From: ynk@ome.toshiba.co.jp (Yasushi Nakahara) Subject: Draft RIN 9210 Minutes Reply-To: ynk@ome.toshiba.co.jp Date: Thu, 05 Nov 92 18:01:55 JST Hi all RIN members, Attached please find the draft RIN 9210 minutes, which Mr. Hopper kindly typed up on the last day (evening) of the meeting and I slightly modified (such as adding page numbers) for printing out for RIN member's circulation during the meeting. This online version also corrects (email) address information of some members. The content is exactly the same as the paper version. Please make sure that Mr. Hopper's minutes are more than administrative and technical work. I agree with Mr. Hopper that the minutes content itself would be "art". Again, I appreciate his "artistic" work! The original one was written in TeX with elegant macros. I think that he is also an expert of document "presentation". So, if you want to enjoy the TeX version, please directly contact him or me. Also, I would express our thanks to Kido-san for his help in offering his notebook computer to Mr. Hopper's type-in and his personal printer to my printing-out. Regarding a formal document number of the minutes, I would ask the Leading Rapporteur Mr. Keld Simonsen to appropriately assign RIN's N number and WG15's N number, I think the WG15's number N323 was reserved for RIN Minutes + RIN Resolutions though. The RIN 9210 Resolutions will follow in another email. Best Regards, Yasushi Nakahara TOSHIBA Corp. Phone: +81 428-33-1346|1347 Fax: +81 428-32-0018 Email: ynk@ome.toshiba.co.jp | ..!tsbome!ynk =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Page 1 of 8 ISO/IEC JTC1/SC22/WG15/N___ ISO/IEC JTC1/SC22/WG15RIN/N___ Minutes of the Seventh Meeting of WG15 Rapporteur Group on Internationalisation held at the offices of X/Open Ltd, Apex Plaza, Reading, England 26 - 29 October 1992 Present: K Simonsen, Danish Unix Users Group, Denmark (Chairman) keld@dkuug.dk R Barker, Uniforum Inc, US ralph@uniforum.org L Boiteau, Govt of Canada uunet!revcan!DCSEM!256 P Dempster, Canadian Standards Patric.Dempster@canada.ncr.com H Jespersen, Posix Inc, US hlj@posix.com E Kaae, NCR, Denmark Erik.J.Kaae@copenhagen.ncr.com A Kido, IBM, Japan kido@vnet.ibm.com G Leijonhufrud, Posix Ltd, UK -- Y Nakahara, Toshiba Corporation, Japan ynk@ome.toshiba.co.jp B Needham, Amdahl Corporation, US barry@uts.amdahl.com L Randolph, NASA, US lrandolph@nasa.gov N Saito, Keio University, Japan ns@sfc.keio.ac.jp J Spencer, Posix Inc, US spencerj@dg-rtp.dg.com Y Suehiro, Digital Eqpt Corporation, Japan suehiro@jrd.dec.com K Hopper, Waikato University, New Zealand (Secretary) kh@waikato.ac.nz Agenda Action By ------ --------- 1.1 1. K Simonsen was appointed to chair the meeting, K Hopper as secretary and a drafting group formed from R Barker, Y Nakahara and P Dempster. 1.3 2. The temporary document numbers given in Annex A to these minutes were assigned by the chair. Those given permanent document numbers are indicated in that Annex. 1.4 3. The agenda as presented in ISO/IEC JTC1/SC22/WG15RIN/N074 was amended to include Item Detail ---- ------ 3.7 Discuss draft formal specification in RTN003 3.8 "scantoxx" functions -- N059 refers. Page 2 of 8 Agenda Action By ------ --------- 4. The following document allocations to agenda discussion 1.4 items were made --- N055R to 3.2, N057 to 3.1.1, N058 to 4.2, (cont) N059 to 3.8, N062 to 3.1.12, N063 to 3.2, N069 to 3.2, N074-077 to 3.4, N078 to 3.4.2, RTN001 to 3.2, RTN002 to 3.3, RTN003 to 3.7. It was noted that N067 will be reissued when revision is complete and that N060R will be distributed by mail when copies become available. 1.5 5. The minutes of the sixth meeting were approved nem con. 2.0 6. The outstanding actions from the sixth meeting have been actioned as follows :-- 920501 Japanese/US paper by e-mail. Closed " 02 RTN001 " 03 RTN002 summarises Japanese, Canadian, US and French views " 04 RTN004 " 05 This remains open under Agenda Item 3.2 " 06 This has been overcome by events -- considered closed " 07 RTN003 " 08 Members are to inform R Barker regularly of the ALL dates of national meetings 3.1 7. H Jespersen reported on the IEEE meetings in Utrecht as follows :-- a. State-dependent encoding was discussed and it was agreed that applications should not handle the problem. b. Multiple printing width character problems were discussed with the result that the keywords `width1', `width2', `width3' and `width4' will be added to the LC_CTYPE category of the locale definition, each being followed by a list of characters in these new classes. There is also to be a `default_width' keyword taking a