SC22/WG20 N1011

From: Matthew Deane [mailto:mdeane@ANSI.org]
Sent: Monday, February 10, 2003 9:49 AM
Subject: (SC22docs.1878) SC 22 N 3541 - Summary of Voting on SC 22 N 3523, Text for Second CD Ballot, ISO/IEC CD 15897 - Procedure for the Registration of Cultural Elements

ISO/IEC JTC 1/SC22
Programming languages, their environments and system software interfaces
Secretariat: U.S.A. (ANSI)

ISO/IEC JTC 1/SC22 N3541

TITLE:
Summary of Voting on SC 22 N 3523, Text for Second CD Ballot, ISO/IEC CD 15897 - Procedure for the Registration of Cultural Elements

DATE ASSIGNED:
2003-02-10

SOURCE:
SC 22 Secretariat

BACKWARD POINTER:
N/A

DOCUMENT TYPE:
Summary of Voting

PROJECT NUMBER:
1.22.15897

STATUS:
The results of this ballot are forwarded to SC 22/WG 20 for review and appropriate action.

ACTION IDENTIFIER:
ACT

DUE DATE:
N/A

DISTRIBUTION:
Text

CROSS REFERENCE:
SC 22 N3523

DISTRIBUTION FORM:
Def

Matt Deane
ANSI
25 West 43rd Street
New York, NY 10036
Telephone: (212) 642-4992
Fax:             (212) 840-2298
Email: mdeane@ansi.org
_____end of cover page, beginning of document______________
SUMMARY OF VOTING ON
Letter Ballot Reference No: SC22 N3523
Circulated by:                JTC 1/SC22
Circulation Date:            2002-11-08
Closing Date:                 2003-02-08

SUBJECT: Summary of Voting on SC 22 N 3523, Text for Second CD Ballot, ISO/IEC CD 15897 - Procedure for the Registration of Cultural Elements
----------------------------------------------------------------------
The following responses have been received on the subject of approval:
"P" Members supporting this appointment
6 (China, Czech Republic, Denmark, Republic of Korea, Norway, Switzerland)
P" Members supporting this appointment, with comments
1 (Germany)
"P" Members not supporting this appointment
4 (Japan, Netherlands, UK, USA)
"P" Members abstaining
2 (Austria, France)
"P" Members not voting
11 (Belgium, Brazil, Canada, Egypt, Finland, Ireland, DPR of Korea, Romania, Russian Federation, Slovenia, Ukraine)
Note: O-member Sweden voted to approve
___________ end of summary, beginning on NB comments _____________
Germany

editorial comments:

9.2.8

"but need not refer" --> "may refer"

Annex D.2

Check the printing

Technical comments:

Introduction:

Add "XML" after SGML in the introduction and passim (of course, XML is
implied by SGML, but better be explicit)

4.4 text-file

"A file" --> "A human-readable file" (or similar)

5.5 "No modification..."

Allow versioned registrations / updates to existing registrations,
provided all older versions are still accessible and can be uniquely
identified (to a degree, this possibility is implied by Annex A, point
14)

7.3 Identity

Create and quote a URL for registrations in Russian

9.3.7, point 9

Add note on registratiosn for the European Union (EU)

9.4 "contents of a narrative cultural specification"

General:

The list of optional clauses is fairly arbitrary. However, all such
lists are more or less arbitrary, so no change is required at this point
beyond a note that should point out that the list of optional clauses
may be open to additions in the future.

Clause 1:

"Multilingual Sorting" --> "Multilingual Ordering"

The text should refer to ENV 13710 (as it stands, it suggests that
there may be a European Standard in the future, but that it does not
exist already)

Clause 16:

For curiosity: In which culture are family names capitalized
throughout?

Annex A:

The statement on copyright here conflicts with 9.3.6. Please clarifiy
(what is really needed is not a faiver of copyright but a license to use
the registrations without charge)

Japan

There are no market needs for this registration.

Netherlands

GENERAL

1 - ISO/IEC 15897 should be a self-contained standard as much as
possible. That means that references to Posix and to 14651 should be
reduced to a minimum, and where necessary, definitions and explanations
should be added, even if this implies copying lines from Posix or
14651.

2 - The reference to CEN TC304 (see clause 8.1) should be removed.

SPECIFIC

We do not have the means to deal with every clause, but we restrict
our comments to 9.4. The central question here is: Does this answer
what industry wants to know?

Clause 1-6:
Expand the text to explain what is really asked for. For example,
several readers told us that they had no idea what "deterministic
ordering" means. Also, a more logical sequence of clauses would
facilitate answering a questionnaire.

Clause 1:
This requires knowledge of the national alphabet, which is only made
available in clause 9. Thus put this clause first.
The question of ordering is repeated in clause 10. Why? What is the
difference? Are there two possible approaches of ordering?
Which of the two produces "culturally expected results"?

Clause 7:
This presupposes that such a thing exists. What if it does not?

