Busy times for the Debian developers: There's posts about GCC 4.1 now the default GCC version for etch, Summary of Debconf i18n/l10n activities and Report from the “Dzongkha Linux launch. Enjoy.
Report from the “Dzongkha Linux launch
From June 1st to June 5th, I have been invited to attend the Dzongkha
Linux launch event, in Thimphu, Bhutan.
For those not aware of this, Bhutan is a 700,000 inhabitants country
located between India and China, in the eastern part of the Himalaya
range. The size of the country is somewhat similar to that of
Bhutan's national language is Dzongkha, a language from the
Sino-Tibetan family. Recent laws in the country have enforced the use
of the national language in all official events and all official
communication. Therefore, even though all the (free for everybody)
education system is bilingual in English/Dzongkha, it is very
important for the country to be able to use the Dzongkha language on
After a quite deceptive attempt with Microsoft to include support for
Dzongkha in Microsoft operating systems ($523,000 have been thrown in
this attempt), the Ministry of Information and Communication launched
the DzongkhaLinux project 2.5 years ago.
During that time period, the Department of Information Technology
(DIT) has been able to build a complete system with complete support
for the Dzongkha language. The system is based on Linux and more
specifically on Debian. It consists of one CD which can be either
installed or used as a live CD (the installation system is using
Morphix, not D-I which was not ready at that moment).
The CD embarks a complete set of Dzongkha-localised applications,
namely the Gnome environment, the OpenOffice suite, the Mozilla web
browser, the Evolution mail reader and GAIM as instant messaging
On June 2nd (national holiday in Bhutan as anniversary of the
coronation of the King), the DIT was officially launching the
DzongkhaLinux system (“Our language….our software”).
Because of the recently very productive collaboration with the
technical project head, Pema Geyleg, for inclusion of Dzongkha support
in Debian Installer, I was invited to attend the event and give there
a keynote lecture about “Free Software and the Global community”.
The event was very widely advertised in Bhutan: it was covered in all
newspapers and got a strong importance in the national television.
Two ministers of the Bhutanese government were attending the event:
the minister of Information and Communication (Chief Guest) and the
minister of Education. Nearly all other ministries were represented by
Secretaries. The country's Prime Minister, originally scheduled to
attend, had to cancel because of other commitments abroad.
As the project is part of the PANLocalization project (http://www.panl10n.net), aimed at bringing localized computer use in
several Asian countries, and funded by IDRC (a Canadian governmental
organization), several representative of these organization, or other
governmental or non governmental organizations, were represented.
More specifically, Panl10n was represented by Dr. Sarmad Hussain, from
Pakistan, head of the Center for Research on Urdu Language
Processingin Lahore, Pakistan..
Also invited were the Nepali localization team, who launched a very
similar distribution for Nepali language, as well as Guntupalli
Karunakar, recognized expert in Indic languages handling in Free
Software, and one of the leaders of the Indlinux project
The event itself featured:
-Introduction by the head of DIT
-My own keynote lecture about FLOSS and the Global Community
-Dr. Sarmad Hussein, lecture about the PAN localization project
-A pre-recorded demo of the DzongkhaLinux dostribution
-A conclusion by the ministry of Information and Communication
I have been incredibly impressed by the very wide coverage of this
event (the next days, barely anyone I was meeting could talk about it)
and the strong commitment showed by local officials to support the
My own keynote lecture focused on the main key aspects of Free
Software, especially in developing countries (Openess, independence,
ability to preserve the local culture and knowledge, ability to
develop a local software and services industry). I, of course, also
introduced the Debian Project, targeting the point on the commitment
of the project to Free Software and enlightning its ability to be
“derived” for specifics needs.
Post-event discussions (including private discussions with the
involved officials and ministers) have shown me that this concern has
been very well received. The ability of Free Software to allow
customization and appropriation of the technology by local people is
very wel understood.
We can safely assume that, in a near future, teachers in all Bhutan
schools who, up to now, needed to prepare their courses hold in
Dzongkha by hand, because of the lack of tools able to process the
language on computers will be able to use DzongkhaLinux for their
Similarly all monks in Bhutan monastery will soon be able to work with
their sacred texts with computers instead of cofying them by hand (the
buddhist religion plays a great role in Bhutan's day to day life).
In general, all needs for the Dzongkha language processing will be
able to be fulfilled, which should improve the coverage in Dzongkha in
many areas in the country.
In short, DzongkhaLinux is promised to a very wide success and can
also be considered as a big success for Debian. When the most
important guest at the event mentions in his lecture that “we had a
collective dream of having our own computer software for our own
needs”, I feel like the dream that the Free Software Community is
having since more than 20 years has become a reality in that small
part of the world.
I hereby want to deeply thank the Department of Information and
Technology, the ministry of Information and Communication, as well as
all individuals who have made this trip to Bhutan a complete success
and a personal deep achievement. Aside from this, it has allowed me to
discover a unique culture and a wonderful country, with a strong
commitment to preserve that local culture and the natural environment.
