Bamatic Blog

Some articles of the United Nations Declaration on the Rights of Indigenous Peoples

2008-06-18T18:42:00.000-07:00

Article 13

Indigenous peoples have the right to revitalize, use, develop and transmit to future generations their histories, languages, oral traditions, philosophies, writing systems and literatures, and to designate and retain their own names for communities, places and persons.
States shall take effective measures to ensure that this right is protected and also to ensure that indigenous peoples can understand and be understood in political, legal and administrative proceedings, where necessary through the provision of interpretation or by other appropriate means.

Article 14

Indigenous peoples have the right to establish and control their educational systems and institutions providing education in their own languages, in a manner appropriate to their cultural methods of teaching and learning.
Indigenous individuals, particularly children, have the right to all levels and forms of education of the State without discrimination.
States shall, in conjunction with indigenous peoples, take effective measures, in order for indigenous individuals, particularly children, including those living outside their communities, to have access, when possible, to an education in their own culture and provided in their own language.

Article 15

Indigenous peoples have the right to the dignity and diversity of their cultures, traditions, histories and aspirations which shall be appropriately reflected in education and public information.
States shall take effective measures, in consultation and cooperation with the indigenous peoples concerned, to combat prejudice and eliminate discrimination and to promote tolerance, understanding and good relations among indigenous peoples and all other segments of society.

Article 16

Indigenous peoples have the right to establish their own media in their own languages and to have access to all forms of non-indigenous media without discrimination.
States shall take effective measures to ensure that State-owned media duly reflect indigenous cultural diversity. States, without prejudice to ensuring full freedom of expression, should encourage privately owned media to adequately reflect indigenous cultural diversity.

Declaration

Cet article en Français Este artículo en español

Bamatic localization project

2008-06-14T19:53:00.001-07:00

When one tries to translate software into a language in which there aren't software translation, one has to face the development of terminology.

All sciences developed their own terminology and is also a work in progress, as new things, new concepts, new ideas are emerging, the new terminology unfolds at a time.

In the case of African languages, work in this field of development of terminology, is stalled. In addition, the few works that are performed, are done in isolation, without seeking a consensus among linguists, computer specialists and other experts.

To remedy this deficiency, the creation of a permanent channel of communication between all specialists, where discuss and develop their work is required. No less important than a media like this, is to have a way of dissemination and publication of the results, so that, the terminology developed serves as a reference to all those who want to translate software into African languages. Share the outcome of the investigations will facilitate the language translation work to people who might come after and interest in making translations. Keeping simple the translation process will involve more people. Due to the fact that the work to be done is enormous and those who we are involved in this very few, the priority, is for the moment, to involve public and private institutions, individuals, associations and NGOs in the localization.

I have decided to develop a website which will serve as communication between the members of the website, a forum for discussion, for creating new work teams, for publication of glossaries, and a way to to publicize final works.

The address of this site is PL10nB and I expect to have it 100% functional by the end of the month. All people interested in becoming a member of the network or in keeping in touch with me can do so via vzugadi@bamatic.com or visit occasionally PL10nB to see the progress.

The website should have these characteristics:

A forum for discussion among members.
A list of members of the network, available on line and with the profile of members.
A mailing list from wherel receive the news.
A Networking system for the translation of terminology on line.
A place for publication of the terminology developed to make it searchable online.
A multimedia gallery for the exhibition of localized programs and software developed by the members, screenshots, video demos ...
A Web control panel for administrative tasks.

Cet article en Français Este artículo en español

C, gettext, GTK+ and digits localization

2008-06-13T03:15:00.000-07:00

Trying to localizate software in Bambara language with N'Ko script,I have been concern about the digits translation issue. The N'Ko script uses own digits and not the latin ones. Therefore, when the translator finds a string containing a number, he can translate it, however, when the translator finds a string that contains one of the format specifiers of printf functions, it is not clear which strategy to adopt to localizate these numbers to natives N'Ko digits.

The strategy finally adopted, is to define new format specifiers, and use them for translation only, in the translation catalogues. For example, if we need to translate the string "%d" and we want to localizate it with N'Ko numbers, “%d” must be translated by "%D".

Another issue that arises is ,where we define the new format specifiers, one might think that would be better to define them at the level of the glibc or maybe it would be better to define them within each program code. The advantage of the first is that theoretically, should not modify the code of programmes, just need to distribute the modification of the glibc, which could be done within the support package for N'Ko script, along with the keyboard layout and fonts.

In reality we are faced with another problem, most developers do not mark all the stings that contains "%d" , or other numeric format specifier, as translatable strings and thus, modifying the program code to localizate it, is inevitable. So the second strategy, of defining the new format specifiers in the application code itself is more appropriate whenever we need to modify the code of the program or any of its components.

Here is a screenshot of gedit showing the line numbers with native digits:

Cet article en Français Este artículo en español

Writing systems

2008-06-11T15:33:00.000-07:00

Basically, computers only work with numbers. Store letters and other characters by assigning a number to each. There are many different coding systems to assign these numbers to the letters. The Unicode Consortium is a nonprofit organization created to develop, disseminate and promote the use of the Unicode standard, which specifies a number unique to each character and whose goal is to cover all the characters of all alphabets, mathematical and scientific symbols and other signs of common usage.

