We made a quick appearance in the Hacklab yesterday, enough to say "hi" to a dozen of people around and mostlysee how things are organized.
Then bed, which is probably the best way to avoid jetlag.
This morning, I of course woke up quite early (Debian grandfathers, blah blah) and of course I went out running..:-)
The initial plan was to do a short enough run in Central Park. However, I managed to get partly lost and, indeed, after entering the Park in the NW corner, I managed to read Columbus circle...in the SW corner after wandering around the large alleys. So, my only choice was then to come back through the westernmost part of the Park back to Columbia..and thus ended up with a 13km run (for running junkies, the pace was 5'31"/km (8'52"/mile for local runners).
And now, I'm in the hacklab with JB (did I already say that I'm proud of this? Yes, dozens of times...OK...).
And they're waiting for me to go to lunch....so, end of this blog post..:)
posted at: 19:47 | path: /bubulle/planet-debian | permanent link to this entry
Guess what? That was a localization bug..:-)... Joe is doing a lot of work for Danish localization in Debian since a few months and I'm happy to see him rewarded byt this little "success".
Bug #580000 was reported as of May 2nd. It is now confirmed that the rate for 10,000 bugs is more 2.5 months than 2 months as it was in the past.
So, now my next post about this topic will be about Bug #600000 for which a bet was placed two years ago. I already lost the bet as I predicted it to happen on June 30th.
Either René Mayorga or Josué Abarca are likely to win this bet, more likely René, I would say. One should note that all people who did bet where optimistic about the moment where bug #600000 would be reported.
posted at: 08:27 | path: /bubulle/planet-debian | permanent link to this entry
The package suffered from a very longstanding RC bug that made it unusable on amd64 machines. I had to call for help Ocaml gurus and Rémi Vanicat kindly provided a working patch.
So, now, geneweb can be released and I won't have to ask for my pet package to be removed from testing.
I even switched it to dh7 and "minimal" debian/rules file (something I planned to do at DC10 but finally turned out to be easy to do.
That will be my last week-end of activity in Debian until DebConf10 as I'll be away for the upcoming week-end. So, for those of you who attend DC10, see you in NYC. I'll arrive on Wed 28th evening for part of Debcamp and I'll have the great pleasure to introduce my son Jean-Baptiste to the Debian world.
posted at: 05:19 | path: /bubulle/planet-debian | permanent link to this entry
We finally have 65 supported languages. In a final effort, Belarusian, Ukrainian, Hungarian, Tagalog, Kurdish, Croatian and Macedonian were "rescued" and completed enough for being included in the release.
Unfortunately, I had to deactivate Wolof, Serbian and Welsh because of lack of activity. For Serbian, this is very infortunate as new translators indeed tried to contact me with updates, which I missed and discovered only after it wasn't possible to revert the change. Good news are that Serbian is now already reactivated for the next release and we will even have both the Cyrillic and Latin versions of it.
The mass upload of D-I packages happened mostly on Sunday 12th (you probably noticed that if you follow debian-devel-changes).
It is already possible to test what will be D-I beta1: just go to the D-I development web page, then choose to download the daily built netboot image. Boot it with "install mirror/suite=unstable".
This is not *exactly* what beta1 will give as this will install unstable and not testing, for non D-I packages...but I think that, until D-I packages migrate to testing, this is the best way to test beta1.
The next step in D-I release preparation should be the mass migration to testing of most udeb packages.
posted at: 05:09 | path: /bubulle/planet-debian | permanent link to this entry
It has to be awesome. It will be.
posted at: 18:15 | path: /bubulle/planet-debian | permanent link to this entry
You should do it particularly if you care about Belarusian, Ukrainian, Hungarian, Tagalog, Kurdish, Croatian, Macedonian, Wolof, Serbian, Welsh.
(Brazilian Portuguese was also mentioned in that post but updates happened today and I'm confident that Felipe will soon catchup. Croatian is also likely to be OK as Josip is often updating at the last minute...).
posted at: 17:15 | path: /bubulle/planet-debian | permanent link to this entry
The story starts at http://lists.debian.org/debian-l10n-french/2010/05/msg00178.html. In short, someone apparently named "Alain S<shhhhdonttellmyname>" mailed the debian-l10n-french mailing list to request for a message written in march 2006 by someone else whose name is also S<shhhhdonttellmyname> to be removed from the debian-*user*-french mailing list archives.
"Someone" then politely answered him with http://lists.debian.org/debian-l10n-french/2010/05/msg00179.html, trying to use as less irony as possible.
