After my blog post from Saturday and yet another call on debian-i18n, the number of complete languages bumped to 35 with Traditional Chinese, Nepali, Malayalam, Hebrew, Hindi joining the dance. Now complete: bg cs de eo es eu fi fr gl gu he hi it ja ko lt lv ml nb ne nl nn pl pt pt_BR ro ru sk sl sv ta th vi zh_CN zh_TW Incomplete:

 2 099% hu
   099% pa      2
 3 098% be
   098% ca
   098% dz
   098% sq      4
 4 097% bn      1
 5 094% ar      1
 6 093% bs
   093% da
   093% el
   093% et
   093% hr
   093% id
   093% ka
   093% km
   093% mk
   093% tl
   093% tr
   093% uk
   093% wo      13
 7 090% am      1
 8 088% ku      1

Great sunny winter time today over our area. Cold in the morning just as I like. Woke up early enough to have plenty of time in the morning. That was the perfect day to try pushing off my running limits (currently a half-marathon in 2h30). So, I ended up visiting Les Mesnuls, Mareil le Guyon, Bazoches sur Guyonne and all such nice names of my local area. This is the red line on my local Google map. Result: 26.6km, 2h28. W00t. Happy bubulle.

After working on badly encoded debian/control files, I began working on badly encoded changelog files. Again, most packages with that problem are packages that hadn't any upload for a while (the plethora of libroxen-* packages, for instance), so after a short notice, I began NMUs today. Except for a few packages where the maintainer mentioned (s)he prefers doing the upload (hurry up, Javier!), I expect dealing with that issue pretty soon and therefore fill in one of the release goals for Lenny.

We now have 30 complete languages in D-I: bg cs de eo es eu fi fr gl gu it ja ko lt lv nb nl nn pl pt pt_BR ro ru sk sl sv ta th vi zh_CN Some progress was made for others, but I doubt we reach my unofficial target of 40 languages by Thursday February 14th which is the deadline set by Otavio, D-I release manager. Incomplete languages:

 2 099% he
   099% hu
   099% pa      3
 3 098% be
   098% ca
   098% dz
   098% ml
   098% sq
   098% zh_TW   6
 4 097% bn
   097% ne      2
 5 094% ar      1
 6 093% bs
   093% da
   093% el
   093% et
   093% hi
   093% hr
   093% id
   093% ka
   093% km
   093% mk
   093% tl
   093% tr
   093% uk
   093% wo      14
 7 090% am      1
 8 088% ku      1

Translators, hurry up, please! We are in the process of *losing* translations for lenny if some of you don't wake up.

In today's annoucement of the release update, Marc Brockschmidt mentioned the release goal to make UTF-8-clean control and changelog files. I was also working on Pootle for i18n.debian.net and such non clean files for linux-doc-html-pt and linux-doc-text-pt were annoying me. So, I just decided to NMU these packages and fix the 20th century ISO-8859-1 encoding they use in debian/control. As this was easy, I decided to NMU (0-day....all these packages had longstanding bug reports while the fix is trivial) all other packages with badly encoded debian/control files and, as of tonight, there should not be such files anymore in unstable. Maintainers with badly encoded debian/changelog, watch out for your asses, I could decide for such a campaign for your stuff. So, please fix this: this is as easy as using "iconv".

(no, I won't make a daily post) After one day only, and a general call for updates, we went up to 22 complete languages for D-I level 1. +4

Otavio is currently busy preparing the first release of D-I for lenny. He is still polishng the release schedule but, as part of the process, I sent a call for translation updates today. The starting point is: 59 activated languages (including English) but 'only' 18 of them are 100%. For others:

1 100% bg
   100% cs
   100% de
   100% es
   100% eu
   100% fr
   100% gl
   100% gu
   100% it
   100% ja
   100% ko
   100% nl
   100% pt
   100% ru
   100% sk
   100% sv
   100% vi
   100% zh_CN   18
 2 099% hu      1
 3 098% be
   098% ca
   098% dz
   098% eo
   098% he
   098% ml
   098% nb
   098% nn
   098% pl
   098% pt_BR
   098% ro
   098% sq
   098% ta
   098% th
   098% zh_TW   15
 4 097% bn
   097% lt
   097% ne
   097% pa      4
 5 094% ar      1
 6 093% bs
   093% da
   093% el
   093% et
   093% fi
   093% hi
   093% hr
   093% id
   093% ka
   093% km
   093% lv
   093% mk
   093% sl
   093% tl
   093% tr
   093% uk
   093% wo      17
 7 090% am      1
 8 085% ku      1

Languages at 93% had no single updates since Etch. For these, I have some worries and I fear we could 'lose' some if their translator disappeared. Amharic and Kurdish never reached 100% yet but they have translated the most important strings, so we'll probably keep them. So, in short, the message to translators is: *move*. You have about 2 weeks.

