Re: "French+" support by Unicode

From: JFC Morfin (jefsey@jefsey.com)
Date: Wed Apr 09 2008 - 19:53:14 CDT

  • Next message: Otto Stolz: "Re: Using combining diacritical marks and non-zero joiners in a name"

    I agree with Kenneth on most of his points which reflect an
    anglo-saxon culture.

    My own approach is depends on a French culture and mother language.
    So do most of the debates I participate to : ISOC France,
    france@large (for governance aspects), MLTF, INSDO project. For
    example, the MLTF work concerns multilinguistics, i.e. what in the
    lingual diversity does not belong to linguistic but to language
    plurality (technical support, politics, communication - documentation
    - information protocols), and facilitation technologies (what may
    computer-assist a better representation exchange [psi2psi]).

    1. English is a convenient language but is not precise enough alone,
    both for description and working. French, Russian, Chinese and
    Japanese are precious complements. But this has to be analysed,
    validated, documented to the advantage of the governances.

    2. The Asiatic, English and French languages; each support more
    easily a different logical process. This has to be analysed, studied,
    tested, benefited from in R&D, relational, educational etc.
    protocols. Some semiotic process could be language indifferent (we
    all remember the GU)?

    3. More important : the normalization, strandardization and best
    practice documentation processes need to be far better
    metalingistically understood in order to address the current needs of
    globalization, virtualisation and new forms of governance in a
    pertinent manner. In particular, it appears that most of the SDO's
    still share a decentralised copernician/netcentric point of view,
    which tends to be increasingly at odds with the consensual
    information society paradigm.

    - pragmatic linguistic and utterance operations are certainly an
    easier and cheaper place for study and experimentation of the
    architectonic convergence than nuclear physics or biology.

    - a very interesting study case are the practical support of
    multilingual naming (IDNA, IETF), convergence between multilingual
    DNS, IRIs, and multilingual mail addresses within the IETF,
    interoperability with semantic addressing R&D, and with usage(s), and
    idiolect recogntion.

    I also agree with Kenneth about the "commercial" term. Unicode is a
    commercial consortium (which means that Reasearch is conducted under
    commercial sponsoring with the cons/pros documented in the IAB's RFC
    3869). Other SDOs are "regalian" like ITU and ISO. IETF wants to
    "influence the people who design, use and manage the internet" (RFC
    3935) and as such is certainly influenced. This is not a very
    important point, as long as private innovative contributions can be
    infused in the reflection process, as it is the case here.

    What we found more important during the WSIS regalian domain, civil
    society, private sector, technical community, and international
    entities polylogue is what I called ethitechnics. This is not
    technoethics: how technology can be ethically developped and used. It
    is how to develop technology so its deliveries will be more ethic. An
    equivalent to ecology about standardization pollution.

    On these matters we try to start from what we know best :
    English/French bilingualism.
    Thank you for your help.
    jfc

    At 22:09 09/04/2008, Kenneth Whistler wrote:

