Tendances aux Pays-Bas en matière d'usage légal de jeux de caractères

[Alain LaBont� <alb AT sct.gouv.qc.ca>]

Rappel : les Pays-Bas ont l'obligation légale de respecter tous les signes
diacritiques composant le nom d'une personne, lorsque ce nom est soit écrit
d'origine en alphabet latin, soit transcrit à partir d'un autre système
d'écriture... le non-respect de cette obligation peut, par exemple, faire
échouer un procès intenté à une personne, pour vice de forme... l'identité
d'une personne  est considérée là -bas comme quelque chose de plus sérieux
qu'ici dans les faits. La connaissance de ce fait est un préalable à ce qui
suit. Cette obligation très sérieuse leur fait rechercher des limites quant
à l'aplication de leur loi, et ces limites ne vont pas pour l'instant
au-delà de l'alphabet latin, sauf peut-être timidement jusqu'à l'alphabet
arabe, un petite ouverture en raison de la communauté marocaine importante
en sol néerlandais.

Alain LaBonté
Québec

>Date: Thu, 26 Feb 1998 16:58 +0100 (CET)
>From: Johan van Wingen 
><PRECAL AT rulmvs.leidenuniv.nl>
>X-Sequence: 
>SC2 AT dkuug.dk
> 392
>Errors-To: 
>SC2-request AT dkuug.dk
>Subject: (SC2.392) Developments in the Netherlands
>To: SC02 List 
><SC2 AT dkuug.dk>
>
>Dear Colleagues
>It may interest you to be informed about recent developments in the
>Netherlands regarding the national policy on the use of coded character
>sets, and in particular on subsets of 10646.
>
>There are two projects going on here, in both I participate.
>1.  Revision of:
>    NEN 1888, General Personal Data
>    NEN 5825, Addresses
>
>    This is prepared by NNI NC 380 007
>    Represented are:
>    Ministry of the Interior
>    Ministry of Justice
>    Ministry of Finance (Taxation)
>    Police Information Centre
>    National Chipcard Platform
>    Ass. of Neth. Communities
>    Social Security
>    Ediforum
>    GBA (Communities Data Exchange)
>    PTT Post Media service
>    State Agency of Road Transport
>    Health and Welfare Informatics
>    Health and Welfare Insurance
>    Housing Corporations
>    Broadcasting Subventions
>    Direct Marketing Ass.
>    Some Service bureaus, among them:
>    Human Inference
>    Directview
>    (some I left out for which I did not know the English equivalent)
>
>    A subgroup is working on character set matters.
>    This work covers:
>    -- selection of coded character set standards to be used,
>    -- conversion between Latin character repertoires,
>    -- transliteration of non-Latin characters to Latin.
>
>2.  Police Information Centre
>    To select character sets for use with Police systems, and their
>    mutual conversion.
>
>Both projects work in close cooperation.
>
>There are no decisions as yet, but the directions are already clear.
>These may be summarized as follows:
>------------------------------------------------------------------------
>                                                            VERSION 0.3
>                                                             1998-02-17
>                                                        J. W. van Wingen
>
>Principal trends in character handling in systems of the Netherlands
>Government or related institutions
>
>1.  It cannot be expected, nor required, from a civil servant to be able
>to handle non-Latin scripts. Thus handling of these scripts will not be
>included in the normative parts of the standards under development.
>
>2.  What matters primarily for data exchange and storing is the
>repertoire of characters, not their coding.
>
>3.  In principle three repertoires will be specified for use:
>        a small one,
>        a middle one,
>        a large one.
>The selection will be directly related to the levels of Edifact (ISO
>9735).
>
>4. The small repertoire may be ASCII (ISO/IEC 646 IRV:1991) or Edifact
>Level B (invariant subset of ISO/IEC 646:1991, ISO-IR 170).
>
>The middle repertoire will be that of ISO/IEC 8859-1 or 8859-9. (Whether
>there is, or will be, any actual use of it is a matter of doubt.)
>
>The large repertoire is the GBA set, that is that of Teletex (T.61 or
>T.51, subset of ISO/IEC 6937), without the IJ as a single character.
>(This is part of the specifications for the GBA system which forms the
>basis for all personal data communication and for authentication in the
>Netherlands.) No additional characters will be permitted.
>
>5. ISO/IEC 10646-1:1993 is considered not mature nor stable enough to be
>recommended, apart from a carefully selected subset, which is identical
>to the GBA set. It may offer coding for the large repertoire, but there
>is no intention to change the teletex coding with the GBA system to full
>two-octet coding.
>
>6. For the coding of the GBA set three methods are considered suitable.
>    a. that of ISO/IEC 6937
>    b. that of UCS-2
>    c. that of UTF-8
>The choice will depend on the application (database, network).
>------------------------------------------------------------------------
>
>>From the trends it should be clear to SC2 that there is no interest at
>all for promoting larger subsets than the GBA set for use with any body
>related to the Government. Even the interest for transliteration rules
>from other than Latin scripts is very restricted. The Committee is only
>permitted to discuss that if time allows.
>
>It should be realized that not keeping to the principle of Latin script
>could even involve the risk of conflicts with the Unions. Thus there is
>no demand, and thus no market, for any large scale implementation of
>other scripts in the Netherlands, apart from a very limited production
>of Arabic texts for Moroccans living here.
>
>It is under consideration to make the GBA set a EN from CEN. This may
>take some time, because the specification of the GBA set (taken as a
>repertoire or as a subset of 10646-1) is under review at present. It
>has been discovered that there are subtle differences between ISO/IEC
>6937, ITU T.51 and the former ITU T.61 (Teletex). These pertain only
>to special characters, not letters. For letters the GBA set has the
>same repertoire as ISO/IEC 6937, and that is fixed.
>
>Because this set of letters (subset of the GBA set) is also in use with
>ITU, and with X500, it will become in the future a very important thing
>for the whole of Europe, in particular because it has been stable since
>1983 at least. This may disappoint some people, but the cost of change
>will be very high, and as has been decided, will have to be borne by the
>proposers.
>
>Best regards from J. W. van Wingen
>P.O.Box 486, NL-2300 AL Leiden, Netherlands,
Email:PRECAL AT rulmvs.LeidenUniv.nl