceptable. (Will send the report to Messrs. Mike Ksar and Michael Everson). (There was a joint meeting of ordering Urdu in Pakistan - some of the Arabic characters are also used in Urdu. I have suggested to them to submit their input to WG 2. Mr. John Clews was advised not to present himself as SC 2 rep. If there are any contributions they are encouraged to contact the convener. Note: this comment on Urdu is unrelated to the topic of Thaana script.)

1. Mr. Michael Everson: The font is true type - not a felt pen. It was from some student body contact in UK (?) -- not sure if they have a thinner set. The Maldivians use this font.
2. Dr. Asmus Freytag: I will be using available fonts to print the next edition. I do not want to end up in having to do arbitrary changes to the standard. What we should put into the document is to get the typical fonts. We can possibly invite the editor to find a better font.
3. Mr. Michel Suignard: The font used is NOT typical in my opinion. Editor should possibly get a better more typical font.
4. Ms. Joan Aliprand: There is a font used in the Library groups. The font used in that document is less broad-brushed.

The US comment was accepted in principle, a better font would be used if the editor could find it.

Disposition: Accept in principle for further processing as FPDAM - hold till the ordering and font issues can be improved. See revised N1841, section on PDAM 24 on Thaana script.

Action item: Mr. John Clews to do some research to get the contact names, mailing address, fax etc. in the Maldivian Ministry of Education and get the information to the convener.

Relevant resolution: M35.4 (PDAM-24 on Thaana)
WG 2 accepts the disposition of comments in document N1841R, and instructs its project editor to prepare the final disposition of comments and the final text of FPDAM-24 with assistance from the contributing editor, and forward these documents to SC 2 secretariat for further processing with unchanged target dates.

Mr. Bruce Paterson: Document N1835 contains the ballot results. The FPDAM was approved unanimously with 15 approval votes and none against.

Sweden and Israel comments - no actions needed.

1. Dr. Ken Whistler: I have reviewed these names and have no problems with changes proposed by Ireland.
2. Dr. Asmus Freytag: Unicode consortium accepts the proposal (the changed names are in the Unicode database).

Accept all proposed changes from Ireland.

1. Mr. Christopher Fynn: Is the US comment based on the knowledge of the script or based on the Brahmi based script?
2. Dr. Asmus Freytag: The request seems to be for Spacing Characters and can be separated from the previous consonant character. If the character cannot be split from the previous one then it has to be marked "non-spacing". Also, the US comment about font availability has been satisfied.
3. Mr. Michael Everson: We can check the combining or non-combining aspect with experts.
4. Mr. Hugh Ross: Based on the work I have done on Khmer in the past, these characters should not be COMBINING.
5. Dr. Ken Whistler: Based on the documents we have available at this meeting, we can double check and verify before the end of the meeting.

Japan: Mr. Takayuki Sato:

Technical comment A is on the currency sign:

1. Mr. Michael Everson: Referring to document N1179 -- the RIAL SIGN has been derived by drawing a line through the Khmer letter RRO.
2. Mr. Johan van Wingen: Currency sign related comments from Japan - there are precedents -- like in Thai. The BAHT sign is in the Thai block, and is not derived from the Thai script. I think we should keep the sign with the Khmer block.
3. Dr. Asmus Freytag: Keeping the currency sign with the block of Khmer script can assist typographers in keeping the styles consistent. Putting the currency sign in the currency symbols block will help us find it in the same block. There are also precedents by keeping the currency signs with the script. The space available for coding the currency sign is not large enough. If Cambodians ever design a Latin-based script, it will be desirable to keep the Riel sign with the Khmer script. Technical comment A - as to the name of the currency symbol - we can use the consistency argument and keep the KHMER in the name; we accepted the IA versus IE change for transliteration and can accept the change to RIEL from RIAL. As to the comment B.1.-- 17A2 and 17A3 seem to have identical glyphs -- we can verify if it is a duplicate GLYPH error or not. Similarly for the others. Rationale for item B.5 is not expressible in the PDAM document.
4. Mr. Hugh Ross: As to where the currency sign is to be encoded -- primary use of the symbol gives guideline. A currency in use within country typically should be used within the script.
5. Dr. Ken Whistler: Suggest all B.1, .2 and .3 -- the apparent duplications should be checked with Mr. Maurice Bauhahn (Dr. Ken Whistler has his telephone number in London and will get it resolved).

• B.4 - means a standalone form - the conventional way is to combine with a space. Cambodian people seem to state that a standalone form is needed. To be checked.
• 17B6 and 17C6 -- these are combining marks for Combining Characters. Non-spacing standalone character is to be represented by combining with space.
• B.5 - Cambodia has a large number of subscript forms. The model used is to use the Cambodian Virama to create the subscript form.

1. Mr. Takayuki Sato: Cambodian users should be consulted on these to check the model used and the coding used.
2. Mr. Michael Everson: Document N1729 - report on the ad hoc group on Khmer is the basis of the document. A simple mapping table between the model chosen and other existing models can be done.
3. Mr. Maurice Bauhahn: The Khmer proposal has a lot of input from various experts. I have 14 years of personal experience. I went through the Khmer dictionary. Some questions have been raised:

• Same glyph 17xx and 17xx -- first glyph is a consonant and the second one is an independent vowel used in transliteration. First one is used in Khmer and the other one in transliteration. Some of the experts have insisted that both of these characters should be coded.
• 17A4 - transliteration character (for Sanskrit).
  There was hot discussion (among the experts) on the Virama model -- one of the persuasive reasons was that Khmer has been written in several ways. Often subscripts have been used to avoid characters that are considered to be insignificant. The Chung Virama use allows the use of many characters versus the alternative of encoding all possible subscript characters. The Virama model was chosen to economize the coding.
• Technology in Cambodia is at the moment primitive. Only myself have done sorting. It is risky situation for them to go with a technology that they are not familiar with. Main thing for us is to encode all the characters for Khmer - heavily researched and debated -- and the current proposal does meet the goal. As to the currency sign, I agree to change the spelling of RIAL to RIEL. I do not have any particular location as preference.

1. Mr. Takayuki Sato: Based on what I hear from Mr. Maurice Bauhahn, let us go with the new proposal. We need a small note from the governmental organizations that the current proposal is acceptable encoding for them. Why is it so difficult to send some expert in Cambodia a note?
2. Mr. Mike Ksar: WG 2 has the expertise - and usually we do not go to the government organization. We would like to add some explanation to the disposition of comments to address the Japanese comments and possibly change their ballot to positive.

Disposition: Based on the explanation presented by Mr. Maurice Bauhahn (to be captured in the disposition of comments) Japanese questions have been answered and the negative vote will be reversed. Japanese comments were accepted in part: 17DB KHMER CURRENCY SYMBOL RIEL (comment A.1)

US Comment - the forms are always combining and therefore should be shown with a dotted circle. Accept the US comment -- marking positions 17C7 and 17C8 as combining.

Relevant resolution: M35.5 (PDAM-25 on Khmer)
WG 2 accepts the disposition of comments in document N1841R, and instructs its project editor to prepare the final disposition of comments and the final text of FPDAM-25 with assistance from the contributing editor, and forward these documents to SC 2 secretariat for further processing with unchanged target dates.

Mr. Bruce Paterson: Document N1837 contains the ballot response. 14 - 1 - 3 -- was the voting result. There were some comments also. The amendment is approved.

UK - negative - 3 characters should have SYRIAC in their names.

Ireland and US - have similar comments

1. Dr. Ken Whistler: US and Ireland have similar comments - we should accept the UK comments
2. Ms. Joan Aliprand: The UK comments are rigorous - and we should accept that along with the proposed US comment to move them to the Arabic block.
3. Mr. Mike Ksar: We will use ARABIC to these characters and move them to 0653 -- 0655 (from their current 074B --074D).
4. Mr. Hugh Ross: There are characters similar to these in Arabic block. The existing ones are non-combining.
5. Dr. Ken Whistler: These three were added to Syriac block only because they were not in the current Arabic block as combining. Dr. Ken Whistler is to check the correct spelling of names -- consistent use of OBELUS versus OBLLOS etc. for 0D0B --0D0D.

Disposition: All comments have been accepted. The editor is to prepare a revised disposition of comments and text of FDAM-27 for further processing

Relevant resolution: M35.6 (FPDAM-27 on Syriac)
WG 2 accepts the disposition of comments in document N1841R, and instructs its project editor to prepare the final disposition of comments and the final text of DAM-27 with assistance from the contributing editor, and forward these documents to SC 2 secretariat for further processing with unchanged target dates.

Mr. Bruce Paterson: Document N1796 was distributed earlier. It is the text part of 10646-1 together with all editorial corrections, and other changes in the text due to Amendments over the last four years. Several amendments add text to main body of the standard -- in clauses 1 through 26. Annexes A through N, which were in the first edition, and Annex P from TCOR.1. Any changes to the text arising out of additional amendments approved by the Japan meeting (M36) can be included. The document includes all amendments and corrigenda that have been already published by ITTF or are already approved and are in the pipeline for publication. All have been approved at SC 2 and JTC 1 level.

