- From: Addison Phillips <addison@yahoo-inc.com>
- Date: Tue, 2 May 2006 20:03:21 -0700
- To: "'Baden Hughes'" <badenh@csse.unimelb.edu.au>, <www-i18n-comments@w3.org>
RFC 3066bis was developed with the participation and active consultation with the ISO 639-2 and ISO 639-3 standards folks and contains a number of features related to the eventual adoption of ISO 639-3 and the future direction of ISO 639 in general. The reason 3066 and 3066bis do not reference 693-3 today is that it isn't officially a standard yet. Standards that adopt ISO 639-3 before it is officially complete are open to the instability that can affect draft standards. > It seems to me that in the context of LTLI, we > should use the > finer grained standard, minimally as a supported option (note this is not > in RFC 3066/RFC 3066bis) and arguably from a linguistic standpoint by > default. Actually, this is 3066bis, provided that the registry is updated with the additional subtags when that standard is published. That is currently the plan and I don't foresee any roadblocks to adoption of 639-3 at an appropriate time later this year. Multiple standards for language tagging are a bad idea, in my opinion, since they will lead to fragmentation of tagging and the confusion of users. Addison Addison Phillips Internationalization Architect - Yahoo! Inc. Internationalization is an architecture. It is not a feature. > -----Original Message----- > From: www-i18n-comments-request@w3.org [mailto:www-i18n-comments- > request@w3.org] On Behalf Of Baden Hughes > Sent: 2006?5?2? 17:41 > To: www-i18n-comments@w3.org > Subject: Comment on LTLI WD > > > > Hi > > One issue which concerns me is the choice of RFC 3066 (ISO639-2) as the > basis for defining language values, particularly in the context where a > new ISO > standard (ISO639-3) will be shortly adopted [1]. ISO639-3 extends > ISO/DIS 639-2 to cover all known languages and represents the merging of > the two internationally authoritative sources on language classifications > (the Ethnologue [2] and the LinguistList [3]), and is far more fine > grained than ISO639-2. It seems to me that in the context of LTLI, we > should use the > finer grained standard, minimally as a supported option (note this is not > in RFC 3066/RFC 3066bis) and arguably from a linguistic standpoint by > default. It is interesting to note that other large communities are > actively considering > the use of ISO639-3 as a preference (eg Dublin Core). > > Regards > > Baden > > [1] http://www.sil.org/iso639-3 > [2] http://www.ethnologue.com > [3] http://www.linguistlist.org
Received on Wednesday, 3 May 2006 03:04:09 UTC