Clause 9:
This question is worded in such a way that it may be interpreted by
readers quite differently, with as effect that answers are no longer
comparable, to the distress of industry.

Clause 10:
See above.

We may continue this way, but the most significant improvement to
9.4 would be that the questions should no longer put riddles to the
reader, which he cannot solve. A questionnaire in the present style
will
motivate very few people to give honest and true answers.

United Kingdom
The WG20 meeting in Tromso agreed that an alignment with the registration procedures in 2375 should be made; this has only be partially carried through, leaving a procedure which is far from adequate. In particular it is not acceptable that a registered set of cultural conventions should only be checked for conformance to some
format, we need to recognise that a set of cultural conventions could be transnational and need to be agreed by several national bodies through some appropriate procedure.

USA

Subject: US Comments on CD2 Text for the Revision of ISO/IEC 15897
From: US National Body
Date: 2003-01-29

References (SC 22/WG 20 documents):
N 893, Summary of voting on CD registration and CD ballot for ISO/IEC CD 15897. - Registration of cultural elements
N 957, Disposition of comments on CD of 15897
N 987, Information technology - Procedures for registration of cultural elements (ISO/IEC CD2 15987:2002(E))

Notation:
A substantial number of the US comments on the first CD were not accepted, for reasons that the US does not agree with. In addition, other US comments that were accepted (either in actuality or in principle) have not been adequately incorporated into the text of N 987. If text from an earlier comment is quoted, the original number of the comment (as it appeared in N 893 or N 957) is shown, with the prefix "CD1" to distinguish it from comments on the second CD.

Organization:
US comments consist of:

·       general comments on technical issues and on editorial issues;
·       technical comments on specific clauses; and,
·       editorial comments on specific clauses

supplemented by four appendices.
Numbering of US comments on the second CD is continuous (including the appendices).

GENERAL COMMENTS ON TECHNICAL ISSUES

General Comment #1 - Technical issue: On use of TR 14652
As specified in Annex E of the JTC 1 Procedures (<http://www.jtc1.org/directives/main.htm>]), registration requires two standards:

· a technical standard ("The standard containing the definition of the classes of objects requiring registration"), and

· the procedure standard ("The standard containing the specific procedures for the JTC 1 Registration Authority to follow")

[Quotations are from Annex E.] ISO/IEC 15897 is the procedure standard.
For a POSIX Locale or definition of a POSIX category, the applicable technical standard is ISO/IEC 9945:2002. In other clauses, this CD2 references TR 14652.

TR 14652 is being published as a Type 1 Technical Report rather than as an International Standard because consensus could not be reached on five significant topics: LC_CTYPE, LC_MONETARY, LC_TIME, LC_XLITERATE, and REPERTOIREMAP. All of the clauses dealing with these topics are marked "controversial," i.e., there are no agreed-upon specifications for these topics.

In several places, the CD2 references one of the controversial topics in TR 14652 and sometimes specifies that the controversial topic shall be used (for example, in Clause 9.3.9). Before a controversial topic can be used, the controversy must be resolved. If this is not done, the information in registrations will be unreliable, and there will be problems with interoperability.

It is unacceptable that there would be an attempt to elevate material that is specifically labeled "controversial" in TR 14562 to normative status in ISO/IEC 15897 by specifying its use. Instead, ISO/IEC 15897 must forbid use of any of the clauses in TR 14652 that are identified as "controversial".

In comments on the "Scope" clause (which is normative), the US has supplied new text to ensure that this fundamental requirement is met.

General Comment #2 - Technical issue: On specification of procedures
The US commented on CD1 clause 4 (now clause 7 in the CD2) as follows;

CD1 OBJECTION #10
Section: 4 REGISTRATION AUTHORITY
Problem and Action:
It is vital that cultural specifications be reviewed by those who represent varying viewpoints. Existing cultural specifications registered under ISO/IEC 15897 have often been written by the editor of this IS, and often accepted into the registry by the same person. This is a serious conflict of interest. The rules of the registry must be written such that a person who writes or proposes a cultural specification is not also the person who decides whether it is accepted. Further, the registration authority must be made up of representatives from different geographic areas and representing different interests (for example, industry, standards committees, government agencies). Although DKUUG is to be congratulated for volunteering to be the Registration Authority, a group with more varied backgrounds and expertise must take on this task for the registry to be successful.

The Editor responded in the DoC (N 957):

10. Accepted in principle. The proposed RAC will address this problem, as well as the N 945R contribution, which will be taken into account when writing the next draft.

The clauses in N 945R describing the procedures in detail have not been incorporated into the CD2. (See also specific technical comments between US comments on clauses 9 and 10.) The US is concerned about the lack of detailed information on procedures for the reasons given above.

JTC 1 Procedures (Annex E, Clause E2.6) require:

The procedure standard shall define the process for the JTC 1 Registration Authority to review and respond to applications to ensure fairness and shall define the maximum time intervals between steps of the process.