1. Dr. Asmus Freytag: Just to clarify -- it includes all amendments that have been approved to date. The deadline for review of this text is December 1998. Future ones will be deltas to this document.
2. Mr. John Clews: Are there annexes only up to Annex P?
3. Mr. Johan van Wingen: This text primarily reflects new Amendments and new corrections. We did not review the text for clarity etc. Another one that may influence the text is the new TR on CGM. To what extent can we suggest changes towards clearer text of 10646 etc? The concern is that any proposed changes may be considered as out of scope. The matter is of what is the extent of changes that could be requested / suggested.
4. Mr. Bruce Paterson: There are additional annexes Q, R, S, T etc. through different amendments -- these are not included in document N1796. Up to Annex N are from the first edition. The cover page gives more information on what is included and what is not. If the changes are purely typographical - then you can give it to the editor. If it is a significant change that may be considered to be technical or change of concepts, a working document should be produced discussed and then could be accepted.
5. Mr. Mike Ksar: We want this document to be reviewed by December 98. In the future we will be reviewing only the changes to this document. We had published a previous version of this document (document N1396). The Netherlands had time to review this text. There is a freeze date for change requests. The window is very narrow. If there are any changes that could be technical changes then we should do quickly.
6. Dr. Asmus Freytag: I invite Mr. Johan van Wingen to have an off-line discussion to understand the nature of his concerns etc. (After an off-meeting discussion) I agree personally with Mr. Johan van Wingen - an enhanced suggested wording was proposed to satisfy the concerns of Mr. Johan van Wingen.

Action item: All national bodies and liaison organizations to review and inform Mr. Bruce Paterson of any errors or corrections on document N1796 before the next meeting -- deadline is end of December 1998.

For information and use by WG 2 members. There was no discussion.

After some discussion the following responsibilities were identified.

• Action item on font providers national bodies /experts should be completed in time
• Han code charts from IRG
• the name list database is the responsibility of US/Unicode
• Printed version of the non-Han is AFII
• The editors will provide textual content
• Editor to do the final merging of all the pieces.

This only addresses the paper version – not the electronic version.

Mr. Michel Suignard: Presented the US recommendations in document N1839. Recently a number of character registrations have been sent for review in SC 2. We discovered a number of errors in the registrations that were sent for review. Registrations should keep the characters as it was in the source publications. Mapping of these characters to 10646 should be subject to review. If the name of the character is the same and the glyph is the same, the mapping decisions may be simple. The registration probably should not be the place for mappings and mapping decisions should be with an oversight committee. In order to make the above happen, it is recommended to resurrect the Registration Advisory Group.

1. Ms. Joan Aliprand: There are registration procedures used by the registration authority. They seem to be setting procedures in place without proper oversight. The registration should reflect the original published standard and the registry is not the place to fix if the original publication is not in the correct format. The US has some general philosophies about no more 7-bit and 8-bit sets. The library needs do have an exception at the present time for using the existing sets. But the library community is moving fast towards the use of 10646 and the need for 8-bits will probably go away. Not everyone can move to 10646 right away.
2. Mr. Bruce Paterson: This paper is aimed at SC 2. ISO 2375 and the procedure for registration are due for review. The procedure document is a booklet published by ECMA - "Procedures adopted by the registration authority". The registry itself is not a SC 2 document. There is no oversight from the SC 2. The document is based on old technology of using pre-computer-typographic technology. OCR-B was the preferred shape.
3. Mr. Johan van Wingen: If applications for registrations are received according to ISO 2375 then it must be accepted. If the US says that we do not need registrations there is no way out. We can adopt a resolution that WG 2 and SC 2 perhaps can recommend that the sponsoring body withdraw the registration.
4. Dr. Umamaheswaran: There was a discussion in WG 3 on helping the registration authority with a set of revised procedures. This set of revised procedures should be requested for review. I believe the RAG should be resurrected. We should endorse the set of recommendations to SC 2.
5. Ms. Joan Aliprand: One of the areas of concern is the mapping to 10646 - and the expertise of WG 2 can be brought to bear to check these in the registrations.
6. Mr. Hugh Ross: The registration authority (RA) should be encouraged to produce an updated version of the practice of the registration authority. It is not just the 2375 standard. The new RA is quite open to the idea to revise their procedures document.
7. Mr. Michael Everson: We used the ECMA procedures in the context of TC 46 related registration. We should recommend to SC 2 that the US contribution is accepted and followed. ISO 2375 itself does not specify the details of the registration. The original set of rules was extremely rigid and the previous RA had relaxed the rules. We would like to have the 10646 mapping correct.
8. Mr. Keld Simonsen: We should coordinate the work with the character set in the cultural registry. I do not agree that there should be no more 7-bit and 8-bit registrations etc.
9. Mr. Takayuki Sato: There are some Japanese experts acting as informal RAG to the current registration authority. I would like clarification on the 2nd recommendation -- will not the existing review by SC 2 be sufficient.
10. Mr. Mike Ksar: Thanks to the US national body for bringing this matter to WG 2's attention -- apparently there is an agreement among several of the delegates that this is a good idea.

Disposition: Accept the document N1839 - forward it as recommendations to SC 2 for endorsement and forwarding to the Registration Authority for consideration and adoption.

Action item: Convener to forward document N1839 to SC 2.

Dr. Umamaheswaran: Introduced the document UTF-8-EBCDIC. Outlined the basic problem and the solutions. The transformation is similar to UTF-8 in the standard and provides similar capabilities for EBCDIC platforms like UTF-8 provides for ASCII platforms. The intent is to make the final version of UTF-8 EBCDIC an informative annex of 10646 and also a part of the Unicode standard.

1. Mr. Hugh Ross: It seems to me that this is an extremely valuable contribution. The amount of commitment to EBCDIC systems is quite large. The detailed contribution should be extremely valuable.
2. Mr. Mike Ksar: My first concern was that it could potentially cause confusion with the existing normative UTF-8. Since the proposal is to make it informative annex, I have fewer concerns.
3. Mr. Alain La Bonté: I do not see how it would work exactly. On the Internet - would it be used?
4. Dr. Umamaheswaran: See section 6 in document N1848. The transformed data is unsuitable for direct use of Internet.

Action item: National bodies and liaison organizations are to review and feedback to Dr. Umamaheswaran on the document N1848.

Dr. Asmus Freytag: Document N1873 contains an analysis of a number of issues regarding the current definitions of zones. Document N1874 is a defect report proposing how to fix the problems identified in document N1873. Since we have the roadmap document we do not need so called A- I- O- and R- zones in the standard. We will retain the Hi-Half and Lo-Half zones in the standard.

1. Dr. Ken Whistler: On behalf of the US we are strongly in favour. It will make it much easier to maintain the standard. It eliminates the confusion on the use of these zones. There are no implementation impacts.
2. Mr. Bruce Paterson: There is no impact on the normative interchange aspects. It is not a normative change to the standard.

Disposition: Accept the defect report in N1874 along with the problem statement in N1873.

Document N1791 consists of a cumulative list of repertoires since the first edition of 10646. Dr. Ken Whistler pointed out the allocation concerns related to CJK Ideographic radicals (marked with a ?? in document N1791). Mr. Bruce Paterson will be updating this document to reflect deliberations at this meeting.

Document N1738 is a contribution from the US in response to action item from meeting M34 in Redmond, WA. This is the outcome from the ad hoc in Redmond meeting on this topic. An explanatory note to Clause 24.1 is proposed.

Action item: Since Korea was not here, the convener is to check with Korea before accepting this.

Dr. Ken Whistler: An ad hoc met Tuesday night. A written report is being prepared and will be distributed at this meeting. We believe that Mongolian script work item can progress to an FPDAM stage at the end of this meeting.

1. Mr. Richard Moore: All the outstanding items on Mongolian were addressed and resolved. Comments in document N1734 (replaced by N1865) are addressed in document N1878. The outstanding issues and their resolutions are listed in document N1878. The document consists of draft text - with a number of items to be corrected and filled before it can be forwarded for use as PDAM text.
2. Dr. Ken Whistler: The suggested coding location in the Mongolian block - 1800 -- 18AF. It departs from the roadmap for the following reason - the proposal has one too many columns than what is available in the proposed roadmap location. The next available location is used, leaving the current roadmap proposed location unassigned.
3. Mr. Chen Zhuang: Item 6 – "TODO SOFT HYPHEN" should be "MONGOLIAN TODO SOFT HYPHEN"; In the code tables it is OK.
4. Dr. Asmus Freytag: I request True Type fonts to cover the characters are provided. He also requested that fonts are required for the variation forms. The acceptability of a font is to be measured only from the acceptability of the printed form of the glyphs in code tables.
5. Mongolia: Fonts are available right now.
6. Mr. Chen Zhuang: PUBLISHER product has the true type fonts for all the basic and variant forms are available in China.
7. Mr. Zhang Zhoucai: We have to ensure that the shapes from the different fonts have to be checked.
8. Dr. Ken Whistler: Ad hoc recommendation felt that the document is ready for an FPDAM.

Action item: China and Mongolia are to send the true type font to the convener for printing the Amendment text and the next edition of the standard.

Disposition: Accept the Ad Hoc report and process as FPDAM.

