Recipient Institution:
Open Forum for Cambodia

P.O.Box 177 Phnom Penh Cambodia

Tel/Fax: 855 12 415 734/855 23 360 345

Email: javier@khmeros.info

Project Leader:
Mr. Javier Solá

Amount and Duration: US$ 29,732 / 24 months

Commencement Date:
May 2004

Project Proposal
Facilitation of Software Localization for those who are not directly involved in the Open Source movement.

Project background and justification
In order to enter a digital world without forfeiting its culture, a country must do it by using software in its own language. Software in a foreign language exacerbates the digital divide, makes basic computers training difficult and expensive, impoverishes local culture, and blocks computer-based government processes, as local language scripts cannot be used in databases.
The Open Source movement has most of the applications that 80% of the users will ever need, and has also developed mechanisms to "easily" localize (translate) these applications, but... living in a world of computers, the members of the Open Source Community sometimes fail to see what is easy and what is not for those who are not used to work with Open Source software. It takes several months, for a trained computer scientist, to understand the mechanics and the philosophy of Open Source software and the Open Source Software community.
Figuring out how to develop a localization project, locating the pieces of software that need to be translated, finding what are the steps that need to be followed to translate the software and, finally, discovering how to come out with applications in a local language... is a long learning process which is repeated once again by every localization project in every country, multiplying the effort, and pulling again and again from the same multiple sources of information, when they find them. As the expertise acquired by each one of these countries or projects is not written, it is lost. New projects have to start from zero.
Several countries in the Asia/Pacific region are at this time considering starting localization projects, but given the size and industrialization level of the countries, they lack the human resources to start these projects without outside expert advice. Simple but extensive centralized information on the particular issues of localization, addressed to non-OpenSource-experts, including how-to manuals, is a must for these countries to be able to attack a localization project.

Project Objective
The Open Source Software Localization Toolkit project objective is to reduce as much as possible the necessary research, work and expertise that a country or minority needs in order to undertake a localization project.
The development of an Open Source Localization Toolkit, formed by a Manual and a CD. The Manual will consist of a series of "How to" sub-manuals that include everything that 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 the most technical parts, including installation of translation tools, finding and processing the files that need to be translated and building (physically creating) applications in local language. The toolkit will be based on the real work and experience of a localization team.
The first version of the toolkit will be made available six months after start-up. The final one will be delivered two years after the project starts.

Beneficiaries
All countries in the Asia/Pacific region, and mainly those with the least available technical human resources for localization.
The Toolkit project is specially addressed to small countries that do not have the necessary expertise and knowledge to start a localization project by themselves, as well to national minorities that use their own script or language and wish to use for its people to enter the digital age.
Nevertheless, we believe that the information contained in the Toolkit will be of use to all Asia/Pacific localization projects.
Project sustainability
The Toolkit is a tool for other projects. What is really important is that such localizations project care about sustainability (and therefore the Manual should talk about it and ask users to consider it in the project). Sustainability will be assured if the Manual helps create local expertise in the different countries it is used.
Nevertheless, it is important to understand that the Manual will stay quite up-to-date for at least a couple of years after termination, as it will not only address versions of the software that are current at that time, but it will also point at the websites of the different OpenSource projects, and give instructions on how to look and use the new versions... but again, new software comes out, build forms change... and at some point the most technical parts of the Manual will be outdated. The CD will be outdated in one year.

Project methodology
The Toolkit project will be developed in parallel with the Khmer (Cambodia) localization project being coordinated by Open Forum of Cambodia. The Cambodian project will be used to assure that the information being included in the Toolkit is correct, the manual being written before the actual Cambodian implementation takes place.
To avoid getting carried away by Open Source terminology and know-how, the project will include a computer scientist that has never worked on Open Source, under the guidance of the Cambodian localization project manager, and will be implemented in the Cambodian project by a different computer scientist that will only have this document as guidance.
Information will be collected from Open Source communities, on-going localization project and other sources, and then translated into extensive, simple, how-to sub-manuals addressed to computer scientists with low level of training, to translators who need to know where to find and how to translate files, and to a project manager.
All information about the different Open Source programs will be checked with the members of the communities that are in charge of the given program.
If possible, the Manual will also be tested with a second localization project that will help monitoring and assuring the correctness of the methodology.