When filing an ITP, Thou shalt read Developer's Reference 6.2.2:

• leading articles suck
• capital letters suck
• sentences suck
• jargon sucks

A lot of Debian packages description really suck, indeed. Bubulle resolution of the day: read ITP and nag proponents about synopsis writing style. Other developers/maintainers resolution of the day: get back to your packages descriptions and think about them twice after reading DevRef 6.2.2 and 6.2.3.

...install a Babelbox, de-activate the "db_info babelbox/info" line in preseed_early and put the following in big capitals on a sheet of paper attached to the monitor:

BABELBOX CONTEST
Guess what language is used to install
that Debian system.

...and wait for the crowd to slowly gather around the box and start guessing. If you can offer stickers pins CDs to winners, that's even better. Hint: do not activate easy to guess languages. Good candidates (apart from all non Latin ones which are all *very* hard to recognize for the regular French geek, except maybe Japanese) are Hungarian, Basque, Slovenian. Arabic is too easy to recognize and so is Greek. Indic languages are a great success but Dzongkha is generally the winner. Hint 2: do not sort languages alphabetically by locales. Geeks are too smart and the geeks manning the booth will learn too quickly. Watch also people trying to switch to a console during the D-I runs and typing "cat /proc/cmdline". Hint 3: train yourself to recognize the languages before the contest so that you look clever... That worked pretty well during 1st day of Solutions Linux.

The first of my talks at LCA 2008 gave me a chance to talk about the current state of the Debian project, which got me my first LWN.net coverage with a photo, even (subscribers only for now, the article will become public on 7 February). Thank you, Jonathan Corbet, for a very good article which nailed all the main points! Slides are here. I agreed to this talk on short notice because I like to talk about Debian and was honoured by the chance to represent the project in this form. I would not have been able to do it without plenty of helpful input from colleagues in the last few days. Since I didn t get a chance to display the final slide with the acknowledgements during the talk, I would herewith like to thank specifically Andreas Tille, Michael Banck, Kevin Mark, Josip Rodin, MJ Ray, Cyril Brulebois, Stefano Zacchiroli, Frans Pop, Moritz M hlenhoff, Russ Allbery, Steve Langasek, Luk Class, Andreas Schuldei, and Christian Perrier. No guarantees for the completeness of the list. The talk ended in an open discussion on how Debian could improve. I took notes and shall forward them to the project mailing list, once I get a chance. Thanks to all the participants, as well. Update: during the talk, I mentioned that there was no security support around the time of the etch release. Thanks to Moritz M hlenhoff, who spotted my error: that should have been sarge . The problems with security support had long been resolved by the time etch was being prepared, and this was in (large?) part thanks to Moritz. Sorry for the screwup! I also called Linux to tend towards multimedia more than one might like. I should not have made this comment as a representative of the Debian project, and I probably did unjust to the Linux kernel in whole. This is entirely a personal issue, I have a number of problems with Linux memory management, scheduling, and some other points relevant to production use. I ve had some of these problems for years, but they seem never to get fixed, while development is fast-paced. Then I look at some of the work being done and I wonder what the priorities are. Regardless, I should not have made this comment and I apologise for it.

That's what happened to my /usr this morning during my daily dist-upgrade. Sigh. I probably have too many things that I installed once to get a look at and forgot to remove. Time to clean that laptop out again..:-)

Joey, just don't delay your releases waiting for incoming translations. Gettext is anyway smart enough to not use translations when they're non-existing (aha) of fuzzy. So, an incomplete translation for a program will just end up in a partially translated program. That's actually the case for *many* localized software, including the big ones. I suggest that you apply such a schedule:

• Plan a release
• Send call for translations (hint: podebconf-report-po is your friend even for non debconf translations...see its man page) and give translators a deadline. The deadline has to be reasonable and leave some room for translation proofreading. Most often, when few strings are impacted (say not more than 20 but, again, that's hard to judge)
• When the deadline is reached, upload release with what you have
• A few days weeks later, eventually release a new version with late translations

But, as you suspected, releasing without synchronised translations is the better choice....as long as you care abou tsending call for updates when needed. BTW, if you need some help to coordinate incoming translations for apt-key, maybe ask on debian-i18n and someone will eventually pop up (hopefully, not /me).