202F NARROW NO BREAK SPACE

2048 QUESTION EXCLAMATION MARK

2049 EXCLAMATION QUESTION MARK

in the BMP. WG 2 further instructs its editor to:

- create a new sub division proposal, with the following target dates:

WD 1998-09, FPDAM 1998-10, FDAM 1999-04 and AM 1999-07

- prepare registration request and FPDAM text, with assistance from China, Mongolia, and the contributing

editor (for the three additional characters)

forward these to SC 2 secretariat for further processing.

Several experts have discussed this topic on the email among themselves – experts were from China, Ireland, UK, Russia, and US.

1. Mr. Mike Ksar: The working document should be formatted clearly with the label Working Document (not PDAM etc) to remove any connotation that this may be a PDAM. Names are requested to be using Latin alphabets (for the document registry).
2. Mr. Michael Everson: Document N1660 was the original document starting the discussion. It resulted in almost 4MB of email on the topic among the various experts.
3. Mr. Christopher Fynn: The existing set of extensions is needed so that several Tibetan words are not currently representable with the basic set of Tibetan characters in the standard. Point 1 in document N1864, there is disagreement. These are new / corrected glyphs for them (from Amendment 8).
4. Dr. Ken Whistler: We would like to take to PDAM or FPDAM level at WG 2 level - objective is to get a consensus document at this meeting suitable for PDAM. Points 4 and 5 of document N1864 are agreed upon. Consensus at the meeting was that it was a MISTAKE before to show them as combining characters.
5. Mr. Michael Everson: China sent these questions to the ad hoc. Answers were sent back by the ad hoc.
6. Dr. Umamaheswaran: Points 1, 2 and 3 -- it seems that the responses from the ad hoc to Chinese Tibetan experts in the email has not been included in document N1864 and why they do not agree with that response.
7. Mr. Bruce Paterson: The original was BASIC TIBETAN – I suggest that we change this to TIBETAN. I do not want to imply that there are more characters to come on Tibetan. The editor's note should contain a summary of differences instead of the current list of authors.
8. Mr. Christopher Flynn: There may be more characters - rarely used by them -- to be added. On the name of the script, Bhutanese experts -- it also their national script -- would prefer not to call it Tibetan. They do have native names for these.
9. Mr. John Clews: A document from Mr. Christopher Flynn containing an explanation to Chinese on a few characters is given for WG 2 distribution. Document N1880 contains the response to Chinese comments prepared by Mr. Christopher Fynn. The recommendation is to process the extended Tibetan as a PDAM.
10. Mr. Chen Zhuang: Chinese comments are in document N1864. Chinese experts are not here at this meeting. I have to take this back to the Tibetan Language Committee in China. We tried to reach our Tibetan expert - have not been able to reach him yet.
11. Mr. Mike Ksar: China is to look at the input from Mr. Christopher Fynn. The subject is on hold till Tibetan experts from China can be contacted. If further input is not available before the end of this meeting we will process the document as PDAM instead of FPDAM.

- create a new sub division proposal, with the following target dates:

- WD 1998-09, PDAM 1998-10, FPDAM 1999-04, FDAM 1999-08 and AM 1999-12

- prepare registration request and PDAM text, with assistance from China, and the contributing editor

and forward these to SC 2 secretariat for further processing.

Dr. Ken Whistler: There was an ad hoc between Dr. Ken Whistler and Mr. Michael Everson. There is a set of characters that is not controversial. Some others are potentially for further discussion and the ad hoc has agreed to remove these from the proposal.

1. Mr. Mike Ksar: Dr. Ken Whistler and Mr. Michael Everson to prepare a document into these categories.
2. Dr. Ken Whistler: It would be appropriate to start with a process of creating an FPDAM into which all the non-controversial characters can be included. The first candidate into this collection could be the agreed upon from TC 46.
3. Mr. Johan van Wingen: If I had known of the ad hoc was held I would have participated.
4. Mr. Hugh Ross: When I was the editor of 10646 I was making a number of contributions and I referenced many of the TC 46 standards. A number of characters came directly from the Bibliography standards. Several of these standards were only drafts (in 1990s) and the characters from these could not be introduced into 10646. The TC 46 members got tired of participating in standards and cooperating with SC 2. I am quite satisfied to see what Mr. Michael Everson has done in doing sound work and the useful comments made by the US. The sources of standards in 10646 should include the list of Bibliographic standards.
5. Ms. Joan Aliprand: Dr. Ken Whistler has no objection for participation from other experts from the RLG and British Library etc. who are at this meeting. All the published standards from TC 46 are in current 10646. More references may be added if we accept more characters -- as several of the TC 46 standards have been now published.

Dr. Ken Whistler presented document N1884 - the ad hoc report on "Bucket 35" (a number of characters from meeting 35). A whole series of documents were considered -- from TC 46, Cyrillic Sami and others. The characters could only be listed by hand writing. We have 10 Cyrillic Sami characters in the list to be added. A long list of characters is for further study requiring more information. There are two groups of characters to be added in two different parts of the report.

The revised document N1884R is a much cleaner version of document N1884. There are two parts to be discussed. The first part is TC 46 characters and the Cyrillic Sami characters. Description of this set is included here. The second part consists of all other contractions from document N1747. Livonian characters were not discussed at this meeting but we had taken a decision earlier (from document N1322). We have taken care of document N1817 (Nenets character), document N1812 (Swedish accent), document N1838 (character required for binary normalization completion), document N1857 - Tugrik sign, and document N1845 - IPA characters for disturbed speech. The four Sinhalese letters from document N1809R for writing Tamil have been incorporated into the disposition of comments and will be processed as part of the Sinhala amendment.

Document N1886 - sub-space, document N1882 - Interlinear annotation, document N1887 (Three symbols from Dr. Asmus Freytag) -- will also be considered. Another revision of document N1884 will be done. Reference document N1847 for the best summary of a number of documents. Documents N1741, 43, 44, 45, 46, 47, 48, 49 are summarized in 1885.

1. Mr. Bruce Paterson: Have we addressed all the characters requested by TC 46 to us? YES.
2. Mr. Johan van Wingen: I have to check with the Netherlands experts as far as Cyrillic Sami is concerned. I have expressed my concern about this script earlier. LATIN CAPITAL LETTER OU and GREEK CAPITAL LETTER OU - these are ligatures. These are of so little use that there should be a mention about these in Annex P.
3. Mr. Mike Ksar: We need to check off the ad hoc report for the various documents that were considered by the ad hoc. Thanks to this ad hoc group also for the good work and such a fast turn around and produce the revised document N1884. We should look at additional characters also and see if they could be accepted along with the proposals in document N1884. I hear what you (Mr. Johan van Wingen) say - we cannot make Annex P a repository of explanatory text.
4. Mr. Evangelos Melagrakis: There is no formal request or authorization for the so-called TC 46 characters. It should be changed to TC 46-related characters. 03D8--03E1 -- we were not in favour of including these in the standard earlier. These are used in Ancient Greek scripts. The rationale for these is for purely typographic reasons. This should be pointed out. Greek Kai symbol is used mostly as a short hand way of saying end. Do not distinguish Kai with and without Varia. It is always with Varia. GREEK OU - is a ligature - it is used mostly in calligraphy - o and u put together. It should be included as a ligature with small and large forms. It is also found in everyday use but only in calligraphy. The Greek national body has the above feedback -- we have no strong objections but we cannot strongly support inclusion of these characters either.
5. Mr. Mike Ksar: Names of characters to be refined -- LETTER OU changed to LIGATURE OU.
6. Mr. Michael Everson: I will have no objection to LIGATURE OU. The ad hoc rejected KAI SYMBOL with Varia. KAI SYMBOL with VARIA was not accepted as a separate symbol. The Glyph for KAI symbol includes the Varia sign in it. The Finnish national body should be happy that there is place in 10646 for the Cyrillic Sami.
7. Mr. Klaas Ruppel: The Cyrillic Sami script does exist. Do you plan to change LATIN CAPITAL LETTER OU to LATIN CAPITAL LIGATURE OU? -- The answer was NO. It is also used in transliteration of Uralic, and in some Canadian native scripts.
8. Mr. Bruce Paterson: We do have the Greek Stigma etc. called GREEK LETTER STIGMA. The small forms have SMALL LETTER in them.
9. Ms. Joan Aliprand: Latin Capital Letter OU -- the British Library Character has the Romanian Capital Letter OU (?) it is a letter and not a digraph. The small forms of Greek Stigma etc. came from the TC 46-related standards.
10. Mr. Mike Ksar: Mr. Evangelos Melagrakis has pointed out as to the source of these character proposals. Mr. Randy Barry TC 46, SC4 WG 1 was the person who provided the input to this effort. We could use TC 46-derived etc.
11. Mr. Evangelos Melagrakis - we have consistently objected to the presence of coded character standards in TC 46. The Greek national body was not using some of the TC 46 standards because several characters have been included.
12. Ms. Joan Aliprand: The Greek letters came into Unicode 1.0, from TC 46 standards - from the Research Library Group. At the request of the convener of one of the WG-s of SC-s of TC 46 Mr. Michael Everson helped prepare the contribution. Some other members are the Libraries – the RLG and British Library are present here.
13. Mr. Michael Everson: Some of the characters have come also from my own research. Document N1743 gives all the details on the use of the Greek characters. The Latin Capital Letter OU is not a ligature in use. Algonquian, Huron, Wendat languages in Canada use this character. Would like to see Romanian OOK character also to see if the Glyphs can be rationalized.
14. Dr. Asmus Freytag: Should we be going into so much details for some of these?
15. Mr. Michel Suignard: The small forms of Greek seem to be ligatures. The ligatures should be moved into the block of ligatures. The ligature as a name is confusing. I suggest moving these into the ligature block. There are several ligatures with rationales similar to the OU.
16. Mr. Evangelos Melagrakis: I am the secretariat of TC 46 -- and all the SC and WG material is brought to my attention. I have not had discussion in the Greek national body on this subject. If Greece has to respond formally, I need more time.
17. Dr. Ken Whistler: In this particular case of 03F4 and 03F5 - Greek Capital and Small letters OU - we have the option to move these four characters into another group.
    Disposition: ACCEPT to move these under "for further study".
18. Mr. Hugh Ross: I hear you were going to be open minded -- does this mean I can open up the Yoruba characters? (Not in conjunction with this set of characters under discussion).
19. Mr. Bruce Paterson: Are there code positions for Nenets and Swedish accent? Yes - 02EE and 02DF respectively.

