Locale help

Locale explains how people can translate this project into worldwide languages.

Each locale subdirectory must be named using a standard locale code.

Examples:

- `locales/en-us/` is for English - United States regional dialect

- `locales/en-gb/` is for English - Great Britain regional dialect

- `locales/zh-cn/` is for Chinese - China mainland dialect with simplified Mandarin

- `locales/zh-tw/` is for Chinese - Taiwan island dialect with traditional Mandarin

List of BCP-47 language tags:

• https://developer.mozilla.org/en-US/docs/Glossary/BCP_47_language_tag

List of ISO 639-1 codes:

• https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes

Content subdirectory names must be translated into their corresponding languages.

Examples:

• locales/en-us/hello/

• locales/zh-cn/你好/

Content subdirectory names must use a slug format, which means a format that is user-friendly for reading, copying, pasting, typing, browsing, and searching:

Example:

• locales/en-us/hello-world

• locales/zh-cn/你好-世界

You have options for the slug format:

• Our projects use a slug format for English that uses alphanumeric lowercase characters, with words separated by a hyphen-minus character, and no other characters.

• Your projects may use whatever you prefer; for example Wikipedia uses alphanumeric mixed-case characters, with words separated by an underscore character, and other characters such as parentheses.

This project keeps track of corresponding content across multiple locales by using the computer science technique named "locale peer id".

A locale peer id is a secure randomly-generated 32-character lowercase hexadecimal string.

For example, b82232cff8e5bb785fbc855b65a83653.

Each directory contains a file named .locale-peer-id that contains the locale peer id then a newline.

When anyone translates any project directory from one language into another, then the locale peer id is identical.

For example each directory below contains a file named .locale-peer-id and each file contains the identical locale peer id:

locales/
  en-us/
    hello-world/
      .locale-peer-id - b82232cff8e5bb785fbc855b65a83653 then newline
      index.html - translation with English United States dialect
  zh-cn/
    你好-世界/
      .locale-peer-id - contains b82232cff8e5bb785fbc855b65a83653 then newline
      index.html - translation with Chinese China mainland dialect

To create a translation of a directory and its contents, you must use the same locale peer id and file name .locale-peer-id, because this consistency is what enables the project team to track directory translations among languages.

If you are a programmer, then see these tools which may help you:

• bin/locale-peer-id: generate a locale peer id

• bin/grep-locale-peer-id: find files that contain a given locale peer id

• bin/markdown-read-to-headline: markdown tool to read a file or stdin, then print the first headline

• bin/slug-case: input a line of text; output the text converted to slug case

A locale can have a region, such as:

• locales/es/ is for Spanish - officially doesn't have a region

• locales/hi/ is for Hindi - officially doesn't have a region

A locale can have an unspecified region such as:

• locales/en/ is for English (unspecified region) - better to be specific en-us or en-gb.

• locales/zh/ is for Chinese (unspecified region) - better to be specific zh-cn or zh-tw.

For worldwide standards, international organizations, cross-border communications, and the like, we prefer we currently prefer using the Oxford locale:

• locales/en-gb-oxendict/ is for English - Great Britain - Oxford English dictionary

For example, the World Health Organization (WHO) and United Nations (UN) use this locale.

en-001 replaces en-US as the generic fallback.

Historically, many applications used American English as the default global fallback, which forced everyone to use MM/DD/YYYY dates and imperial measurements. en-001 solves this.

Defined in CLDR as equivalent to “English (United States)” except:

Almost equivalent to “English (United States)”, except for some locale and formatting differences including these:

• Date format DD/MM/YYYY instead of MM/DD/YYYY.
• First day of the week is Monday instead of Sunday.
• Currency symbol is US$ instead of $.
• Use metric measurements.

en-150 replaces en-US as the Europe fallback.

Historically, millions of Europeans used American English as a professional lingua franca but are deeply accustomed to European standards like the 24-hour clock and comma-separated numbers.

Defined in CLDR as equivalent to “English (World)” except:

• For decimal separator, use comma instead of period.
• For time format, use 24-hour instead of 12-hour
• For currency symbol, use ¤ of US$.

For our major worldwide projects, we consider these locales.

The United Nations languages:

• ar-001: Arabic (World)
• en-001: English (World)
• es-001: Spanish (World)
• fr-001: French (World)
• ru-001: Russian (World)
• zh-001: Chinese (World)

The locales where we receive funding:

• en-us English (United States)
• en-gb English (Great Britain)
• cy-gb Cymraeg (Great Britain)

The top 10 most-spoken languages in order:

• en-001: English: ~1.49 billion total speakers (372M native + 1.12B non-native)
• zh-001: Mandarin Chinese: ~1.18 billion total speakers (988M native + 194M non-native)
• hi-001: Hindi: ~611 million total speakers (347M native + 264M non-native)
• es-001: Spanish: ~561 million total speakers (487M native + 75M non-native)
• ar-001: Modern Standard Arabic: ~335 million total speakers (mostly second-language or formal use)
• fr-001: French: ~334 million total speakers (75M native + 258M non-native)
• bn-001: Bengali: ~274 million total speakers (234M native + 43M non-native)
• pt-001: Portuguese: ~269 million total speakers (252M native + 18M non-native)
• id-001: Indonesian: ~255 million total speakers (78M native + 177M non-native)
• ur-001: Urdu: ~246 million total speakers (78M native + 168M non-native)