Lang Gérard <gerard dot lang at insee dot fr> wrote:
On 29-03-2004, ISO 3166-1 Newsletter V-11 decided to create inside ISO 3166-1 the three new entries GUERNSEY (GG, GGY, 831), ISLE of MAN (IM, IMN, 833) and JERSEY (JE, JEY, 832) and suppressed any remark concerning GB. At the same time UNSD, that had created the numeric-3 code elements 831 and 832 at ISO 3166/MA request, added Guernsey and Jersey as two new entries inside M49 (but did not suppress the entry Channel Islands, as should, or could, have been the case). The sixth edition of ISO 3166 (second edition of ISO 3166-1), published on 15-11-2006, integrated the Newsletter V-11.
We were there. A fair amount of discussion during LTRU 1.0 revolved around the existence of four M.49 code elements that did not exist in 3166-1. It was considered important to have a way to distinguish language X (English? French? Dgèrnésiais?) as spoken in Guernsey from language X as spoken in the rest of the UK. Numeric region subtags based on the M.49 code elements were our solution.
The introduction of the 3166-1 code elements GG and JE (and IM) made the original passage in RFC 4646 look very awkward. The continued existence in M.49 of a "country" code element for "Channel Islands" as well as "Guernsey" and "Jersey" individually also adds a bit of weirdness.
2-Coherence: if it is correct that Channel Islands is today the only entry of M49 that is not an entry of ISO 3166-1, it must also be remarked that some entries of ISO 3166-1, like BV or HM, are no more today's entries for M49.
But we aren't worried about this. In the Language Subtag Registry, alpha-2 code elements based on 3166-1 take precedence over numeric code elements based on M.49.
Moreover, M49 added to the section concerning "countries and territories" another section concerning macroregions and giving, on one side a list of geographical macro-regions (continents, subcontinents and continental regions, that is considered inside RFC 4646 (and also indirectly inside ISO 3166-1 because, as application of the clauses 8.1.2 and 8.1.3 concerning user-assigned use of reserved code elements for aggregations of countries, ISO 3166/MA recommands alignment for such use with a notified assignation) and, on another side a list of economic grouping of countries (like developped, developping, least developped, landlocked, ..) that is not used by ISO 3166 or RFC 4646. .
Because the developed/developing and island/landlocked distinctions have no conceivable relevance to language identification, whereas the geographical groupings do have relevance. In particular, there is an established need for some applications to distinguish "Latin American Spanish" from Castilian as spoken in Spain, without breaking the former into Colombian Spanish, Venezuelan Spanish, Argentinian Spanish, etc. There are other examples.
3-Terminology: In the common terminology of ISO 3166 and ISO 639, a code is a table making a link between on one side a list of entries (Country names, Language names) and a corresponding list of code elements, built by use of a coding scheme defined by a syntactic format (like alpha-2, alpha-3 or numeric-3) and maybe other rules (like a visual association between the alphabetic code elements and the corresponding entry names). So that "830" is a "code element" and should not be considered as a "code".
This is a useful distinction which is lost on many English speakers who don't work with ISO standards. The English term "code element" is unwieldy and it is extremely common for English speakers to speak of the individual elements as "codes." French benefits by having the two very short and convenient words "code" and "codet."
4-Anyway, in my opinion, this text "E" could very well be suppressed.
Glad we agree. -- Doug Ewell * Thornton, Colorado, USA * RFC 4645 * UTN #14 http://www.ewellic.org http://www1.ietf.org/html.charters/ltru-charter.html http://www.alvestrand.no/mailman/listinfo/ietf-languages ˆ
Note Well: Messages sent to this mailing list are the opinions of the senders and do not imply endorsement by the IETF.