Disposition: Accept the set of characters in first part recommended by the Ad Hoc (minus the two Greek OU characters). FPDAM is desired.

There was an action item on Mr. Metra to give some answers to questions for clarification. There is a fax in response from Mr. Metra. Document N1888 contains the response from Mr. Metra.

Disposition: Accept Livonian characters from document N1322 (refer also to document N1888)

Mr. Michel Suignard: Introduced the document N1886 containing a proposal for a single character "CONDITIONAL SPACE" (instead of SOFT SPACE), a contribution from Mr. Maurice Bauhahn. He explained the need for a special kind of space to represent inter-word spaces in languages like Khmer. The Khmer script is using x204F - SOFT SPACE.

Dr. Asmus Freytag: There are common uses of Square Foot, Square Inch and Property Line symbols. I have seen them on signboards. Property line symbols are used in marking boundaries. Document N1887 proposes to add these three symbols.

1. Ms. Joan Aliprand: I have seen a variant of these symbols.
2. Mr. Michael Everson: Would like to see more evidence of use of these characters.

Disposition: Mr. Mike Ksar - We will put these three characters on HOLD till we get more documentation.

Action item: Dr. Asmus Freytag to provide more information on use of these symbols.

Mr. Evangelos Melagrakis: The Drachma Sign is being considered for proposing to be included in 10646. Will be examining the 10646 character set for any missing Greek characters -- likely 3 or 4 combinations in the extended Greek set only if the national body decides that these are valid for inclusion. Issue of Combining Characters --will be examined to ensure that correct correspondence between existing fully combined characters elsewhere and combining characters in the standard would be possible. CD on ISO/IEC 8859-7 will have the Drachma Sign (and the Euro Sign).

- create a new sub division proposal, with the following target dates:

WD 1998-09, FPDAM 1998-10, FDAM 1999-04 and AM 1999-07

- prepare registration request and FPDAM text, with assistance from the contributing editor, and

forward these to SC 2 secretariat for further processing.

Mr. Bruce Paterson: Document N1821 contains the ballot responses on PDAM 14 -- 13 approvals with a few comments and 3 disapproval votes - from China, Ireland and US.

1. Dr. Ken Whistler: I have reviewed document N1814 - and the proposed ordering and names are correct and meets the objections from the US. I have not had the time to review the glyphs.
2. Mr. Mike Ksar: We have input from China - reflecting the national standard. I would like to take document N1814 as feedback on PDAM ballot rather than as a revised proposal for a PDAM.
3. Mr. Hugh Ross: My understanding of the situation is that there are two suggestions for the Yi syllable code tables. One of these is derived from the Chinese 1991 national standard and the other is derived form traditional use of the Yi syllables. Document N1814 - uses the 1991 Chinese national standard order. These two proposals agree on - - the script repertoire, the glyphs and the names. The only difference is in the order. This is a situation we had before -- the national standards have one way and WG 2 experts identifying another way. This has happened before with Sinhala. My suggestion is that we should provide emphasis on the maximum use by Yi people. Proposals put forth by Mr. Michael Everson, US etc. is such a proposal. However, we should do our best to satisfy the national standards in China also. Suggesting a mapping table between the national standard and the code table arrangement that is agreed to in WG 2 could satisfy the need. Such a mapping table is simple to generate and authorized to be correct, and could be done in one day. My suggestion is to resolve the problem in that way.
4. Mr. Mike Ksar: Document N1814 seems to be a consensus of the three parties.
5. Dr. Ken Whistler - N965 (GB xxxxxx) had a specified order and has been followed in the previous working drafts. The order of the characters that went into the PDAM was in that document. US comment is addressing errors within sub-groups within that set. Chinese input corrects the errors and reflects that original order. It addresses all the comments from the US and corrects a few other items. There are implementations using the 7+7 national standard coding system. These implementations have to do some conversion any way to implement the UCS.
6. Mr. Michel Suignard: We do not have any interest in the mapping itself.
7. Mr. Michael Everson: I have been involved in the Yi ad hoc since the San Francisco meeting. The first document I saw was based on the Chinese standards and my request was always to change that order to the way used by the Yi users. There are several -- Burke - 1984 reference and other documents -- which reflect the traditional users. There was an email from China indicating that the preferred order was Burke's. Our position is that the order to be found in the reference materials, dictionary etc. should be used. China has not changed their national standard from before.
8. Mr. Bruce Paterson: Can we get a rationale why document N1814 adopted a different order from the Burke.
9. Mr. Chen Zhuang: Our Yi experts from China are here. Some experts may have no specific preference for any particular order. Chinese national standards may be revised on the basis of 10646 final version. The Chinese standard has errors in it and needs revision any way.
10. Dr. Ken Whistler: The missing question is - are they planning to just fix minor errors only or to revise it to reflect the traditional order.
11. Dr. Umamaheswaran: Does it mean that China does not have an objection to change the order to traditional order?
12. Mr. Chen Zhuang: We will not have objection.
13. Mr. Mike Ksar: Ordering of the standards is outside the encoding arena - though we can accommodate the ordering as much as possible.
14. Dr. Asmus Freytag: Unicode would prefer to have the ordering not to diverge from the national standard ordering. China's willingness to adopt the order that WG 2 comes up with is good news. Due to the size of the repertoire I would request that special precaution needs to be taken that the next draft that is prepared is submitted for verification by other experts before we proceed with next FPDAM ballot submission. The statement that China can accept any order that this committee decides on -- we can proceed -- China should be aware of and be able to stick to that decision made on this item.
15. Mr. Mike Ksar: Due to the number of characters involved, before this document is sent for further processing to SC 2, the document be sent to me and I will circulate to the experts in the Yi area for verification before sending to SC 2. The question that has not been raised so far -- are there any implementations in the Yi community that may be using the existing Chinese national standard that would be impacted if we change the order? If there are implementations that would be impacted then one has to prepare the mapping tables for those implementations.
16. Mr. Chen Zhuang: Not that I am aware of.
17. Dr. Umamaheswaran: There seems to be a way forward -- to use the traditional order instead of the current Chinese standards.
18. Mr. Michel Suignard: I would like to see if the traditional ordered names and code positions (not glyphs) can be made available. It would be useful for us to check it before we can accept.
19. Mr. Mike Ksar: Yi ad hoc group should meet one more time with the revised ordering proposal by Mr. Michael Everson.
20. Dr. Asmus Freytag: The report should contain the final list of names, and should contain the statement that China has no objection to use the order agreed to by WG 2.
21. Dr. Ken Whistler: In view of the Chinese input, there are several questions on the names of the characters -- to be taken up in the ad hoc on Yi script.
22. Mr. Bruce Paterson: If we change the order according to the above - we will satisfy the Irish comment 1.

1. Document N1863 - China's feedback. China's position is that the four radicals proposed by Mr. Michael Everson are considered to be same as some of the existing radicals.
2. Mr. Michael Everson: Ireland believes that this is not acceptable - since we have included some but not all the radicals from the Burke's input.

Disposition: The editor will prepare DOC. Prepare the FPDAM text. There will be no change in schedule. China to assist the editor in cooperation with Mr. Michael Everson.

Action item: National bodies are invited to provide the sources for the scripts from Amendments 10 through 27.

Document N1836 contains the table of replies and summary of responses on PDAM 26 on Burmese script. An ad hoc on Burmese script met on Monday afternoon. Mr. Michael Everson has prepared document N1883. This report consists of point by point disposition of comments. A Myanmar delegate presented document N1883. The script has been grouped into Myanmar character set and Myanmar extended character set.

(Note: My apologies to the Myanmar delegation for not having captured the name of the individual delegate who spoke during the discussion below, wherever I have indicated as "Myanmar delegate" -- Umamaheswaran.)