The legendary friendliness and openess of the Bhutanese people is
definiteily not usurpated.
I am deeply confident that the future collaboration between Debian and
similar projects all around the world will became very common and will
justify all the energy that all individuals involved in Debian in Free
Software invest daily in this dream.
Official announcement of Dzongkha Linux:
Bhutan's Department of Information Technology:
The Dzongkha Linux project home (needs update):
IDRC, International Development Research Center:
Summary of Debconf i18n/l10n activities
(this report is a little bit late as it took time to finalize
it…sorry for the inconvenience)
The work on internationalisation (i18n) and localisation (l10n) at
Debconf6 has been particularly interesting and productive.
The main topic has been the discussion on l10n infrastructure, both
summarizing existing features and services (most of them being
summarized in the paper I published along with Javier Fernandez
Sanguino) and future features.
The two main topics were:
-begin to work on the l10n infrastructure specifications
-setup a precise project for the proposed work inside the Google
Summer of Code project by Gintautas Miliauskas
Two BOF sessions have been organized and the l10n talk by Javier
Fernandez Sanguino and Christian Perrier concluded the week.
Many other informal discussions also happened with several people
involved among who we can cite: Javier Fernandez Sanguino, Otavio
Salvador, Michael Bramer, Nicolas François, Javer Sola, Gerfried
Fuchs, Margarita Manterola, Raphaël Hertzog, Christian Perrier, Frans
Pop, Javier Sola, Aigars Mahinovs, Daniel Glassey.
Some cntributions from the mailing list by people not present at
Debconf have also been included and taken into account.
During our first session and the initial discussions, the main point
was trying to setup ideas about the needs for the infrastructure: what
are its targets (aka users) and what features might be needed by each
of these targets.
The second session happened after some informal work between
contributors and was aimed at being a summary of the ideas that were
All this lead to the following conclusions:
We identified the following targets, or categories of users:
They are in charge of managing the system and the users
They work on translations of original strings coming from “upstream”
They check the work from the translators and certify that it meet the standard of each translation team
Either originating inside Debian (the package maintainers) or outside
(upstream software authors and maintainers), they are the source of
translatable material and the destination of translations.
They are occasionnal visitors of the web site, or potential users
such as governmental institutions. This category include “regular”
users who will use the system to occasionnaly contribute (error/bug
reporting and the like)
– Team coordinators
They are in charge of coordination the work of one
translations. There may also be transversal maintainers who just work
on a single directory or software.
Needs of each user category
-delegate –> backup admins, but also delegate tasks to translation
-get the information about the needs and priorities
-“book” a translation. Reservation should be valid only for a certain amount of time. After it (calculation can be automated), the translation is released (post by the robot in d-l0n-foo).
-get material (web, mail, SVN…) or work online
-derive translations (Spanish, translate from other langs than en)
-choose their preferred format (XLIFF, PO…)
-licence translations (???)
-ability to merge reviews and ACK proposals one by one
-avoid collisions with files translated in other projects
-need to enforce the concept of “owner” of a translation
-optionnally more than one ower in some projects
-glossary (able to propose several translations)
-get work assigned, on request
-express “Intent to review” (released after a calculated amount
-do work in public
-see what other reviewers have proposed
-Modify the source location
-send the material
-Ask for updates in release time by priority raise
-get the updated material
-Be notified of updates (opt-in). Options:
– every commit
– when “Ready”
-visitors (or ordinary users)
-learn about the system (stats..) (references to l10n)
-propose changes in translations
-can be per project and per lang
-get status and stats about their field of expertise
-manage assignments (in addition to automated unassignments)
-setup and use different processes from team to team
(nr of reviews, etc.)
-set some goals, and see/show if the goal is achieved
These design goal raise the importance of the project for being as
modular as possible. The core of the project should be a backend on
which all other modules will plugin to.
Going with WordForge
The presentation by Javier Sola of the Pootle and related projects has
finished convincing us that working with the members of the WordForge
project is certainly the way to go. Their project is aimed mostly at
these goals and even more which we had never formalized completely.
That project doesn't suffer the concerns we have with the Rosetta
project from Ubuntu/Canonical. We indeed have been later confirmed
that WordForge also work on open standard for communication between
localisation projects and these could be used to communicate between
the Debian infrastructure and that of Ubuntu.
These early specifications for features certainly have to be enhanced
and completed in the future but they will probably already allow
Debian to merge them in the Pootle/WordForge specifications which are
still under work.
As a consequence, we will complete WordForge specifications with ours.
Google Summer of Code project
The challenge with the GSoC project is multiple:
-complete the project prepartion in a very short time
(less than 1 week)
-give a precise goal to the student
-make it reasonable to achieve in the 3 month time frame
The first idea that came out has been requesting some work on some
“bounties” of the WordForge project. However, some of us prefer that
the specifications are ready before we enter such path. Given that
they won't obviously be complete, we finally decided to launch a quite
conservative project just to ensure that the work done has still some
benefit for Debian without compromizing the future.