    >Marion Gunn said:
    >
    > > I believe it means, Dan, that ISO standards are international standards
    > > published in at least two independent languages (European French and
    > > British English), but Unicode standards are only commercial standards
    > > published only in US (Anglo-American) English, but I could be wrong on
    > > that point, so please do ask Unicode reps to confirm.
    >
    >Well, I'm a Unicode rep, and I'll confirm that that is incorrect.
    >
    >ISO standards *are* in fact International Standards (ISO's
    >capitalization), but most of them are only published in
    >English.
    >
    >ISO/IEC 10646, the main ISO International Standard in question
    >here, is developed under JTC 1, by the rules codified in
    >the JTC 1 Directives. Those directives state, regarding language:
    >
    >7.9.1 The languages of JTC 1 are English, French and Russian. In
    >general, the work of JTC 1 and its subsidiary bodies may be in any one
    >or more of the above-mentioned languages. However, meetings are
    >conducted in any one of these. The Chairman or Convener is entitled
    >to authorise participants to speak in a language other than that
    >in which the meeting is conducted....
    >
    >That's it for the languages for the meetings. Regarding *documents*,
    >the JTC 1 Directives state:
    >
    >8.3.4 One copy, in English or in French, or in both languages if
    >available, of documents relating to JTC 1 ... shall be posted
    >by the JTC 1 Secretariat to the JTC 1 web site in an acceptable
    >document format as specified in Annex H ...
    >
    >And with similarly worded clauses for documents related to the
    >SCs (subcommittees) and WGs (working groups) of JTC 1.
    >
    >There is nothing *requiring* JTC 1 documents, including CDs (committee
    >drafts) and ballots, to be bilingual. On the other hand, there
    >is nothing *preventing* it, either.
    >
    >The actual way this works out depends greatly on the editors
    >of the standards in question, and the preferences of the SCs
    >responsible for them.
    >
    >For ISO/IEC 10646, the standard is developed, balloted, and
    >published in English. There is then an active volunteer effort
    >which takes the published English version, translates it,
    >and makes it available to ITTF for publication as an "(fr)"
    >version of the standard.
    >
    >For ISO/IEC 14651 (collation), the standard is developed, balloted,
    >and published simultaneously in both an "(en)" and a "(fr)" version,
    >but as separate documents. And the understanding of the SC
    >is that in case of any difference of interpretation between
    >the (en) and the (fr) versions, the (en) prevails.
    >
    >For ISO/IEC 15924 (script codes), the standard is developed, balloted,
    >and published as a bilingual (en/fr) document,
    >with English in the left column and French in the right
    >column, as parallel text, as can be seen here on the
    >Unicode website:
    >
    >http://www.unicode.org/iso15924/
    >
    >The reason why those particular standards are available in
    >English and in French, although by 3 entirely different methods,
    >has largely to do with the fact that their editors, to one
    >degree or another, are all bilingual (at least) in English
    >and French.
    >
    >Many *other* ISO standards are only ever developed
    >in English and never get translated. This is easily confirmed
    >by browsing the online ISO store. Of the 14 ISO standards
    >published this week for which language and other ordering
    >information are available today, 3 are available in English
    >and French, 11 in English only.
    >
    >As regards the local varieties of English and French involved,
    >JTC 1 rules don't favor British English and European
    >French over North American English and French -- or for
    >that matter Australian or Indian English or any other variety.
    >Such issues as are distinguished in written forms, such
    >as spelling and occasional word choice, tend to depend
    >largely on the preferences of the respective editors.
    >It is true that ITTF (www.iso.org/ittf) is based in Switzerland,
    >and may in general favour continental usage ;-), but the ITTF editors
    >seldom interfere in the actual balloted text of standards
    >merely to fix spelling conventions or Americanisms in the
    >text.
    >
    >For standards published by the Unicode Consortium, the text
    >tends to be developed in American English, because the
    >majority of the editors are domiciled in the U.S. -- although
    >the editors do make an effort to comb out any regional
    >quirks that would be puzzling to international English
    >readers.
    >
    >And there is certainly no aversion by the Unicode Consortium
    >to having the standard available in other than English, but
    >for the Unicode Standard in particular, translation is
    >very challenging, because of the standard's size, its technical content,
    >the production difficulties with charts and figures, and
    >most of all the frequency of its updates.
    >
    >A French translation of Unicode 3.0 was developed with a
    >great deal of effort by Patrick Andries and others of
    >his colleagues in Canada, and posted online at:
    >
    >http://hapax.qc.ca/
    >
    >That was updated to Unicode 3.1, but the amount of effort required
    >to update it again to Unicode 5.0 (and now 5.1) has outstripped
    >the volunteer resources to do the translation, to date at
    >least.
    >
    >As regards the statement that Unicode standards "are only
    >commercial standards", that is an outdated bromide that doesn't
    >really reflect the reality of standardization in the IT
    >field in the 21st century.
    >
    >ISO, IETF, W3C, and the Unicode Consortium are all SDOs
    >(Standards Development Organizations) developing important
    >IT standards in use around the world. The fact that only
    >ISO is a treaty-based organization, while W3C and the
    >Unicode Consortium are industrial consortia, and the IETF
    >consists mostly of self-appointed experts working by email,
    >doesn't change the fact that all of us depend on the technical
    >standards developed by those organizations (and numerous
    >others) working well together for the international
    >information infrastructure.
    >
    >--Ken
    >
    > > Scríobh Daniel Ehrenberg:
    > > >> Also the differences between ISO
    > > >> 10646 and Unicode 5.0, since ISO 10646 is bilingual.
    > > >>
    > > >
    > > > What do you mean by "ISO 10646 is bilingual"? (I'm sorry, to your
    > > > particular question, I don't know the answer, but I'm just wondering
    > > > about this.)
    > > >
    > > > Dan



    This archive was generated by hypermail 2.1.5 : Wed Apr 09 2008 - 19:59:26 CDT