Warning: include_once(cookbook/pmfeed.php): failed to open stream: No such file or directory in /misc/32/000/115/128/8/user/web/bisharat.net/wikidoc/local/config.php on line 63

Warning: include_once(): Failed opening 'cookbook/pmfeed.php' for inclusion (include_path='.:/usr/share/pear') in /misc/32/000/115/128/8/user/web/bisharat.net/wikidoc/local/config.php on line 63
Pan African Research on L10N Workshop and Localization "Blitz" | Marrakech / Background
Recent Changes - Search:
Home

Background

Agenda

Locales

Venue

Participants

Working Group

Resources

Organisers

Photos

FAQ


RSS

Background

Localisation is a broad process of transferring knowledge between communities that are representing different values, cultures and languages. In the course of this proposal we use the basic definition of localisation (l10n) developed by the Localization Industry Standards Association that says: it is a process of adapting text and cultural content to specific target audiences in specific locations. The process of localisation is much broader than just the linguistic process of translation. Cultural, content and technical issues must also be taken into account. Localisation involves taking a product and making it linguistically and culturally appropriate to the target locale (country/region and language) where it will be used and sold.

Africa is one particular example for the importance of localisation. Across the continent post colonial dominance of English, French Arabic and Portuguese continues to overshadow the fact that at least 1000 distinct African languages have been identified (not including dialects, which would double that number), of which more than 800 are spoken today. African languages can be largely divided into four main groups: Afro-Asiatic, Niger-Congo, Nilo-Saharan, and Khoisan and they reflect the great ethnic diversity of this continent. Yet the majority of computer software interfaces and content are available in very few languages only, and a tiny subset, less than 20 according to one expert, of African languages is represented by content on the world wide web.

In the context of l10n, language itself is the major issue. While many consider English or French and Portuguese (or even Arabic) universal languages in the African context, findings of computer users elsewhere show the limitations of this perspective. In Sweden and Denmark, countries with a high degree of English proficiency, 64% and 63%, respectively, of users prefer software in their own language. On the global scale over 100 million people access the Internet in a language other than English. As for 2005 over 50% of Web users speak a native language other than English (http://www.glreach.com(approve sites)). As Internet access in developing and transition countries is increasing the importance of today's dominant languages is expected to further decrease.

In addition, language influences users' perception of technology and their interaction with it. Web users are four times more likely to purchase from a site that communicates in the customer's language (http://www.idc.com(approve sites)) and stay for twice as long -site stickiness is doubled. (Forrester Research).

In the discourse of globalisation and the role of ICTs in fostering economic agglomeration localisation is often omitted as parallel process where local/regional economy becomes more important to individuals and businesses operating locally. (Richard G. Harris –Economic Approaches to Language and Bilingualism in New Canadian Perspectives) In that regard forcing majority languages as primary communication vehicle on indigenous communities is working against economical principles and common sense.

The current status of localisation in Africa

A number of projects are working on l10n issues in Africa, including the Bisharat project "PanAfrican Localisation," maintained by Kabissa (and funded by IDRC's Acacia program) whose goal is the mapping of localisation issues, documentation of localisation trends on the continent and collection of best practices for localisation processes. Additional groups are promoting l10n in more practical ways: Translate.org.za is leading a localisation framework project that develops distributed translation tools; a group supported by Open Society Initiative West Africa is working on a localisation framework for Western Africa and also includes Translate.org.za. In addition to these larger projects, there is vast number of smaller local projects, including:

  • The Ethiopian Ge'ez Frontier Foundation's work on software support for Ethiopian languages
  • The Ethiopian Computer Science department at the Addis Ababa University (AAU), which has localised the OpenCMS content management system for the regional administrative offices to manage web content on the WoredaNet?. Handwriting recognition algorithms, which can identify Ethiopic characters with 99.4% accuracy, have been developed with an eye towards PDA support.
  • The African Virtual Open Initiatives and Resource (Avoir) initiative brings together a network of 13 African Universities, spanning regional and language boundaries. The AVOIR partners have focused on software development, but are working with local partners to localize their learning management system and are exploring l10n needs especially in the area of open courseware.
  • POLLES: Plate-forme de logiciels libres pour l'enseignement supérieur or Open Source software for North African Universities. This project aims to develop or adapt a series of localized software in French and Arabic for use by the maghrebine universities. Partners come from Algeria, Morocco, Tunisia and Canada.
  • More recently, the University of South Africa (Unisa), and the University of Limerick's Localisation Research Centre (LRC) announce their intention to establish closer links between the two institutions to promote internationalisation and localisation-related research and teaching. http://www.localisation.ie/resources/LRCNews/2006/July.htm

In addition there is at least one notable commercial localisation initiative underway. Mark Shuttleworth's London-based Canonical, Ltd. is working with African governments to create localized versions of the Ubuntu Linux distribution.

Key l10n issues in Africa

Regardless of the promising efforts a number of key problems remain. While working in Africa (organising both Africa Source I and Africa Source II in Namibia and Uganda respectively) TTC together with our African partners developed localisation curricula appropriate to the continents' context. These curricula were further enhanced during the localisation sprint we hosted in Warsaw in 2004. During this work a number of major problems faced by l10n in Africa could be recognised:

  • Very little work is put towards integrating isolated localisation efforts on a regional scale
  • There is an extensive gap between development and adaptation, between people involved in localisation and the broader group of users and supporters, which also includes policy makers
  • The aspect of community building is inter-connected with the solutions to these problems. As observed in other countries or regions, for successful adoption of localisation a critical mass of computer geeks, translators, developers, academics, policy advocates and users needs to be established. One of the most advanced examples of that case would be the Regional Initiative to Develop Local Language Computing Capacity in Asia (http://www.panl10n.net/) or work done by Spanish Extremadura project building communities across South East Europe.

See also

Research on the current situation and potential of localization in Africa for the PanAfrican Localisation project has produced two documents, one a more extensive survey, and the second being a brief overview and comparison with Asia:

Page last modified on 2007-09-08 15:42