As a consequence, it has been decided that the requested work will be
separating the frontend from the backend in Pootle, which will allow
future work to be concentrated on the backend used by all future
software in the framework.
Draft specification for the Debian i18n infrastructure
This summarizes the various discussions that occurred during Debconf
and on the mailing list during the month of May, mostly. Listed below
are the various modules we identified as needed in the target
– Import from po-debconf (need standardization action to provide
up-to-date POT. At the minimum, reproduce the current layout)
– Import from programs PO (ditto)
– Import from man pages converted with po4a (standard layout mandatory)
– Documentation (action from maintainers==opt-in)
– Web site (no action, source under control)
– Translation teams define their own processes from a set a standardized
– TTD (Translation To Do)
– TTU (Translation To Update)
– RFR (Request For Review)
– Reviewed (with counter) == LCFC (Last Chance For Comments)
– Pushed to maintainer via Debian BTS
– Pushed to maintainer via another BTS
– Ready for Use
– Different processes for different types of translations
– Branching translations with merge features
(manually or automaticall for stable/testing/unstable)
– Individual PO files
– Set of PO files as a tarball
– Individual XLIFF files
– Online work
– Web interface
– Mail interface
– Other Pootle servers
– Rosetta servers
– TP server (?)
Future plans for Debian l10n contributors
Reviving the DDTP – translated package descriptions as a release goal
While the work on the new infrastructure advances, Michael “Grisu”
Bramer will stabilize the current DDTP code to allow some
maintenance of existing translations of package descriptions. Most of
this work is already achieved, indeed.
In parallel, and because APT 0.6 now includes support for translated
packages descriptions, the use of these will be promoted. Temporarily,
the Translations files will be hosted on another server, namely
Some discussion has to happen with the ftpmasters team to decide
whether the use of Translations file on FTP servers is considered
suitable and when their inclusion can be possible. Of course, for this
to happen, the DDTP must have a working maintenance system so that
maintainers of packages can be sure that the bug reporst they might be
received from the DDTP, will be maintained.
The plan here is to temporarily use http://ddtp.debian.net as the demo
case of what can be done with “Translations-*” files and the new
APT. We should (re-)advertise this, have it used during a few weeks
and then discuss with the ftpmaster to have it integrated in the main
repositories, along with Packages files.
Having working translated package descriptions as an “etch pet release
goal” for the Debian i18n team seems possible.
A meeting will be organized in Extremadura, from Thursday September
7th to Sunday September 10th.
This meeting will use the specifications previously finalized by the
Debian l10n community to allow contributors to start building a
consistent backend to the future Pootle system, benefitting from the
work of Gintautas Miliauskas to separate both.
Inviting the Pootle developers to the meeting is considered highly
wished. Hopefully, in the meantime, enough will have been achieved,
especially by Gintas during his work.
The goal could be setting up the first Debian Pootle server.
For all information about Extremadura sessions, the #extremadura-2006
channel can be used on freenode (irc.debian.org).
GCC 4.1 now the default GCC version for etch
—–BEGIN PGP SIGNED MESSAGE—–
The compilers from GCC 4.1 provide now the default compiler for etch
for Ada, C, C++, Objc, ObjC++, Fortran95 and for the Java language.
The packages are currently in the incoming queue and will hit the
archive on Wed. June 7. Compilers for Fortran77 and Pascal still
default to the compilers built from GCC 3.4. Although we did not get
feedback from all port maintainers, we did switch the compiler
versions to 4.1 for all architectures .
Test builds on different architectures did reveal compiler bugs and
hundreds of bugs in packages, where the compilers became more strict.
The remaining 14 open reports against these packages can be found at
. Many thanks to Martin Michlmayr for endless archive rebuilds and
working on these bugs.
During the test rebuilds, usually only reports from package building
failures were reported and fixed. There may miss some bugs
resulting in miscompilation of a package; in some case that can be
compiler bug, in other cases it's an overviewed compiler warning,
leading to wrong code.
We did pick two compiler warnings and scanned the build logs of one
archive rebuild on alpha (64bit), where wrong code may be generated.
These warnings can be found in 1600 packages ; they are:
– dereferencing type-punned pointer will break strict-aliasing rules
For a description see the gcc(1) man page or ; a common
workaround is to lower the optimization using -fno-strict-aliasing.
See -Wstrict-aliasing and -Wstrict-aliasing=2 as well.
– cast from pointer to integer of different size
cast to pointer from integer of different size
These warnings may point to code which is not 64bit clean. They are
most likely not seen on 32bit architectures. See the amd64, alpha
and ia64 build logs for these architecture specific warnings.
Bug reports for all these maybe-bugs could be a way to audit these
packages and close them if the package is looked at for the problems.
—–BEGIN PGP SIGNATURE—–
Comment: Processed by Mailcrypt 3.5.8, an Emacs/PGP interface
—–END PGP SIGNATURE—–