numeric argument which indicates the width occupied by all characters not specifically allocated to a different width class. c. Collation syntax problems are still unresolved (see RTN009). d. It was discovered that `era_d_t_fmt' had been inadvertently omitted from the LC_CTIME category. This is to be corrected using wording from the X/Open Portability Guide 4. e. The copying forms of locale category specification will have some name syntax options. f. It was agreed that portable character sets must not be merely a single octet coding. g. Mappings for other matters in the LC_CTYPE category was not considered to be within the scope of POSIX. h. The pax file format is now based on ISO 1001. User names, group names and file names are in UTF-File System Safe encoding. A new/better format is still looked for. Page 3 of 8 Agenda Action By ------ --------- 8. Discussion of RTN015 on PJ Plauger's book definition of a localedef translator revealed that it is definitely code-set oriented. This is deemed dangerous in a POSIX environment. A liaison statement to WG14, WG20 and WG21 on this danger is to be drafted. 3.1.1 9. No extended identifier proposals have yet been seen. 3.1.2 10. It was agreed that all API interfaces must be given specifications in the LIS version of ISO 9945-1. A resolution is to be produced asking the US development body to consider this matter. 3.1.3 11. Collation weight naming is still under discussion in Japan. 3.1.4 12. The use of `substitute' in collation was suggested. A review of the history of this shows that this gives recursive definitions between the locale and regular expressions -- which cannot in general be shown to be finite. DIN 5007 and the Canadian standard on sorting do not use this, but the highest level of the Danish sorting standard (DS377) does. 13. The Danish national body is to produce a paper before the Denmark next meeting on its perceived need for the use of substitution in the collating order category of a locale vis-a-vis DS377 and in particular the level at which that appears to be necessary (921001). 3.2/3 14. (The group was joined by WG15 convener for this item) The Japanese members introduced RTN001 -- a NWI Proposal for a TR for a national profiles guideline. Document WG15/N212 is a proposed initial draft. 15. SGFS is producing TR10000 in parts covering specific profiles (eg TR10000-3 -- OSE Profiles). None of these parts addresses national profiles. These seem to be needed by WG15 as ISPs. 16. Each functional category of the POSIX locale is seen as a functional specification. For each there may be instantiations which are registered as separate entities. Combinations of already registered and (possibly) new variants (to be registered) then form a national profile. 17. The Japanese national body is to expand RTN001 in light of Japan the above discussion. This expansion is to include provision for a separate part for a TR covering POSIX locale guidelines (921002). 3.1.5 18. Discussion of RTN014 resulted in a decision not to proceed with either `reorder_after' or `replace_after' mechanism in locale ordering. 3.1.6 19. It was reported that the requirement for NUL to be handled separately had been dropped. It was suggested that NUL would be defined as in ISO 6429:1988 for all possible character sets. This is to be checked. Page 4 of 8 Agenda Action By ------ --------- 20. The Danish national body is to provide a proposal for Denmark a definition of NUL to this group and to the US development body for consideration at its January meeting (921003). 3.1.7 21. A review of Uniforum and X/Open documents on state- dependent text encodings has led the Japanese C-language group to develop a minimal set of functions for their manipulation. The whole matter of state-dependent encoding is agreed to be necessary, but the question of exactly what needs to be included is left for later consideration and further discussion. 3.1.8 22. Character printing width discussion is closed. 3.1.9 23. Character set conversion based on charmaps is now provided by the iconv function. 3.1.10 24. Field display width has been added to the date category in the locale definition. 3.1.11 25. The pax utility does support extended character sets. 3.1.13 26. BASE64 encoding is now in draft 4 of POSIX.2b. 3.1.14 27. The Japanese C-language group Multi-byte character Support Enhancement draft was submitted to the US development body for information. 3.1.15 28. The Danish draft on invariant ISO646 is seen as a rehash of the original trigraph proposals to digraphs. This should be approved by WG14 before this issue may be re-opened in this group. Closed pending such approval. 