1. Mr. Bruce Paterson: Why do we need Myanmar Extended and Myanmar Base blocks? Do we understand that the names of ALL the characters and the block names are being proposed to be changed -- to MYANMAR replacing BURMESE?
2. Myanmar delegate: There is a new Myanmar orthography prepared by the Myanmar language commission. The characters that are in the proposal but not in the Myanmar new orthography were separated into a new block. The characters that are currently in the extended block is for characters that are outside the new official list.
3. Mr. Mike Ksar: We do not reserve any specific code positions for any script. What we have done in the past is not to leave any holes within a code table as much as possible. We normally reserve holes for the purpose of future extensions. This document is prepared in the context of helping the editor to prepare the disposition of comments. We try to accommodate all the national body needs, but the user community at large for a given script is what WG 2 considers. We should make sure that the space in the BMP is efficiently used. Empty positions should be reconsidered between the base and extended sets. Looks like the ad hoc has not completed the work as is presented in document N1883. Statements like expectation of space being reserved for future extensions should be deleted from disposition of comments. Needs more work
4. Comment: Characters at 1774 and 174B -- are different. They are swapped in document N1883. Other editorial errors were pointed out to the contributing the editor.
5. Myanmar delegate: There are some more characters where the names are in question.
6. Mr. Michael Everson: Some of the errors in the document are because of editing of the ad hoc report. Some of the names related issues were not discussed.
7. Dr. Asmus Freytag: There seems to be some more outstanding items to be addressed by ad hoc. I would like to separate the Collections and Block names. The direction is to move towards using a single block name for showing the tables and if any specific subset -- like Myanmar basic -- collections should be used. We have similar use of Block names and collections for other scripts. Collections could be discontiguous also (can have holes). There seems to be large amount of consensus in the ad hoc.
8. Dr. Ken Whistler: I agree that this matter needs to be taken back to ad hoc and cannot be resolved in the WG 2 meeting. I need access to a computer so that we can finish the work.
9. Mr. Hugh Ross: There is a special feature of the set that it is designed in the alphabetical order and it is important to the Myanmar people. It is important to have 6 blank cells to respond to the UK comments. I hope you will allow some gaps in the table to be able to accommodate the UK needs.

A revised ad hoc report was prepared (document N1883R) and presented by Dr. Ken Whistler. He explained the organization of the new report. Change the collection number for EXTENDED MYANMAR to 90 (from 89). A single Block Name - MYANMAR and two specific collections are proposed. The ad hoc has met after document N1883R was produced. There is one more technical change -- characters 104C through 104F will be reordered reflecting the agreement in the ad hoc. This change will be documented and provided to the editor in a revised document. Also, on page 7 - change XXXX MUY MUERDO to 266F MUSIC SHARP SIGN. All the experts in the ad hoc have reviewed and reached a consensus at this meeting.

As to the ballot comments:

• Ireland was a member of the ad hoc group. Their comments have been addressed to their satisfaction.
• Japan and UK had negative ballots.

The disposition of comments was reviewed.

• Japan - the disposition of comments addresses their concern and will change the ballot to positive.
• UK - the request for two vowels were rejected.
• Mr. Bruce Paterson: UK's concerns are accommodated - if the additional vowels are accepted in the future, space has been left in the code table to introduce them later . UK changes its vote to positive.

1. Mr. Hugh Ross: This ad hoc has been a splendid example of international cooperation. Our friends from Myanmar had brought a proposal to the meeting. The procedures of the meeting permit such cooperation to be effective.
2. Mr. Mike Ksar: The ad hoc group has certainly reacted quickly to the requests on them and has produced excellent results and documentation on the subject.

Accept the ad hoc report document N1883R with further identified changes (see document N1883R2 that was distributed later at the meeting). Code tables should be sent to the editor as soon as possible -- we prefer to get these documents in the hands of SC 2 by middle October 1998, to be in time to finish all the ballots by March 99. We can process the amendment as an FPDAM. The name of the script will be changed from Burmese script to Myanmar script.

Mr. Michael Everson: As the contributing editor most of these documents are already in electronic form. The FPDAM text can be prepared very quickly.

Mr. Michael Everson: CEN ISSS is working on Multilingual European Subsets (MES) definitions. Three sub repertoires of 10646 are proposed as candidates for inclusion in Annex A of 10646. Draft versions are included - subject to finalization by CEN Workshop Agreement (CWA) in the near future. The request is for 2 fixed and 2 open collections.

1. Dr. Asmus Freytag: The collection seems to be containing several complete collections. Would it be useful to list the code positions or list the full collections only and some delta collection. We need to make a decision. For an implementer it may be difficult to trace if one has to chase down if a character exists within a range or not. Would it be better to show a list of collections rather than cells?
2. Mr. Michael Everson: The model we are following was the document from the ad hoc on collection identifiers.
3. Mr. Johan van Wingen: The current work of CWA is still a draft and cannot be used as a basis for defining sub repertoires. The source of the subset should be more stable.
4. Mr. Mike Ksar: We can put a qualifier that these sub repertoires are subject to finalization of the CWA. Finland and Ireland are the requesters.
5. Mr. Bruce Paterson: Document N1877 contains the principles and procedures. The suggestion from Dr. Asmus Freytag is relevant to that discussion. The intent under 4.2 of "contain .. etc." is to state characters from the collection are in the collections registered. The suggestion is to make the proposals list the collections contained in the form similar to 4.2.
6. Ms. Joan Aliprand: Is this the first time the collections are being proposed on a language or specific use basis.
7. Mr. Mike Ksar: No - we have collections in Annex A.
8. Mr. Johan van Wingen: We have a set of characters in the Netherlands called GBA set which we are intending to identify as a collection in 10646. In one of the documents an example based on my input was used. That should not be misinterpreted to be having any official status.
9. Mr. Bruce Paterson: The collection in 4.1 shows EN 1923 LL8 is identical to the example in document N1877 and looks the same as 6937.
10. Mr. Keld Simonsen: I think it is premature to accept the request in principle at this time. I would suggest postponing the acceptance to the next meeting.
11. Dr. Asmus Freytag: Since the repertoires submitted is not final, we cannot get into Annex A (the next edition). If we postpone accepting this till the next meeting it may not get into the next edition of the standard.
12. Mr. Mike Ksar: Whether the collection is expressed as a grouping of collections or as a set of code position ranges within a row, is for discussion. If there is no agreement among the European national bodies on the stability or the content of these sub repertoires then it should not be entered in the standard.

Disposition: WG 2 notes the contribution. No action is needed.

Mr. Hugh Ross: This contribution is to assist the work of IEC. The set of Electrotechnical symbols is widely scattered in our standard and users will need assistance. The names used in the standard are more general than the names used by the user community. I was attempting to do a mapping. There are several characters from the standard could be mapped ONE of their symbols etc. They look the same. Some queries came up during this work -- identified as notes in the contribution. Other experts are requested to assist. In several cases, the glyphs we are using in the standard, several of these could be improved from the Version 1 of the standard in our next edition. The IEC contact we have is a convener of a SC -- from Munich, Germany.

1. Mr. Bruce Paterson: This is very interesting information. The information is needed by the IEC documentation. We have agreed to put some mappings in the user's standard. We could recommend to IEC that they provide similar mappings in their standards themselves.
2. Dr. Asmus Freytag: The suggestion for Centre Line CL symbol could be useful. Circled C and circled R do have two different forms in 10646. The Unicode Consortium also has the facilities to post the mapping related information and we will be glad to accept such information.
3. Mr. Mike Ksar: We thank Mr. Hugh Ross for his contribution. The Unicode consortium has offered their resources to host such information on their site.

Action item: National bodies and liaison experts of WG 2 are to review and comment to Mr. Hugh Ross, copying the convener and the editor. Mr. Hugh Ross can approach IEC to ask them to include similar mapping information in their standards.

Dr. Ken Whistler: This proposal made by Ireland is for a "single modifier letter double apostrophe". This is used in an orthography that has no digraphs in the script. The ad hoc on the TC 46 reached consensus on this. The rationale is provided in document N1817.

Dr. Ken Whistler: This document is a contribution from Mark Davis, Unicode Consortium. This is a first of a number of documents - which can be included in the set of characters being addressed by the ad hoc on TC 46 characters. The request is for four "binary completion letters" to the BMP -- a and A dot above, and e and E with Cedilla. There are some fully composed forms with these as components in them. Failure to encode them causes technical difficulties in the composition algorithms. The composition algorithms will be widely used in the World Wide Web.

