Last call for ISO 15924-based updates

Lang Gérard gerard.lang at
Tue Mar 17 08:13:20 CET 2009

Dear Peter,

I understand your position very well, but as I already had the occasion to write it at least one time before, there is in my opinion a fundamental error in your reasoning.
It is a very god thing that IETF's BCP (de facto standards, whose use is restricted to some forms of use) make use of ISO standards (de jure standards, that are designed for very large categories of users), but they must use them only for what they really are for and not twist these ISO standards for the sole advantage of Internet or Language industry, that is maybe a growing part of the world, but not all the world we live in.
 ISO basic standards like ISO 3166, ISO 639 and ISO 15924 have not IETF for unique users and the code they produce have certainly now information systems as principal use, but these are not at all the only possible and real ones (for example, ISO 3166-1 alpha-2 code elements were initially mostly taken from indications of nationality on license plates of vehicles). So, if information systems need creation of special code elements to code special scenarios, that certainly make sense in these restricted area,then they should systematically reserved user-assigned code elements that were created specially for such cases.
Gérard LANG 
-----Message d'origine-----
De : Peter Constable [mailto:petercon at] 
Envoyé : mardi 17 mars 2009 06:33
À : Lang Gérard; ietf-languages at
Objet : RE: Last call for ISO 15924-based updates

From: Lang Gérard [mailto:gerard.lang at] 

> May be "Zinh" is not directly a "sin", but it is at least a "trick", 
> and in all case it is far more out of the spirit of "Code for the 
> representation of names of script" that a proposition like "Zipa", for 
> International Phonetic Alphabet (IPA) that was refused by ISO 15924.

I have no qualms with your comment in relation to IPA, but it seems clear that you think "Zinh" is not in the spirit of the "Code for the representation of names of script", and here I very  much disagree: coding systems are created for use in information systems, and those systems need ways to code special scenarios. Sometimes, the scenario is very specialized and local, and a private-use code element is appropriate. But there may be cases in which public interchange is needed, or scenarios that aren't likely to involve public interchange but yet are very common in information systems, and in those cases it makes sense to include coded entities in the standard encoding.

ISO 15924 has two major consuming open industry specifications, BCP and Unicode, and the latter makes significant use of the notion encoded by Zinh and exposes it broadly to many information system implementations. For that reason, it is, IMO, completely appropriate and in the spirit of ISO 15924 to code Zinh.


More information about the Ietf-languages mailing list