Project time-line
A final deliverable is planned two years after the project starts (may 2006), but given the pressing need for this type of Toolkit in the Asia/Pacific region at this time, a first version will be released six months after the project starts (december 04). From that moment on, this first version will be updated bi-monthly and made public in the web.
Project output
As mentioned above, the project will output several versions of the Toolkit, each one including:
1) The Manual. It will include instructions on how to:

• Prepare a localization project.
  • Justification for the Project.
  • Choosing the software to be included in a complete final localized system.
  • Preparing a project plan (roadmap).
  • Sustainability
  • Preparing a project budget
• Find out if the language script has already been included in the Unicode standard. How to have it included if it is not. Why this is important.
• Find out if there are advanced fonts in the target language (TrueType or OpenType Unicode fonts). Pointers to resources related to Unicode, font development and to language specific typographic information. Pointers to companies that may develop such fonts if they don't exist.
• Find out if there is support for the target language script in MS Windows and in different Open Source platforms and applications. Explanations on how to get support, either through the open source community itself or through private people or companies that may do such work on a for-profit basis.
• Develop a Glossary of computer terms in the target language, as well as translation style directives.
• Develop a LOCALE for the language and the country, including all local cultural information, such as date formats, number formats, etc...
• Choose, [purchase in necessary,] and install translation and editing software that is adequate for the script of the target Language.
• Choose, [purchase in necessary,] and install a Translation Memory system. Translation Memory allows easy re-use of sentences or parts of sentences that have already been translated.
• For each one of the applications included in the toolkit: 1) Explain how to acquire the files that need to be translated. 2) Detail how to prepare them for translation and 3) Clarify how to edit them.
• Again, for each one of the applications: 1) Explain how to set up the environment for building the application, including compilers, systems variables, sources, etc... 2) Explain how to build the application with the menus in the local language, and 3) Detail how to prepare it for distribution.
• List Manuals that it is interesting to translate for each application, including help files if any.
• List Open Source training materials that may be translated as support for the translated applications. List non-free materials that may be translated after negotiation with the owners.
• Explain how to develop a deployment plan, taking into consideration hardware and software vendors, computer training schools, etc.

2) The CD (or set of CDs), including the following digital information:

• Open Source programs and libraries needed to set up the environments to build the applications in the target language.
• Last-stable-release versions of the sources of target applications.
• Separately, the files that need to be translated for the above applications, so that they are easily findable by non computer-experts and can be translated before computer scientists intervene to build the applications in the target language.
• Open Source training materials that may be translated.

Project monitoring
The project output is very clear, and can be easily monitored after the first release takes place six months after the project starts.
In order to assure the quality of the Toolkit, it could be interesting to link to this project a second localization project (under its own control and budget) independent from the one used for development. The results of this second project would not only allow measuring the quality of the Toolkit, but would also lead to improvements in the Manual. This second localization project would get direct help from the Toolkit project team. All help requested and given would be recorded as improvements needed in the Toolkit.

Project applicant
The Open Forum of Cambodia is a local Cambodian NGO established in early 1995 with the aim of providing opportunities and resources for Cambodian citizens to develop new skills and insights into policy questions in order to provide timely and structured contributions to decision-making in social, political, cultural and economic spheres of life in Cambodia and to use their acquired abilities to help build a stronger civil society.
In order to fulfil some of these goals, Open Forum has played a pioneer role in the introduction of Internet e-mail in Cambodia, as well as in establishing the .kh Internet domain. It has also helped Cambodians send e-mail in Khmer and coordinated interconnection of Cambodian ISPs.
The launching of the KhmerOS (Khmer Software) Initiative has been the next step of a long process within Open Forum to provide the Cambodian society with the necessary tools to advance in technology at the same time that they maintain their own culture and language.

Project staff
The Toolkit project will be managed by Javier Solá, a Spanish computer scientist with over 20 years of experience who is now a resident of Cambodia, coordinating also the Khmer Software Initiative.
Javier Solá (Spanish, 43 years old) holds a BSc degree from Duke University (1984, North Carolina, USA) and a Masters Degree in Computer and Information Sciences from Ohio State University (1987, Ohio, USA).
His last job in Spain has been as Director of the Spanish Internet Users Association, where for seven years- has pushed the development of Internet in Spain through participation on Internet related legislation and coordination of competing telecommunication companies to work and compete under better conditions, helping them to offer better conditions to the users. His work also included the creation of a non-profit Internet user support structure, and the dinamization of the market through conferences and trade-shows.
Part of this job was his participation -from 1997 to 1999- on the creation of ICANN, where he coordinated the creation of the Names Council and chaired the Working Group that decided the creation of new international Internet domains (such as .info). He was also VP of ISOC for Conferences in 1998 and 1999.
Prior to his work around the Internet, Javier Solá has worked for four years (1984-1985 and 1988-1989) in Paris, France, as a researcher on Artificial Intelligence for BULL S.A., France's largest computer company. From 1989 to 1995 he worked in Spain as Director of a Software Development company that did strategic and software development work for Spanish corporations as well as for the European Commission's body that specialises on vocational training.
He has taught at seven different universities, including three Spanish universities, the Ecole Central and Ecole des Arts et Metiers in Paris (two of France's most prestigious universities), Ohio State University and -finally- Boston University's MBA program in Spain, where he taught "Strategic Impact of Information Technology".
For the last year he has lived in Cambodia, working on cooperation projects. He speaks fluently Spanish, English and French, and has a basic knowledge of Khmer.