1. Mr. Hugh Ross: I have concerns about the character with a dot above. The combined character with the A WITH DOT ABOVE is a mistake in the standard - we never got around to fixing it with a defect report. It is from Sami Finnish and came into the standard when the quality of typography was poor. The suspicion is A WITH RING ABOVE became to be interpreted as A WITH DOT ABOVE. The dot above is rarely used on top of a vowel. On the other hand Macron is used on top of a vowel to indicate a long vowel. The 27th letter of the Sami Finnish – A WITH RING ABOVE WITH MACRON -- is what was really needed. I have known this about two years ago but not able to confirm this. It originated in SC 2 WG 3. What I am worried about propagating the error. We should get another opinion to see if there should be a defect report instead.
2. Mr. Klaas Ruppel: The A WITH DOT ABOVE is used in Kilden Sami. It is in between two sounds of A. This particular letter has been used -- in a recent dictionary. The dot above is used to indicate a middle vowel.
3. Dr. Asmus Freytag: Independent of the linguistic pro and con for the A WITH DOT ABOVE. The A DOT MACRON does exist in the standard and we cannot eliminate it from the standard. We cannot simply change the name or the glyph. The defect report will therefore be out of the question. Therefore the rationale behind the request still valid. If the A WITH DOT ABOVE exists, it is all the better. The argument that it was an error in the standard is not valid.
4. Mr. Hugh Ross: I am grateful to hear the explanation that the character does exist. I am sure others will also make the error. We can request the Finnish representative to provide some information for A WITH DOT ABOVE. (NOT for the A WITH DOT ABOVE AND MACRON).
5. Mr. Mike Ksar: We do not take characters out from the standard. Finland has explained that the character is in use today. Whether it came from WG 3 etc. cannot be helped and it is history.
6. Mr. Johan van Wingen: If there are concepts that are used in Unicode and not in 10646 we should update the concepts by updating the text of 10646 to keep consistency. I have never seen a vowel with Cedilla.
7. Dr. Umamaheswaran: There is NO need to add any explanatory text -- to just add four characters.
8. Dr. Ken Whistler: I want to emphasize Dr. Umamaheswaran's comment -- this proposal is not asking for any additional terminology. I would suggest Mr. Johan van Wingen look at 1D1C and 1D1E for example.

Action item: Mr. Klaas Ruppel will provide additional explanation of the use of "A with Dot Above" to supplement document N1838.

Dr. Ken Whistler: The consensus was to accept these in the ad hoc group (on Bucket 35).

1. Mr. Johan van Wingen: I have agreed to these characters, however, I still have some misgivings. Why are there no reasonable people who would refrain from inventing these characters in the first place? Should we be accepting all characters from misguided linguists? There is new orthography in Sami -- which includes stress marks. We have taken a position that we should not be encoding characters for stress marks only.
2. Ms. Joan Aliprand: US were skeptical about the uppercase forms. We have evidence from Mr. Klaas Ruppel that we are satisfied.

This contribution was discussed in the ad hoc group on Bucket 35. The ad hoc's recommendation is --

Cross Accent should be added to the Spacing Modifier Letters block -at U+02DF.

There was no discussion. Carried forward to next meeting.

Action item: National bodies to review and feedback on documents N1755 on Philippines repertoire.

Mr. Michael Everson: Document N1686 was the original document. There is an active discussion of the subject in two different interest groups. They have settled on an 8-bit standard -- base and ligatures. They have indicated that ligatures themselves should go into 10646 - I have discouraged them. There is also debate on whether these should be RTL or LTR. Modern computer users think that it should be LTR but there is another group for RTL. They are awaiting the recommendation from WG 2. I recommended to them RTL - traditional way. Use of Directional Overrides is known. There are a large number of Internet sites using this script -- just like Ogham and Runic. The script will require 3 columns and there is room in the RTL area.

1. Mr. Bruce Paterson: In 10646 there is no implication as to the directionality of characters. We may want to place these characters in some preferred areas.
2. Dr. Asmus Freytag: While 10646 may be silent on the properties of characters, in the interest of synchronization between Unicode and 10646 we need to address the directionality of character. The associated issue is Symmetric Swapping - this is dealt with in 10646. We may be doing the scholarly community and to the user community a disservice if we require BiDi.
3. Mr. Klaas Ruppel: The glyphs shown are for RTL -- and if someone reverses the direction, then the glyphs will need mirroring. There are several, which are bilaterally symmetrical.
4. Dr. Ken Whistler: The issue of mirrored glyphs for direction switch is not generally addressed by WG 2. We have addressed some characters in particular - but entire scripts. Mirroring is an issue in general for Runic, Greek etc. It is also as a general issue. If you want to see these generally available on the web a LTR choice would be preferred -- since BiDi support is not going to be available early.
5. Mr. Michel Suignard: By choosing the RTL as the default - we are not helping the implementers - since this requires BiDi support. The user community has decided that the traditional order is required. If the entire script is in one direction - not mixing directions.
6. Mr. Mike Ksar: There is no specification in the standard to specify that the glyphs will be mirrored. Do we need an immediate answer to this -- do we have time to get feedback from the user community on the directionality.
7. Dr. Asmus Freytag: We have Ogham and Runic in a similar situation. We took the decision of LTR even if they are written RTL. For historic scripts we should take the position that it should be LTR - to get the scholarly community go be able to use it early rather than having to await BiDi. Only if there is strong evidence that the RTL is really needed, then we can do so.
8. Mr. Bruce Paterson: We have gone with LTR for Ogham and Runic - we should follow this model.
9. Mr. Michael Everson: What Dr. Asmus Freytag says make sense.
10. Dr. Ken Whistler: Given the level of controversy on this item - I propose that a revised proposal reflecting the LTR version, containing also more information on how unification related items, and an indication as to how the final set of characters was arrived at.

Action item: Mr. Michael Everson is invited to take the comments and prepare a revised contribution. The document is not ready for processing at this time.

Dr. Ken Whistler: The number of IPA characters from document N1845 has been pared down by the ad hoc committee- by taking into account unification, and other ways to represent them. 5 IPA letters, 2 modifier letters and 11 combining marks. 0362 is a character with a mark that extends over to the next character.

No discussion – carried forward to next meeting.

Action item: National bodies to review and feedback on documents N1740 and N1807 on Hebrew Tetragrammaton.

No discussion – carried forward to next meeting.

Action item: National bodies to review and feedback on documents N1846 on Ethiopic extensions.

Mr. Takayuki Sato: Phillipino language requires n and g together as a single letter. Sequence of characters "ng" is required.

1. Mr. Michael Everson: Why can't n and g be used?
2. Ms. Joan Aliprand: In the Library area we see a character n, g with a tilde. There is possibility to use the current mechanisms in place.
3. Mr. Evangelos Melagrakis: There is transliteration of characters n, g with marks on it. We do have the transliterated character without the original character ng.
4. Dr. Ken Whistler: I would suggest - the encoding of this character should be contrasted with the need for Ch and LL in Spanish -- as input from Philippines.

Action item: Mr. Takayuki Sato to provide more information on the usage of and the need for "ng" in light of the above discussion.

Mr. Takayuki Sato: Document N1861 contains the review comments from Japan on document N1727. In the context of Ruby the solution in document N1727 should be OK. For the general case of inter-linear annotation the proposal is not adequate. Japan is of the view that we have to make a choice of narrowing the scope to Ruby only or expand the scope for general interlinear annotation. Items from document N1882 were pointed out. document N1882 is an updated document N1727 addressing some of the Japanese comments.

1. Mr. Michel Suignard: The narrow and wide terminology is possibly confusing. These are intended primarily for processing purposes only. The kind of objects could be generic or narrow. There is no need to place any limitation. We need lot of additional information to do rendering properly. It is no different from Object Replacement Character. We need anchors to attach the interlinear object. Document N1882 is the rationale behind why we need these interlinear annotation characters.
2. Mr. Takayuki Sato: We may place restrictions on what we mean by interlinear object. SC18 group seems to indicate that there are three kinds of interlinear annotations and these controls are applicable only to one of these. As Mr. Michel Suignard explained, we need more information to present these annotations correctly. These annotation characters alone is not sufficient.
3. Dr. Asmus Freytag: In the context of 10646 we only need to have information described under Proposed Characters and Proposed Glyphs. We could possibly add some text in the standard to explain the use of these characters. The Object Replacement Character(OBJ) is another such anchor in the standard. If we have the precedence of not describing the use of OBJ we have the option of leaving the text out. These characters are not all that meaningful in the context of plain text -- needs additional support from implementations. There is no question that these characters are needed and are in use except that they are using the wrong character codes.
4. Mr. Takayuki Sato: The Ideographic Combining Mark - due to the potential misunderstanding we did add explanatory text on usage. As a minimum, I would like to see some notes (for example in Annex P) on the use of these -- such as in the Notes on page 3 of document N1882.
5. Mr. Michel Suignard: We can add some annotations to the characters in Annex P.
6. Dr. Asmus Freytag: We seem to have consensus to add short annotations in Annex P.
7. Mr. Mike Ksar: We cannot use "the OBJ having no usage notes" as a rationale. Mr. Michel Suignard can write some text for Annex P to satisfy Japanese comments. In the past we did not write long annotations in Annex P.

1. Mr. Takayuki Sato: The program of work problem gets resolved by removing the words - non-graphic character i.e. Where the Tagging characters are described as non-graphic characters. There is one document indicating how the language tags will be used. I would like to see the difference between language tagging, tagging character etc. Need clarification on the use of these characters.
2. Dr. Umamaheswaran: Suggested the use of the term "Special Purpose Use".
3. Dr. Ken Whistler: The term Special Purpose Use could be used. In the BMP there are already several characters of different kinds in the BMP. Some of these are from standards, which have taken them from other standards. Alphabetic, Syllabic - go to Plane 1; Ideographics go to Plane 2 and the General Purpose Plane is used to put the other characters which do not have a visual representation.
4. Mr. Bruce Paterson: The current example of the Tag Scripts is ASCII based. Later we may have a need to provide Tag Script in other languages like Greek. There can also be other kinds of tags -- for example, Colour of the Text.
5. Dr. Ken Whistler: The tag-alphabet is not the only kind of characters that can go into Plane 14. It is meant to be for general-purpose use.
6. Mr. Takayuki Sato: My request is to write down the purpose of these planes. If there are similar characters in Plane 0 why do we need to go for Plane 14.
7. Dr. Ken Whistler: The justification is simply - Plane 1 is allocated to different kinds of scripts in our roadmap.
8. Mr. Bruce Paterson: We could define the characteristics of Plane 14 as characters that need not appear as graphic characters -- they can be rendered visibly if you choose.
9. Dr. Ken Whistler: They are like the BiDi controls where a glyph is attached to make them visible where there is a need.

