Locales are primarily identified by their base language. For example, English [en], Arabic [ar] or German [de]. We also label scripts explicitly, where a language is typically written in multiple scripts, such as Cyrillic or Latin. For example, Serbian (Cyrillic) [sr_Cyrl] and Serbian (Latin) [sr_Latn]. Each language + script combination is treated as a unit. (i.e. People do not mix different script in the same data set.) If a language is not typically written in multiple scripts, then the script sub-tag is omitted. For example, en_US or ko_KR. Locales may also have regional variants. For example, English (US) [en_US] vs English (UK) [en_GB], or Serbian (Cyrillic, Montenegro) [sr_Cyrl_ME] vs Serbian (Cyrillic, Serbia) [sr_Cyrl_RS]. Regions may be countries such as China [CN], parts of countries such as Hong Kong [HK] or multi-country regions such as Latin America [419]. Also see Regional Variants. The contents for the base language should be as widely usable (neutral) as possible, but must be usable without modification for its default content locale; this is the locale for the language’s default region, which is typically the region with the most speakers of the language. A default content locale has no data other than identity information, it inherits all data from its parent. For example:
Tips for linguists:
|