Sahana Internationalisation
Languages and beyond
Dominic König, RN
Sweden/Germany
www.nursix.org

Introduction
Use Sahana worldwide?
↓
Translate the User Interface!
Translations of
Resource Files
2

Introduction
Internationalisation: i18n
To design the software in a way that allows adaption
to different languages and regional differences
without having to rework the Code.
Localisation: L10N
To adapt the software to the needs of a particular language or region.
Project Internationalisation:
Interconnection and Integration of a Global Project Community
3

Translation of the Sahana User Interface
All text of the user interface is

stored in resource strings.
The resource strings are stored in

a separate resource file.
GNU/gettext() functions load the

string at runtime as needed
The appropriate resource file is

chosen depending on the current
locale settings
The resource files are relatively

easy to translate
4

Translation of the Sahana User Interface
Pootle = (PO-based Online Translation and Localisation Engine)

FOSS Server Software for Translation, based on the Translate Toolkit

Collaborative Web-based Translation of the Sahana Language Files

Sahana Pootle-Site: http://translate.sahana.lk

5

Maintenance of Translations
Original Version Localized Versions
1x
4x
→ Risk for Delays on Updates
6

Sahana Online Translation
Sahana Online Translation, Worldwide Distribution of Contributors
(Source: Google Analytics, translate.sahana.lk)
7

Character Encoding
UTF-8 encoding

UTF-8 is natively supported

by the most platforms and
client applications.
Additional software or the
●
installation of specific
fonts can be required in
some cases.
8

Writing Direction and other Writing Conventions
Right-To-Left Orientation needed for e.g., Arabic,

Hebrew and Persian
Requires adapted stylesheets and other code

modifications
Should be as pre-packaged and

as easy as possible to apply by the user

Ideally switching all locale settings at once by just

one click
➔ suggested for GsoC 2009
9

Data Model
Names, addresses and Government Assigned

Numbers must be consistently readable and
searchable even when switching to another (parallel)
locale
Identification numbers must be unique across all

contexts
Conventions and tradition, structures and

relationships can differ considerably betweens
regions or different ethnical groups in the same
region
Solutions for data model localization – Locale-

independent part vs. Locale-dependent part
10

Regulatory Compliance
Management of private information

Data protection and integrity

Freedom of information

…

Flexible AAA

Additional functionality for certain regulations

Mapping of data use and access levels

Implementation of Standards (required for e.g. US,

EU)
11

Terminology
Speak a ”common language”

Avoid ambiguous terms and fuzzy phrases

Comprehensibility and Appropriateness of terms

Diverse competences = diverse terminology

Serious risk for translation errors and

misunderstandings
Limited usability, unsafe application

Domain Ontologies and expert level translations of

them
Sahana glossary of terms

Styleguide for interface messages

12

Maintenance of Translations
Original Version Localized Versions
1x
4x
13

Community Integration
Local Language Support

Local Events and Exchange of Ideas

Localisation of the developer documentation

Lower Barriers-Of-Entry for new contributors

Public Relations in local languages

14

Current focus - overview
Translations

Sahana Localisation Knowledge Base

Right-To-Left Support

Automatic Translation/Translation pools

Localisation Add-on Packages

Promotion of Local Sahana Teams

15

Sahana - Available Translations
Ranking Language Total Speakers (millions) Native Speakers (millions)
#1 Chinese 1150 870
#2 Hindi 650 370
#3 Spanish 417 320
#4 English 1500 310
#5 Arabic 422 206
#6 Portuguese 235 180
#8 Russian 275 145
#10 German 180 100
#19 Tamil 80 68
#37 Burmese 42 22
#50 Indonesian 160 20
#63 Sinhalese 15 13
TOTAL up to 3500 2624
covering ca. 40-50% of the world's popoulation
16

Ongoing Translation Projects
Ranking Language Total Speakers (millions) Native Speakers (millions)
#7 Bengali 200 170
#11 Punjabi 105 88
#12 French 360 78
#20 Italian 120 60
TOTAL up to 700 396
Other languages suggested:
Persian, Filipino, Marathi, Assamese, Turkish, Hausa
Goal: 60-70% world's population coverage by end of 2009
17

Sahana Localization Knowledge Base
Useful Knowledge regarding Sahana Localisation

General as well as Region-specific Information

Focus on Sahana 1.0 Release

Portions already available:

Sahana Translation Guide

Sahana Glossary of Terms

UI Messages Style Guide for English

18

Current focus - overview
Translations

Sahana Localisation Knowledge Base

Right-To-Left Support

Automatic Translation/Translation pools

Localisation Add-on Packages

Promotion of Local Sahana Teams

19

Resources and Links
Sahana Project:
Project Homepage: http://www.sahana.lk
●
Sahana Localization:
Pootle Site: http://translate.sahana.lk
●
Sahana L10N Discussion Group:
●
http://groups.google.com/group/sahana-localization
Sahana L10N Blogsite: http://backus.nursix.org
●
Sahana L10N Wiki: http://wiki.sahana.lk/doku.php?id=translate:home
●
Sahana Language Packs:
Sahana Language Packs: http://pub.nursix.org/sahana/locale
●
Technical Resources:
Translate Toolkit and Pootle: http://translate.sourceforge.net
●
GNU/gettext: http://www.gnu.org/software/gettext
●
20

Speaker Contact Information
Dominic König, RN
Sahana Project Management Comittee
Online Translation Site Admin
dominic@nursix.org
www.nursix.org
21