3.1.1 29. The reference in RTN017 to CD 9945-2 p187, Section 3.1 was discussed and it was not liked. 30. The Japanese national body is to reword the proposal of Japan RTN017 relating to Section 3.1 of CD9945-2 and pass to the US development body for the consideration of the appropriate drafting committee as soon as possible (921004). 3.2 31. Objections to the use of N055R as a first draft locale guideline were raised by NZ on the grounds that it contained unacceptable errors, uncertainties and ambiguities. The action item 920507 resulted in these being discovered (as indicated in detail in RTN003). After some discussion it was agreed that a Table of Contents should first be approved before starting to flesh this out with text. 32. The Japanese national body is to co-ordinate and draft Japan the development of a Table of Contents for both National Profile and locale guideline Technical Reports (921005). 33. A draft registration scheme and naming convention for locales is given in RTN012. N080 is a similar document from CEN/CENELEC. As a result of these documents, WG20 will send a liaison statement to all SC22 WGs enquiring if the POSIX mechanism is (or is not) likely to be suitable for the registration of all forms of cultural elements. Page 5 of 8 Agenda Action By ------ --------- 34. XOJIG will also be presenting a draft interim proposal for a registration and naming mechanism in December for review and comment. It is therefore agreed to wait further action until this is received. 35. WG20 has indicated in its liaison statement (RTN016) that it does not intend to consider registration of charmaps which is needed by WG15. In the light of this it was agreed to invite any comments which XOJIG may have on WG15 needs for charmap registration. 36. The RIN lead rapporteur is to write to XOJIG stating the Chair needs perceived by WG15 for charmap registration and inviting that group's comments vis-a-vis RTN012 and RTN013 (921006). 3.4 37. The group confirms that the WG15 response to SC22 requests is satisfactory. 3.5 38. K Hopper spoke briefly introducing some details of the planning for the Ad Hoc meeting on character sets to be held in Copenhagen from 21-23 April 1993. 39. R Barker indicated that through lack of input to him, the promised paper on ISO 10646 would not be forthcoming. It would be overtaken by the Ad Hoc meeting. 4.2 40. Consideration was given to the WG20 liaison report, noting that WG20 wished to retain the use of the term `cultural element' and recommending that WG15 proceed with the registration mechanism. 41. After some discussion it was agreed that further reports be made to WG20 through the WG15 liaison representative. The liaison statement is to be drafted as a recommendation and is to include a request for WG20 definition of the term `cultural element' and its understanding of the relation of that term to the POSIX locale. 3.2 42. The Japanese locale Version 2.1 is a first attempt at a full standard. Development of the collating order mechanism needed is following the Canadian lines, but there are difficulties relating to combination of phonetic, character, stroke, radical and character class needs. 43. It was agreed after some discussion that there is a need for co-ordination and harmonisation in the development of national locales. The Japanese member body indicated its willingness to accept such input for use in the preparation of the guidelines documentation. 44. All member bodies are to provide feedback to the Japanese ALL member body on style issues relating to the definition of national locales and profiles for incorporation into a first draft of the proposed TRs (921007). 3.7 45. K Hopper explained the development reasons behind the draft formal locale and charmap specification. Advice was requested as to whether this group considered that completion of the work was justified, including complete determination of Page 6 of 8 Agenda Action By ------ --------- weaknesses and errors, etc -- with help! Discussion of this 3.7 indicated that there could never be any intention of including (cont) a vdm-sl based specification as part of a revised ISO 9945-2. Since the work was identifying problems, however, it should proceed, obtaining help in resolving complete uncertainties from the US development body as needed. 46. The New Zealand member body is to continue work on the New Zealand formal definition of locales and charmaps (including regular expressions used therein), solliciting help by e-mail to resolve uncertainties when needed. A list of suggested technical changes is to be forwarded to the US development body drafted using the style and terminology of the existing DIS 9945-2 (921008). 