Action item: Editor of Part 2 to take note of the comments in the preparation of the next Working Draft of part 2.

Additional information and relationship with Mongolian proposal in the BMP should also be considered.

Action item: Mr. Takayuki Sato to provide additional information.

Mr. Michael Everson introduced the Blissymbolics proposal in document N1866. He gave an introduction, background and the methodology of encoding. Technical issue - there are arrows / pointers used in different locations, different orientations etc. indicating different words. Potential encoding problems in a general-purpose way.

Mr. Zhang Zhoucai, the IRG rapporteur, presented various IRG documents, discussed under the next few sections.

• Resolution M11.11 - Hong Kong SAR - would like to have liaison status with WG 2. ITSD - would like to attend the WG 2 meetings. They have some independence.

• Resolution M11.12 - Future meetings. IRG Meeting 13 is scheduled for May 1999.

Appreciation: WG 2 thanks the IRG experts for participating in IRG.

Document N1867 - Report of IRG Editorial Meeting # 11-1 is for information.

Mr. Takayuki Sato: Reference resolution IRG M11.9 in document N1784 Two characters were removed from the Extension A because these were in R-zone. Their source information has been lost.

1. Mr. Michel Suignard: There was an action item on me to address the naming of the R-zone ideographic characters. The two characters can have Annex P annotations to rectify the situation.
2. Mr. Takayuki Sato: Would like to see the annotation on the character names in Amendment 17 publication.

Two month ballot for FDAM 17 is scheduled (unchanged from earlier) before 1st of July 99. Amendment 17 dates are 07/98 (FPDAM) and 01/99 (FDAM) and 03/99 (AM). The schedule is unchanged - at the hands of SC 2 secretariat by January 1999. Mr. Bruce Paterson will edit the text part of Amendment 17.

Document N1842 is in response to an action item on the IRG. Document N1842 contains the working draft to create PDAM or FPDAM. WG 2 is to consider this document for suitability Document N1842 is edited version of document N1782 (clause X) provided by Mr. Bruce Paterson to Mr. Zhang Zhoucai.

1. Dr. Ken Whistler: Names suggested for the characters could be improved. I can work with the IRG to improve these. The suggested locations are contentious - these positions are not available.
2. I fail to see justification for distinction from 204C - 4D,4E,4F,..52 -- seem to be variants of 5C. It seems these are redundant. Would like to see some justification for separate encoding of these.
3. Mr. Zhang Zhoucai: Justifications have been provided a few years ago in some of the working documents.
4. Mr. Takayuki Sato: Mr. John Jenkins has provided the reasoning at one of WG 2 meetings on the board. We should be able to get Mr. John Jenkins to explain to Dr. Ken Whistler the need to distinguish these.
5. Dr. Asmus Freytag: The proposed code positions are in the general punctuation block. It seems to be applicable to ideographic characters. In a sense these are not punctuation characters. There is coding space after Extended Bopomofo - can associate these with the IRG(?).
6. Mr. Zhang Zhoucai: I need proposals.
7. Dr. Ken Whistler: This should be made into a new block -- starting x31C0 - in the CJK area.
8. Mr. Mike Ksar: A project sub-division already exists.

Disposition: Accept the draft in document N1842, a new block starting at 2FF0 instead of starting at 31C0, with names improved by discussion between IRG and Dr. Ken Whistler. The set is called Ideographic Description Characters. Instruct the project editor to create PDAM-28 text and submit for further processing. (pending resolution of names)

Document N1869 replaces document N1783 and is the proposed PDAM text for Amendment 15.

Mr. Bruce Paterson: In Singapore meeting WG 2 accepted in principle the KangXi radicals - in row 2F (3/4th of a row). Also a small set of CJK radicals were accepted - could fit at that time at the end of row 2F. In Crete meeting, WG 2 agreed to increase the number of CJK radicals - requiring 8 columns (not enough space at the end of row 2F). It cannot overflow at the end of row 2F. WG 2 failed to make a recommendation on the overflow. As editor, I made the decision to start the CJK Radicals Supplement at row 2E.

This draft PDAM is self-consistent internally.

1. Dr. Ken Whistler: I think it is a mistake to make these separated with a half row in between. We should remove the gap between the two. These are still separate blocks - should not have a half row between these. If any additional characters are proposed we should place these in plane 2 rather than in the BMP. Suggest the starting of the CJK Ideographic Radicals at 2E80. This permits contiguous set of columns for both sets of radicals.
2. Mr. Bruce Paterson: Acceptable - it is rework of the hex entries etc.

Disposition: Accept the draft in document N1869 with changes - change the starting positions for CJK supplement; correct the spelling errors in Table for CJK supplement. Process as FPDAM.

Mr. Zhang Zhoucai: We have collected all the CJK characters including BMP, Extension A and Extension B procedures etc. and have collected as Super CJK in document N1871. This is an internal IRG document for cross checking and unification etc. It is used as the basis for Extension B work.

Format becomes important in the context for all the CJK Extension B etc. The lack of fonts prevents us from following the five-column format. At this time we have only one column format using a single set of glyphs. There is lack of resources -- fonts, expertise etc. Due to the heavy burden on creating the five columns we would like to discuss it at this WG 2.

1. Mr. Mike Ksar: I was hoping that IRG would consider this matter and come up with the appropriate recommendation -- whether to follow a single column format or five column format etc. IRG itself came up with the original five-column recommendation in the BMP. The basic requirement for a single column is always there - i.e. a table of at least one glyph per code position. It would be up to IRG to discuss and make a requirement for other formats.
2. Dr. Asmus Freytag: WG 2 could also show our preference / requirement. The minimum requirement could be stated as at least one column and it will be up to IRG to augment that further to see if other formats are required. If IRG finds that the requirement is not there or that there are no resources then IRG can decide to go with one column. From the Unicode point of view - we have been using a single image with five reference formats. In Unicode 3.0 the publication will have a single set of glyphs with reference information for sources moved to a soft copy table. We do not have to do the names for these characters. From the requirement point of view we do not have one for the names etc.
3. Mr. Michel Suignard: We could at least input to IRG that our preference is to have one glyph per code position.
4. Mr. Takayuki Sato: Probably Mr. Zhang Zhoucai is asking is to make his job easier. If WG 2 agrees and gives the IRG freedom then it will make it easier. As editor of part 2 it will make the job easier if we had a single column format. IRG will certainly provide at least one glyph per character. What about the character names -- currently we have names by reference to existing standards. For future extensions what could be the approach - by reference or other means.
5. Mr. Michel Suignard: I would like a WG 2 resolution to express that our minimum recommendation that a single column of glyphs. The names to source references is optional. IRG is invited to make the appropriate recommendation regarding the sources for character names and for the need for more than a single column format.
6. Mr. Zhang Zhoucai: Another issue is that we do not need the decimal code for the ideographs etc. If this is the case, the PDAM text has to be updated.
7. Dr. Asmus Freytag: Our current direction is to remove the decimal indication and we can make the recommendation to remove the decimal notation also from the text of amendment on Extension A.

Action item: IRG to review the schedule with editor of Part 2 -- should be synchronized with Part 2 plan in SC 2 Program of work.

• Single column glyphs (as compared to the five columns of Extension A)
• Absence of decimal notation for code positions in line with new proposed format for 2nd edition of 10646-1.

Mr. Bruce Paterson: Amendment 5 was sent to ITTF about a year ago. The camera-ready copy (CRC) came from Korea when our Korean expert (Professor Kim) was in the UK. When the second version of CRC came out - we were considerably behind schedule and rather than waiting another 6 months for the corrected version we went ahead and published the amendment. The defect arose because of an error on one of the elemental names and algorithmic generation of the names. Document N1806 proposes fixes to these defects.