The requirement for fairness means that it is incumbent upon the Registration Authority to evaluate an application fully. The administrative material in an application can be checked by a single person, but when it comes to the technical aspects, no one person can be expected to be able to see all the technical ramifications of information in the application. This is particularly true when the Registration Authority is unfamiliar with the culture being specified. The Joint Advisory Committee must therefore be involved in the technical evaluation of each application. This parallels ISO/IEC 2375, where the Joint Advisory Committee is charged with the technical evaluation of all mappings to ISO/IEC 10646 characters included with applications for registration.

Clause E.2.6 also states:

Where the JTC 1 Registration Authority is expected to perform a technical role in determining conformance of the object to be registered to the technical standard, this role shall be defined in the procedure standard. The response to the applicant shall include information pertaining to the results of the technical review.

There is no question that a Cultural Specification registration is a technical document (particularly those that conform to POSIX requirements). Therefore, a technical review of each new or revised application is mandatory, and the complete process must be defined in the procedure standard.

The above comments also apply to revisions or replacements of existing registrations.

GENERAL COMMENTS ON EDITORIAL ISSUES

General Comment #3 - Editorial Issue: Poor Organization
While CD2 is an improvement over CD1, many of the organizational defects of CD1 still exist. In particular, organizational changes based on ISO/IEC 2375 which appear in document N 945R have been disregarded.

Division of the normative general content of this standard (following Terms and conditions) should be in four basic parts:

·       definition of the agencies responsible for this standard and the procedures defined in it
·       definition of the contents of the cultural specification and layout of the registration documents
·       detailed description of the procedures for submission, review, and approval of an application for registration
·       appeal procedures and subsequent procedures associated with a registration.

Specific US technical comments describe the necessary changes in detail.

General Comment US#4 - Editorial Issue: Uncaught errors
From the number of spelling and grammatical errors in CD2 15897, it is appears that the Editor did not perform spell-checking and grammar-checking on the text. This is inadequate and unacceptable -- these functions are readily available in modern word-processing software (including language-specific settings). While the Editor may intend to correct such errors at the final stage (or rely on ITTF to do so), it is preferable to catch even editorial errors at an early stage of the technical work. The US considers that every stage of a standard should be as correct in spelling and grammar as possible.

General Comment US #5 - Editorial: Titles of clauses
The titles of all clauses should follow the practice used in the ISO/IEC Directives, Part 2. That is, they should be in lower case, except for the first letter and any terms that are capitalized in the text of the standard (for example, "Registration Authority").

The ISO/IEC Directives, Part 2, do not appear to have specific instructions on capitalization of the titles of clauses (the relevant clause, 5.2.2 states only: "Each clause shall have a title, placed immediately after its number, on a line separate from the text that follows it.)."

End of General Comments

SPECIFIC TECHNICAL COMMENTS

Foreword

CD2 TECHNICAL #5
The second-last paragraph of the Foreword reads:

This International Standard registers amongst other items Cultural FDCC-sets, charmaps and repertoiremaps as defined in ISO/IEC TR 14652, and POSIX Locales and POSIX Charmaps as defined in ISO/IEC 9945-2 "POSIX shell and utilities".

Delete "and repertoiremaps"
Rationale: As noted above, repertoiremaps are identified in ISO/IEC TR 14652 as "controversial." Therefore, this International Standard cannot register "repertoire maps as defined in ISO/IEC TR 14652" because there is no agreed-upon definition.

Introduction

CD2 TECHNICAL #6
First paragraph, last sentence, and Second paragraph, first sentence:
Insert "the non-controversial clauses of" before "ISO/IEC TR 14652" in both places.
Rationale: The clauses of ISO/IEC TR 14652 marked "controversial" shall not be used. Only the non-controversial clauses may be used.

CD2 TECHNICAL #7
Second paragraph, second sentence:
In CD1 Objection #4, the US recommended a number of changes, all of which were "Accepted in principle." One of these recommendations was:

Thus, the sentence should end "...will also be freely available."

The Editor added the URL for the ISO web pages, but retained the URL for DKUUG, so that the sentence now reads:

The registration will be free-of-charge and the registered cultural elements will also be freely available on the network at the address <http://www.iso.org/mara/> (and initially at <http://www.dkuug.dk/cultreg/>).

The CD2 text now implies that the "registered cultural elements" are available at both URLs. This is incorrect. The ISO "mara" URL yields a list of registration agencies, not "registered cultural elements".

To eliminate confusion, delete both URLs, so that the sentence ends "... will also be freely available." as previously recommended.

Rationale:

(a) If the URL for the English language ISO site is given, the URL for the corresponding French language site should also be given, since French has equal status with English as an official ISO language.

(b) The URLs duplicate information that is available elsewhere:

(a) Both ISO URLs are published in clause 7.3;
(b) The DKUUG URL is published on the two ISO sites.

Including them here is excessive and unnecessary detail for an Introduction.

(c) [From CD1 OBJECTION #4} While DKUUG is the initial maintainer of these cultural definitions, that could change over time, so it seems inappropriate to list the address here in the Introduction.