A glyph is the graphic representation of one letter, number, symbol or character. Depending on the style that the graphic designer gives to it, the same letter can be drawn slightly differently, taking a personal style.

A font is a particular set of characters, numbers and symbols, designed with the same style. Normally a source contains at least the full alphabet of a language, usually fonts contain alphabets of various languages, diacritical marks, punctuation marks, mathematical symbols, numbers and symbols, and in general anything that we wish use for writing.

A font is a computer file containing information of each glyph of the font, information about its size and on how it should be drawn that glyph.

Generally in all operating systems there is a component whose mission is to draw the glyph on the screen. That component known as rendering engine is the responsible for reading the glyph information contained in the records of the font, and draw the glyph on the screen.

Depending on the writing system, Latin, Arabic, Hebrew, N'Ko ... The rendering engine must have additional features that will let you do the right track. Some writing systems require cursive union between their characters, such as Arabic or N'Ko, besides the glyph in the same letter, will vary depending on whether the beginning of word or end or in the middle. The rendering engine must be prepared to handle these requirements or otherwise rendering will be poor.

In this picture we see the voice of ɔ N'Ko, written single and united to form the word b ɔ (leave), if we look at the base, we see that the voice ɔ N'Ko alone or forming a word aren't drawn equally.

As an example, we see in this picture, writing N'Ko with a rendering engine that does not support the union for N'Ko, and below after having added support to the rendering engine.

In addition to the font with our characters, a proper rendering engine, we need a keyboard to communicate with the computer and indicate which letter we want to write .

One piece of software called keyboard layout connects our keyboard with the glyphs. This keymap specifies which letter should be served by each letter pressed. There's really no need to change physically keyboard, enough to install the corresponding keymap, although it is much more comfortable if we have a keyboard where the letters are drawn on the keys. Therefore, a good way to introduce a new writing system is:

1. Ask the unicode consortium takes over the writing and assigned codes unicode.

2. Create sources containing all lso glyphs of the alphabet, digits and symbols of punctuation.

3. Giving support to the rendering engine to take charge of the characteristics of our writing.

Cet article en français Este artículo en español

Localization and internationalization

2008-06-11T06:13:00.000-07:00

The localization is the process by which a computer program is adapted to the language and customs of a particular region. The localization process is much easier if the developer of the programme takes into account that other people might want to localizate your program.

The most used option is the use of translation catalogues. When the program needs to show a text, instead of displaying the text like it appears in the program code, it looks for the translation of that text, in the translation catalogue of the user's language and so the text that will be shown is the translated text .

This operating method requires two things, one, the developer has to oblige the program to seek translations in the catalogues of the user's language and two, the translations catalog must exist in the user's computer.

The process during the manufacturing stage of the program, which instructs the program to use translation catalogues, is part of internationalization. A software running under this scheme, is said to be an internationalized software, as it can be localized to any language with only translate the catalog.

The process of translating catalogues is the process of localization.

Thus the translation process can be separated into two: the technical and the linguistic part.

When the user changes the language of the operating system, all programs use the translations catalogues of that language, and the system and programmes, will automatically appear in that language. To make this possible, in the operating system must be installed a database for each language in which it is hoped that the system can operate. This database is the locale. Usually, to perform a localization as much as possible adapted to the uses and customs of the place, it's usul to create locales as a combination of language-country. So we could have a locale for the Spanish from Spain and other for the Spanish of Mexico, in order to adapt programs more specifically, with the most common terms used in each area.

Este artículo en español
Cet article en Français

About the role of national languages in Mali

2008-06-10T17:49:00.000-07:00

The other day I had the pleasure of attending an event, at gmatic's conference room, Ministry of Communication and new technologies, about software localization in African languages.

The workshop, chaired by Don Adama Samassékou, President - Executive
Secretary of the African Academy of Languages, ACALAN, brought together Malien actors involved in the development of national languages, both public administration, and private, civil associations and NGOs to share their experiences with two experts.

After the presentations by the participants and the two invited as experts, the questions and discussions began.

This initiative has put many points clear, one, efforts by government agencies as well as by private, in the domain of software localization and the promotion of national languages has been virtually non-existent, but at the same time, I have noticed a different attitude, and very positive, towards the development of national languages. Another point that has become apparent is the need for a partnership, a sharing of knowledge and a common effort of all translators, linguists, developers, public and private agencies involved in this problem.

This time yes, I think, that everyone is aware of the importance and vital role that national languages must play for the country's development, a genuine participatory democracy and people's rights.

Cet article en Français
Este artículo en español

Personal information

2008-06-09T12:58:00.000-07:00

my personal page on yahoo 360°

my bloglog profile

Este blog en castellano

Ce blog en français

Some of the projects i am trying to mantain and developpe, all of them around software and African languages issues...

PL10nB is an african localization network.

Some pics of our works

Our new association

Fosebe : Educational typing software