4.2 47. WG20 is uncertain why SC22 rejected its NWI Proposal on registration of cultural elements. It was agreed that this was premature, given that WG20 is as yet uncertain of the definition of the term itself. 4.6 48. New Zealand reported that Leo Pasifika (S Pacific Language Forum) had an interest in having locales developed for its members' needs. This is being pursued by New Zealand. 5 49. The resolution drafts were examined in detail. It was agreed not to ask WG20 for a chance to review their revised draft NWI Proposal. 50. The liaison officer to WG20 is to take note of the intent of the discussion over the liaison statement resolution in relation to ideas on the definition of terms in presenting the statement to WG20. 51. The resolution to propose a liaison statement to SGFS in respect of profile terms and definitions was postponed due to a lack of time to fully develop a text. Development of such text is to be done using e-mail. 6.3 52. The actions attached as Annex B to these minutes were reviewed and approved. 6.2 53. Permanent document numbers will be assigned to the temporary documents outside the meeting. 6.4 54. Thanks were unanimously recorded to X/Open Ltd, our hosts, for the generous offer of their facilities. K Hopper Secretary Page 7 of 8 Annex A to ISO/IEC JTC1/SC22/WG15RIN/N___ Temporary Document Number Assignments Temp No Document Details Perm No ------- ---------------- ------- RTN001 Draft NWI Proposal re `Guidelines for POSIX National Profile' N081 RTN002 Definition of National Profile term from Japan RTN003 Standard POSIX Charmap & Locale Specification from NZ N082 RTN004 RIN Action Items 920501, 920503, 920504 N083 RTN005 Collection of National Profile definitions from Japan N084 RTN006 --- deleted --- (duplicated RTN001 in error) RTN007 Draft Japanese National Profile Version 2.1 N085 RTN008 Draft Amendment to C-language standard for MSE from Japan N086 RTN009 Japanese comments on POSIX.2b Draft 4 N087 RTN010 LC_CTYPE extension for additional character mappings N088 RTN011 IT -- Cultural Elements Specification WG20/N117 N089 RTN012 IT -- Registration Procedures for Cultural Elements N090 Specification WG20/N116 RTN013 Canadian contribution to WG20 N091 RTN014 Danish note on reorder_after and replace_after N092 RTN015 Proposal for localedef format from T Plum (WG14) N093 RTN016 Liaison statement from WG20 N094 RTN017 Extended Identifiers in POSIX utilities from Denmark. N095 Page 8 of 8 Annex B to ISO/IEC JTC1/SC22/WG15RIN/N___ Summary of Outstanding Actions I -- From Previous Meetings 920505 920508 -- permanent II -- From Seventh Meeting -- New Actions 921001 The Danish national body is to produce a paper before the next meeting on its perceived need for the use of substitution in the collating order category of a locale vis-a-vis DS377 and in particular the level at which that appears to be necessary (Minute 13). 921002 The Japanese national body is to expand RTN001 in light of the above discussion. This expansion is to include provision for a separate part for a TR covering POSIX locale guidelines (Minute 17). 921003 The Danish national body is to provide a proposal for a definition of NUL to this group and to the US development body for consideration at its January meeting (Minute 20). 921004 The Japanese national body is to reword the proposal of RTN017 relating to Section 3.1 of CD9945-2 and pass to the US development body for the consideration of the appropriate drafting committee as soon as possible (Minute 30). 921005 The Japanese national body is to co-ordinate and draft the development of a Table of Contents for both National Profile and locale guideline Technical Reports (Minute 32). 921006 The RIN lead rapporteur is to write to XOJIG stating the needs perceived by WG15 for charmap registration and inviting that group's comments vis-a-vis RTN012 and RTN013 (Minute 36). 921007 All member bodies are to provide feedback to the Japanese member body on style issues relating to the definition of national locales and profiles for incorporation into a first draft of the proposed TRs (Minute 44). 921008 The New Zealand member body is to continue work on the formal definition of locales and charmaps (including regular expressions used therein), solliciting help by e-mail to resolve uncertainties when needed. A list of suggested technical changes is to be forwarded to the US development body drafted using the style and terminology of the existing DIS 9945-2 (Minute 46).