Grant Amount: US$ 29,732

Keywords: FREE AND OPEN SOURCE, SOFTWARE, LOCALIZATION, CAMBODIA, ASIA-PACIFIC

Geographic coverage: Cambodia, Asia-Pacific

Objective

The objective of this project is to reduce the necessary research, work and expertise that a country or group needs to undertake a localization project.

Research context

This project addresses the localization of software for countries in the Asia-Pacific region that lack the human and technical resources required to undertake these complex projects. The project is developing a toolkit in layman’s language to allow countries to develop localization projects without the need of specialized help. The toolkit is being developed over a two-year period, with preliminary releases made public during that time.

Target beneficiaries

The project benefits all countries in the Asia-Pacific region, particularly small countries and national minorities that use their own script but do not have the necessary expertise and knowledge to undertake a localization project.

Outputs

The development of a Free and Open Source Software Localization Toolkit, including a manual and a compact disk (CD). The manual consists of a series of how-to sub-manuals that include everything a prospective localizer needs to learn: from management issues such as “how to decide what you need” and “how to write up a localization project” to more technical issues, including installation of translation tools, finding and processing the files that need to be translated and finally building applications in a local language.

Research results and outcomes

For countries or regions that do not have any computer programs in their own language, free and open source software (FOSS) can be translated and localized (adjusted for local standards) making it possible to create a complete set of free programs in a local language. The main barrier localized software removes for local people, is the requirement of learning a new language such as English, prior to learning to use computer applications.

The problem is that the FOSS community is sometimes highly technical and the procedures to be followed to localize text are not well documented. Gathering the necessary information to start a complete localization project, understanding the procedures, evaluating the necessary resources and preparing a plan can take up to six months for a trained computer technician. To address this, the project was first conceived as an umbrella project that would gather information about different FOSS projects and document their localization methods and integration.

Once the project was started however, initial findings made a change in methodology necessary. These include the following. Either little or no localization information was available for key programs, or they were highly technical, which was not appropriate for the toolkit. In many cases the tools for easy technology-independent translation of FOSS did not exist, or were not defined. Finally, the researches found there was little written on how to develop a localization program. In response to these findings, the project considered that it could best reach its goals by defining a translation system for FOSS, thus simplifying the process and helping small economies reach their localization goals much faster.

The project found that coordinated localization projects could be improved by using the same programs and the same format. The South Africa-based Translate Toolkit project does this by converting the files that need to be translated into a standard format and converts them back to the application’s format once the translation is done. However, the project felt that the Translate Toolkit had insufficient documentation. Thus, developing reference documentation of the Translate Toolkit software became the first technical task completed by the project.

As a follow up to this, it was necessary to link (through documentation) the Translate Toolkit to other major FOSS projects. The OpenOffice.org project, started using the Translate Toolkit and the localization documentation developed by the FOSS Localization Toolkit. This has simplified the localization procedure for this product.

The FOSS Localization Toolkit is further developing and testing its documentation through a south-south collaboration (Cambodia/South Africa) project called WordForge (http://www.wordforge.org). The goal of WordForge is to use the latest computer-assisted translation technology to produce the best possible localization tools for FOSS and contents. Where the FOSS Localization Toolkit has worked on documentation to facilitate the work, WordForge goes one step further to create easy-to-use tools that continue the work of facilitating localization. Translators that are not highly skilled can use these tools.

A key document produced by the project is a detailed manual on how to plan and write a localization project for a given country and language. It covers all aspects of the project, from defining its social goals to understanding the workload and preparing a plan and a budget. Documentation from the project is already being used by localization projects in Asia (Bhutan, Lao PDR, Nepal).

Duration

Start Date: August 2004
End Date: July 2006
Total Duration: 24 Months

Contact information

Mr. Javier Solá, Coordinator
Khmer Software Initiative (KhmerOS)
Open Forum for Cambodia
PO Box 177, Phnom Penh, Cambodia
Telephone: +855 23 360 345
Email: coordinator@khmeros.info

Website: http://www.forum.org.kh

Reference website

http://www.translate.org.za
http://translate.sourceforge.net/wiki
http://www.wordforge.org