The dpkg maintenance team just changed the priority of the good old dselect from Important to Optional. Simultaneously, overrides were changed by FTP masters. As a consequence, dselect will no longer be installed by default when installing a "standard" Debian system. Thanks to the dpkg maintainers for the first move (Required->Important), as well as Joey Hess who suggested to downgrade from Important to Optional. Old timers, this is time for a little tear in memory of dselect..:-)...as well as a thought to people who wrote and maintained it (it is still in the archive and part of dpkg source package but noone really maintains it, now). Time to hunt a few more "Important" packages that should be downgraded.

Mausefalle, Karusellkurve, Steilhang, Br ckenschuss, Alte Schneise, Seildamsprung, L chenschuss, Oberhausberg, Hausbergkante, Zielschuss (ach mein Gott!) and the terrible Zielsprung. After last week's Lauberhorn downhill, today was Hahnenkamm's downhill race in Kitzb hel (Austria). Even though I slightly prefer the Wengen race, Die Streif is one of the scariest racest of the Ski World Cup. Great spectacle today again, however saddenned by Scott Mc Carthy's terrible accident on the Zielsprung. We don't have news yet but I cross fingers for him. Today, Didier Cuche won the race wonderfully, just before lastweek's winner, the tireless Bode Miller. Again another great ski week-end, really.

Tricky question, isn't it? Indeed, according to the ISO 639-3 standard which finally became an official ISO standard in 2007, there are 7696 different languages, including extinct, ancient, historic, and constructed languages. Of course, this may become a very debated discussion about what constitues a "language". Let's give the guidelines used by SIL International, who was designated by ISO as the official maintenance agency for ISo-639: There is no one definition of "language" that is agreed upon by all and appropriate for all purposes. As a result, there can be disagreement, even among speakers or linguistic experts, as to whether two varieties represent dialects of a single language or two distinct languages. For this part of ISO 639, judgments regarding when two varieties are considered to be the same or different languages are based on a number of factors, including linguistic similarity, intelligibility, a common literature, the views of speakers concerning the relationship between language and identity, and other factors. The following basic criteria are followed:

• Two related varieties are normally considered varieties of the same language if speakers of each variety have inherent understanding of the other variety (that is, can understand based on knowledge of their own variety without needing to learn the other variety) at a functional level.
• Where spoken intelligibility between varieties is marginal, the existence of a common literature or of a common ethnolinguistic identity with a central variety that both understand can be strong indicators that they should nevertheless be considered varieties of the same language.
• Where there is enough intelligibility between varieties to enable communication, the existence of well-established distinct ethnolinguistic identities can be a strong indicator that they should nevertheless be considered to be different languages.

So, we (iso-codes package maintainers) now have to update that part of our package and buy a big bunch of aspirin to people who will work on translating this. Moreover, contrary to all other such standards (3166 for country names, old 639-2 for languages, 15924 for language script names), there is no bloody official *French* version, so even the French translator has to work. Gues who this was until now? :-) To end this blog entry, let's ask a funny question to the readers (no, comments not allowed on my blog, I'm lazy, so answers by mail or in your blogs): what is the country in the world which as the highest number of languages listed in ISO 639-3?

The traditional gathering of the Linux and FLOSS community in France will this year again feature a Debian booth. The event will be held in Paris-La D fense (CNIT), on Jan 29-31. As the Babelbox was born in that event three years ago, we will again have one, of course (people will get bored of it...). Frans Pop improved the thing so much that it took me less than 2 hours to set one up. We will also probably have a few other demo machines and the usual bunch of developers, contributors and friends popping up at the booth. If you're around, please come up and say "hi"....(don't forget fingerprints, IDs and some aspirin to cope with the verbiage of some of us who definitely talk too much).

