IRMA to Deliver Complete Translations of Linux in 78 Languages by 2006

Linspire, Inc. today announced the release of a new Web-based translating application that will allow volunteers to easily translate leading Linux applications into nearly 80 different
languages. Dubbed the International Resource Management Application, or IRMA,
the project calls on users who speak English and another language to volunteer
to translate parts of the operating system. Currently, 24 languages are
supported through the system, with 54 additional languages to be added over
the next few weeks. To view languages currently supported by IRMA or to sign
up as a volunteer translator, visit http://www.irmateam.com.

Using a user-friendly online interface that automatically selects various
sections of text or code in need of translation, volunteers sign up for one or
more of IRMA's supported languages. Once registered, translators can simply
log in from any computer connected to the Internet then begin translating
small, auto-selected strings of code. When complete, the translations will be
integrated into language-specific versions of popular Linux applications, like
KDE, Mozilla, OpenOffice.org, Nvu, and Lsongs, and made available to a wide
audience around the world.

"Instead of using expensive translation firms, we're turning to Linux
supporters to help make Linux available to new segments of the population,"
said Michael Robertson, CEO of Linspire, Inc. "Soon, a Linux-based operating
system will be available to any user, regardless of their native language,
giving people a level of comfort and personalization with Linux that they may
not have had before."

More than 200 translation volunteers have already signed up, representing
35 languages. Linspire is currently seeking more volunteer translators in all
languages. Those fluent in English and another language who are interested in
joining a translation team should visit http://www.irmateam.com.

Added together, the group of major Linux components contains approximately
168,000 words. Besides cost, challenges such as consistency and encoding make
such a large translation project difficult to facilitate. To have an accurate
translation, text files must be encoded so that all characters in every
language are displayed correctly. In typical translation efforts, much
information is lost between moving these files from one machine to another,
opening them in different text editors, and changing character sets. IRMA, on
the other hand, was created to surmount these difficulties and make it much
easier for numerous volunteers in different parts of the world to coordinate
their translation efforts.

Once a user registers to translate, they are directed to an easy-to-use
interface that presents them with pages that have strings of text in need of
translation. Rather than translate large swaths of text at a time, users
choose which Linux application code strings they'd like to translate using an
easy interface of drop-down menus and simple searches. If similar words or
phrases have already been translated, wording suggestions appear and can be
selected. Volunteers can choose to translate as much or as little code as
they like.

All translation is done through a Web interface, so users do not need to
install any custom translation software -- only a Web browser is needed.
Users can also access the translation pages whenever and wherever they like,
no matter what operating system they run, then save their work to submit it
back into the system. Through IRMA's online interface, lead translators,
called "managers," have the ability to change translations, verify words, and
generate reports on translator's activity levels and consistency rates.

There are many enhancements scheduled to the IRMA project over the next
few months. Since translations do not come from professional translating
sources, checks will be put in place to control the quality of the
translation. When there are more than 25 volunteers translating in a
particular language, translators will work on a double-blind system, where two
people translate the same text and have their translations checked against the
other through the system. When translations match, they will be used. If
they do not match, a lead translator will reconcile the translations to make
sure the correct one is used. Those languages that do not have enough
translators for a double-blind check will be reviewed by language managers,
who will edit the translation for consistency and accuracy.

To learn more about the IRMA project, please visit http://www.irmateam.com.

For more information about Linspire, contact:
Heather MacKenzie
Linspire, Inc.
858-587-6700, ext. 263
858-587-8095 Fax
pr@linspireinc.com

Related Articles

• Czech Post turns to SUSE LINUX to Support Postal Operations
  Sunday, 30 January 2005



• Linux Magazine Becomes Sponsor Of Linux Professional Institute
  Thursday, 27 January 2005



• Forrester Research Surveys Linux Usage and IBM Increases Open Source Programmers
  Tuesday, 25 January 2005



• IBM Introduces New Low-Priced POWER5 Server Based on Linux
  Tuesday, 25 January 2005



• Accent's Caliper with Linux Eliminates 'Blue Screen of Death' Issues
  Sunday, 23 January 2005



• Embedded Linux Development and Testing Tools among Hot 100 Products
  Sunday, 23 January 2005



• MicroWorld announces the launch of eScan for Linux Servers
  Sunday, 23 January 2005



• Runtime Revolution Express 2.2 Free for Novell SUSE LINUX Users
  Thursday, 20 January 2005



• Keynote Speaker Announced for the First Security-Enhanced Linux Symposium
  Wednesday, 19 January 2005



• Eclipse 3.0-based TimeStorm Linux Development Suite Evaluation Kit Available
  Wednesday, 19 January 2005



• Linux and Open Source in 2005
  Wednesday, 19 January 2005



• Bugopolis upgrades Product Offerings with Bugzilla 2.18 and Red Hat Linux
  Tuesday, 18 January 2005



• LynuxWorks Introduces Linux Point of Sale Solution for Worldwide Market
  Tuesday, 18 January 2005



• User-Mode Linux Software for Apple PowerPC G5 Released
  Tuesday, 18 January 2005



• Xybernaut to Co-Chair LINUX Competency Group
  Friday, 14 January 2005