Localization is based around the concept of locales. A locale is a set of conventions, usually associated with a particular country, language, and/or culture, that specifies how information should be presented.

These conventions include such things as:

• The country normally associated with the locale
  
• The language normally associated with the locale
  
• The default unit of currency
  
• The standard number format
  
• The standard time and date format
  
• The standard measurement system
  
• The default paper size
  
• The default codepage(s) used for character processing

You can use locale information to determine the behaviour of your applications as appropriate for the current environment. In addition, several ULS functions take locales as input (for instance, to determine collation or case conversion rules).

There are two types of locale, both of which may be used in this way.

• System locales are standard locales defined by OS/2. They are constant, and cannot be modified. System locales exist for every country and language known to OS/2, as well as for abstract conventions like "Posix", "C", or "UNIV". They are typically used to obtain generic information about a particular locale.

• User locales are locale instances which can be customized by the user. Unlike system locales, user locales can freely be created, modified, or deleted. All locale objects which appear in the OS/2 Locale (Country Palette) applet, including the default desktop locale, are user locales. They are normally derived from system locales, but may be tailored according to the user's own preferences.