A new version of shadow packages (login, passwd, useradd and all other low-level user management utilities) hit unstable this week-end. That release is quite a major update as this is the first release where Nicolas Fran ois introduced numerous changes meant to make the versions of shadow used in various distros reconverge. So that version now really says "hey, Debian took shadow over". He thus incorporated lots of patches from Gentoo and Fedora as well as a few Debian ones (Debian packages were apparently closer to upstream code than others). He also added support for SHA256 and SHA512 encryption methods. Careful users will read /usr/share/doc/login/NEWS.gz to get an idea of these changes. After a small hiccup with 4.1.0-1 (The "Bleu d'Auvergne" release) which was hitting our German-speaking users (login is now providing a German version of su(1) which is also provided by manpages-de, therefore making the package uninstallable), 4.1.0-2 (the "Bleu des Causses" release) is there sitting in unstable waiting for testing. Nicolas is already working on 4.1.1, now focusing on code cleaning. That version should be out pretty soon. Also note that we'll soon be short of "Appellation d'Origine Contr l e" cheeses names for the releases and we more and more have to make our imagination work. General de Gaulle was wrong: there are not 50 million cheeses in France.

Russisprung, Hundschopf, Minsch-Kante, Canadian Corner, Kernen-S, Wasserstation, Hanneggschuss, Silberhornsprung, sterreicherloch, Ziel-S. If that rings a bell, you know what I'm talking about. Today was *the* ski downhill race of the season, namely Lauberhorn in Wengen, Switzerland. The longest (2'30") and fastest (sometimes up to 160km/h) downhill races of the season. Wengen's scenery is among my favourite ones, with the Eigerwand, M nch and Jungfraujoch dominating the entire race (the image of racers in the S curve before the Hundshopf jump, with Eiger, M nch and Jungfrau at the back is usually the image of the year when it comes at ski competitions. And, this year again, the sun was brilliantly shining over Wengen so we spent a great time watching this (on TV...going there to see it live is one of my dreams). And, for this non standard downhill race, we again had a non standard winner with Mr Bode Miller, winning it for the second time in a row. Next week is the Hahnenkamm downhill in Kitzb hel, the ski Mekkah of Austria, on the Streif. Also coming, the Schladming night slalom, Chamonix downhill, Val d'Is re downhill.....January is always busy for those who enjoy competition ski. (and, as I do not only watch sports on TV, I also ran my second longest distance today, with 19.5km in just 2 hours. Yay)

Here's what I found in the recent APT upload as changes to PO files:

 #: apt-pkg/cdrom.cc:678
 #, c-format
 msgid ""
-"Found %u package indexes, %u source indexes, %u translation indexes and %u "
-"signatures\n"
+"Found %zu package indexes, %zu source indexes, %zu translation indexes and %"
+"zu signatures\n"

The consequence of such changes is fuzzying translated strings, thus forcing translators to re-update their files and the translation coordinator to commit them. In short, valuable time wasted. One could object that translators shouldn't try to keep up 100% all time long. However, those who do not do so just end up with huge updates to do...which is sometimes demotivating. So many of us prefers keeping up with small updates. However, when one sees such updates that appear to be cosmetic, it's very hard to refrain hating developers (I even ranted about this in dpkg commit logs, which bringed some argument with Guillem Jover.... I certainly shouldn't have ranted in commit logs but that probably gives a good idea how upset I was). Please think about this next time you change a translatable string, particularly when you *know* that someone is taking care to coordinate *your* software translation work with translators. Side comment: bzr choke up on my local APT copy and I can't even cope with updates for APT. That will not really help me to love those DVCS. Update: many explained me that such changes are not useless which I'm ready to accept given my level of knowledge in programming. So, apologies to programmers who made these changes. I just hope this will however better give the translator's POV wrt such matters. We all know that translators *and* maintainers work really hard to give the best results and both always need to have a better knowledge of other's constraints....and limits! I've also been very wisely teached about possible ways to jiggle with VCS to work on defuzzification by reverting changes, unfuzzy, re-apply changes. Sure. At the expense of a lot of invested time to check how to do this properly with all these neat modern DVCS, I could. Yay, I so much love fiddling with VCS....and spending hours doing this is something that makes my days.

Raphael Geissert reported bug #460000. As bug #450000 was reported as of November 9th 2007, this roughly gives 2 months for 10,000 bugs, so 60,000bugs a year. Bug #500000 should then be reported around Sept 10th 2008. So, the current candidates for winning the 500000th bug contest are Miguel Gea or Kartik Mistry. I should plan a few MBF to keep chances of winning the contest myself.