Since 2000, IBM has been following a strategy of producing software that allows multilingual computing and provides multicultural support without the need for internationalization and localization. By conforming to standards such as Unicode and using open source technology such as International Components for Unicode (ICU) and the Common Locale Data Repository (CLDR) it is possible to create a single software solution that provides a culturally appropriate experience for users worldwide. In addition to providing multicultural support, IBM software is also translatable allowing the user interface to be translated into the languages that are most useful to IBM customers.
This approach to software globalization supersedes the older model of creating software which is 'internationalized' and then 'localized'. One of the disadvantages of the "internationalize and localize" model is that there is no guarantee that multilingual computing can be achieved. For example, software that has been internationalized and then localized for Japan should be able to support Japanese text, the Japanese calendar and currency and have a Japanese user interface (UI) etc, but a customer may or may not be able to work with Cyrillic or Chinese text with the "Japanese version". Moving to a multilingual model makes it possible for customers to work with multiple writing systems simultaneously. Providing multicultural support independently of UI translation enables customers to work in almost any region of the world and have the software work with their data and cultural preferences.
The old approach can be illustrated as follows:
and the IBM single software solution approach using language packs to provide UI translations as needed:
The terminology we use in IBM software globalization is as follows:
In computing, the provision of a single software solution that has (1) multicultural support and (2) a user interface and documentation that is available in one or more languages. See also multicultural support and translation
multicultural support (noun)
In computing, the ability of a single software solution to be translatable and to support the cultural conventions of multiple languages and geographic regions. Cultural conventions include the use of various writing systems, sort orders, different formats for date, time, numbers, and currency, different keyboard layouts, etc.
Additional IBM terminology.