• Language maintainers
• List of state/provinces
• Language codes

Language maintainers

Currently our policy is that the first person willing to translate CiviCRM into a given language creates a Transifex team and becomes the team coordinator; future translators need to be approved by the coordinator.

This makes it possible to guard from spammers and keep a given localisation coherent (e.g. with regards to how different terms, like ‘househould’, should be translated), but it also means that new translators who are not quickly accepted (or denied) by the team coordinator are in limbo and wait for the decision.

Thus we should periodically check whether new translators for existing teams are handled pronto and are not discouraged/blocked when the team coordinator is inactive (or on a leave, etc.); we should poke the coordinators, elect additional ones or, in the worst cases, accept the new translators ourselves.

http://help.transifex.net/features/client/index.html#auto-local

List of state/provinces

We follow ISO-3166-2

Language codes

Language codes are written using the same syntax as linux/libc/gettext locales:

language[_country][.charset]

.. where the charset is assumed to be utf8, and therefore omitted.

Language and country codes tend to follow ISO-3166-1.

However, since we do not have a translation for every country where a language is spoken, and since we tend to avoid creating regional translations (because it's a ton of work), there are some languages that are identified as linked to a specific country, or using a specific charset, only because that's how translators created them initially:

• zh_CN (Chinese simplified)
• zh_TW (Chinese traditional)
• sr_RS uses latin letters, equivalent to sr_RS@latin

See also: new official language.