1. Dr. Ken Whistler: Please go ahead and fix it.
2. Mr. Mike Ksar: Do we need to carry the long names in the next edition - since the algorithm was generating them.
3. Dr. Asmus Freytag: It is a good suggestion and we are in strong support of it - we have done it in Unicode 2.0. The effect in printing was to save pages and algorithm eliminates potential errors. Removal of the long names can also provide economy in space.
4. Dr. Umamaheswaran: There was a concern from UK at the time we balloted on Amendment 5.
5. Mr. Bruce Paterson: I raised this concern with Korea - and they were not in favour of replacing the names with the algorithm and also that in Korea there is no need for them to use the long names - it is used mainly by foreigners.
6. Mr. Michael Everson: I would not like to see names removed from the list of names.
7. Dr. Asmus Freytag: Unicode has a large implementing community. If the names are in fact generated by an algorithm the algorithm is better way - it explicitly describes versus implicitly. The implementers also do not want to refer to a long list of names - an algorithm is preferred.
8. Mr. Michel Suignard: US also supports the removal of the long list of names.
9. Dr. Umamaheswaran: Since the names of the characters are normative and if we decide to replace the list of names with an algorithmic way, which retains the normativeness and generates the unique names.
10. Mr. Michael Everson: I think the removal of long names will make the usefulness of the standard / readability less valuable. We have heard the implemeters' side of the view.
11. Mr. Bruce Paterson: Two years ago - when this topic came up in Tokyo meeting (?) - I had presented a contribution describing a more user-friendly algorithm which was different from the Unicode algorithm of generating the names. At the least such a specification has to be attached to the names of the components.
12. Dr. Asmus Freytag: I am not proposing that the Unicode algorithm must be used - it could be a more user-friendly specification of doing the same job.
13. Mr. Bruce Paterson: I need to check with the UK national body.
14. Dr. Umamaheswaran: Canada will be receptive to replace the list of long names with a suitable specification.
15. Mr. John Clews: I found the specification proposal from Mr. Bruce Paterson very useful. The convention used for romanization of the Korean characters in TC 46 is similar also. Basically the names not in brackets, and those in brackets, are respectively identical to Method 2 and Method 1 transliteration, as described in ISO/TR 11941: Transliteration of Korean. I think it was suggested that this go into a note accompanying the algorithm. It would also require two versions of the algorithm to be shown, if the part of the character name in brackets was also to be indicated. I look forward to hearing how this will be done in practice.

Disposition: The convener to check with Korea national body on whether they have a problem of removing the long list of the names. National bodies to take a position on this approach and feedback their opinion - on whether they will have any objection to replacing the long list of names with a specification of how to arrive at the long list of names from the names of components.

Action items: Convener to forward to Korea for comment. National bodies and liaison organizations to review and invited to express their position on the approach of replacing long names with an equivalent algorithm.

Mr. Bruce Paterson: Defect report is on Tibetan Amendment 6, which was published a few months ago. The list of characters in the Annex E - alphabetically sorted names had some characters, which were not the same as what was in the main body of the document. Since we are in the process of producing a new Amendment on extended Tibetan and the list of characters produced therein will replace the contents of Amendment 6, we could drop the defect report.

Disposition: Working draft on Tibetan Extension in document N1756 replaces the Annex E of Amendment 6 and removes the defect.

Dr. Asmus Freytag: The Unicode consortium is always interested in maintaining synchronization between 10646 and Unicode, in repertoire and other editorial aspects also. We have been working with the editor on a number of editorial alignment items. In terms of the production of the charts - we will reach complete synchronization this time around -- we will use the same names, the same character positions and the same glyphs between the two standards. The Unicode editorial group is interested in getting the kind of feedback required such as Ireland's input to us. In switching over to a combined production process, we may introduce some editorial changes to the 10646 document. We will first prioritize on areas of potential misinterpretation etc. towards minimizing the other minor differences.

Dr. Asmus Freytag: AFII has produced a review copy of the code tables. The binder has been made available for review. I have not had any feedback from anyone except the Irish national body. We are using a custom built tool to layout the code charts. Some items of formatting are programs and any changes would involve program changes. The more massive changes should be brought forward sooner to our attention. Time is of essence. Since document N1810 has been distributed a number of improvements have been made.

Mr. Mike Ksar: I am stressing the point that national bodies and liaison organization should look at document N1810 as samples. The full set of code tables have been made available for viewing / commenting. If you have specific requests contact Dr. Asmus Freytag.

Action item: Delegates attending the meeting are encouraged to review the copy of the code tables and feedback to Dr. Asmus Freytag. National bodies and liaison organizations to review and feedback on the sample charts and name lists for next edition of 10646; in document N1810

1. Dr. Ken Whistler: I welcome this liaison and it will help clear up some of the misunderstandings regarding the use of 10646 in geographical databases.

Mr. Takayuki Sato: Presented his report on his discussion with TC 211 Japanese experts. The pyramid concept in the report is gone from TC 211.

Mr. Mike Ksar: Some improvements can be made - by archiving some of the documents. Feedback can be sent to Mr. Keld Simonsen. Whatever documents we have generated and if we have soft copy -- please send these electronically to Mr. Keld Simonsen for placing them on the WG 2 web site. Old documents are only in paper form. Any WG 2 Numbered Document is candidate -- Word 6, plain text or PDF files can be sent to Mr. Keld Simonsen.

Action item: National body and Liaison Organizations to take note.

Dr. Umamaheswaran presented document N1876 containing the replacement for Annex D in the Principles and Procedures document – N1502.

1. Dr. Asmus Freytag: Editorial -- check the boundaries of cells. Try to get the guidelines part visibly separate from the rest - possibly light shading. Block of Areas using larger fonts. Check alignments in Pictorial view.
2. Mr. Michael Everson: The art used here is ASCII-art. HTML versions are also available. The suggestion was that I maintain a version of the roadmap as WG 2 document and periodically the information in the "procedures" document will be updated.
3. Mr. Michel Suignard: Guideline to use 1K boundary to be friendly to UTF-16, especially for Plane 1. Information about plane 1 roadmap should be refined and included in Part 2 of the standard also.

Action item: Dr. Umamaheswaran to update the principles and procedures document taking into account the feedback at this meeting.

Action items: The ad hoc to enhance the document addressing:

• Stronger demonstrable need for single id for collections of collections
• Listing of all collections in a collection
• Consistency between fixed and non-fixed collection.

Next meeting was to be in Japan March 15--24, 98. There is also Internationalization and Unicode Conference (IUC), March 20-22, 1999 (in Boston) and many of the WG 2 experts are presenting. There is a potential conflict with the SC 2 meeting dates and a misunderstanding on the location. There is a proposal from Japan to move the meeting either starting March 29 or to 11th of April. It may conflict with Easter holidays, and the Greek Orthodox Easter. There are also some Jewish holidays during that time. We have to consider some of these holidays. The second possible date was to start March 1st. There is a third date possible -- 12 April (week after Easter). Delegates are to think about the possible dates and feedback during the meeting.

The meeting location has been changed to Fukuoka from Tokyo -- need to take different flight arrangements than direct flight to Tokyo or Osaka etc. There are 2 direct flights from Tokyo, 8 flights from (Kansai) Osaka. Mr. Evangelos Melagrakis has to be contacted for WG 3 meeting coordination -- before we can get back to Ms. Kimura, SC 2 secretary. If the date is not confirmed - there is possibly some penalty to pay to the meeting location by Japanese national body. SC 2 plenary is the week following WG 2. There is a 19 April 99 CEN TC 304 meeting in Tübingen, Germany.

Mr. Michael Everson: Logistically - is it less expensive in Fukuoka. Internal flights may be expensive.

Recommendation is for April 19 --28, 1998 -- will be suggested to Kimura san, otherwise March 15--24, 1998. Another alternative was to arrange for 8 March 99 (week before the current schedule).

Dr. Ken Whistler: Redmond is offered as a backup location if Fukuoka. If the conflict of dates cannot be resolved, we will stick with original date of March 15, 1999.

Mr. Mike Ksar: We have sent out an email to Kimura san about the next meeting. Proposed date is March 8--13, 1999, for WG 2, with WG3 and SC 2.meetings following WG 2 meeting. The location is Fukuoka with Denmark as the backup.

Mr. Mike Ksar: We have received from Ms. Kimura that dates are confirmed starting on Tuesday, March 9, 1998 instead of Monday the 8th. The series of meetings will end on following Thursday, March 18 (if WG 3 also meets). The next WG 2 meeting will be during the week of September 20--25 in Copenhagen (with US as backup).

• Meeting 36: 09--15 March 1999, Fukuoka, Japan
• Meeting 37: 20--24 September 1999 -- Copenhagen, Denmark
• Meeting 38: March 2000 --China
• Meeting 39 September 2000, Greece, Finland as fall back

WG 2 accepts and confirms the following IRG future meeting schedule:

• IRG 12: 07--11 December 1998 in Redwood Shores, CA, USA
• IRG 13: May 1999, Hong Kong

A drafting committee with participation from Messrs. Bruce Paterson, Mike Ksar, Michel Suignard and Umamaheswaran, drafted the resolutions. These were presented as document N1904draft and were discussed on Friday morning.

Dr. Asmus Freytag: I would like to extend an Open invitation to attend the Unicode Technical Committee meeting - Dec 2--4 1998.

Ms. Joan Aliprand: As the chair of the UTC, I would endorse the invitation and any one requiring visas etc should contact the Unicode offices -- see their web site for details.

Meeting adjourned at 10:15 h on Friday, 1998-09-25.

All the action items from meeting 25, Antalya, Turkey, meeting 26, San Francisco, CA, USA, meeting 27, Geneva, Switzerland, meeting 28, Helsinki, Finland, meeting 29, Tokyo, Japan, meeting 30, Copenhagen, Denmark, and meeting 31, Québec City, Canada, have been either completed or dropped