At least, this "someone" refrained himself to explain that, doing so, yet another copy of Mr S<shhhhdonttellmyname> name was being replicated over dozens of servers all over the world, also exposing that he's probably living in Antony, Bourg La Reine, Chatenay Malabry, Chevilly Larue, Fontenay Aux Roses, Fresnes, L'Haÿ Les Roses, Le Plessis-Robinson or Sceaux, in France.
Later on, this "someone" received many private mails from that Mr S<shhhhdonttellmyname>, which got automatically and gradually classified as spam. Funnily, a few days later, that "someone" received another mail from another person, posting from the very same IP address and requesting for such removal again.
That "someone" being Very Evil and a little bit pissed off answered with http://lists.debian.org/debian-l10n-french/2010/06/msg00136.html which is untranslatable. So, this is why I hope you can read French. Have fun.
posted at: 21:36 | path: /bubulle/planet-debian | permanent link to this entry
A string freeze is a formal agreement among developers to avoid changing localizable material without concertation. This is a commitment towards translators to guarantee them that, while they're doing an effort to complete their work, this effort won't be broken by a developer adding of modifying translatable material.
This can be seen as a first step towards the preparation of a release of D-I. Indeed, the only release we did for Squeeze (D-I alpha1 release) had no formal translation update round.
We currently have 45 languages that are considered to be "in good shape" (even if not all complete) and very likely to remain activated in Squeeze. However, 20 (TWENTY) languages are in danger of being deactivated because of low activity.
To avoid this, I will once again take my superhero suit, add some leather to it, take a whip and start nagging translators, doing noise and, in short, annoying the entire world (including blogging too often), in order to shake people enough so that we reduce this number to zero. Be prepared.
Interesting links:
posted at: 07:44 | path: /bubulle/planet-debian | permanent link to this entry
In short, I achieved two races I was planning to do.
I ran my first trail on May 16th, Le Trail des Cerfs ("deer trail"). The experience of trail running is awesome. Runners are very friendly and all of them are nature lovers. So, the feeling is completely different from mass macadam races (those are also a must-do, still!). I had a great time running that trail (which ended up to be exactly the distance of an half marathon). I completed it in less than 1h50, which is good, as the positive slope was about 300 meters, with sometimes very though rampswhere walking was nearly mandatory.
At this occasion, during the "after", I discovered some members of an online community of runners, named Kikourou ("Who is running where?"), and I'm now a member of this online community (yet another one...). That makes the promise of lots of beers after races...:-)
Today, I completed my "Maureparathon", a full marathon ran alone on small roads and bicycle lanes in my home city, and was happy to break my personal best by 6 minutes, down to 3h43'14". Yay. For sure, still very far from SuperNoèl (3:03:56 in Mainz), but he's younger..:)
As I planned, I'll be running the Sur les Traces du Loup ("The Wolf Track"?) 33km trail in La Ville aux Clercs, 2 hours away from our place. Elizabeth is very happy to come with me in a 2-day trip in Chateaudun neighbourhood. Renaissance castle visit mandatory for the following Sunday...:-)
Nice planning, eh?
And, guess what? I'm already trying to spot some local 10km races while being in New York. I would find it an interesting experience to run with local runners. Just like free software...and beer...and cheese, running is much better when shared.
posted at: 12:54 | path: /bubulle/planet-debian | permanent link to this entry
For lenny users, I also build backported packages. They are not on backports.org as the policy for these "as official as possible" backports is to have the same version than testing.
So, they are provided "as is", for i386 and amd64, on the pkg-samba team's unofficial repository. The Release file on that repository is signed by my GPG key.
Please note that these backports also need talloc from backports.org so you need to add entries for backports.org in your sources.list files to be able to use them.
Since I last blogged about samba packages, the 3.4.8 version was also released. It entered testing without any problem and I also uploaded a backported version to backports.org.
Finally, on May 5th, new versions of the 3.2.5 packages of lenny were updated to 2:3.2.5-4lenny11, currently in stable-proposed-updates. So, they will be part of the next point release of lenny. They fix two important bugs (#575951 for interdomain trust with w2k8 servers, and #538819 for memory leaks in some situations).
posted at: 12:32 | path: /bubulle/planet-debian | permanent link to this entry
The main point was, in general, to gradually resume running activities, after the marathon peak.
Which I actually didn't respect at all, as usual..:-)... I already wrote that my "training program" is mostly made of "run as you feel, with no strongly hardcoded program". So, the week-end after the marathon, I ran 11km on Saturday, followed by an half-marathon (alone) on Sunday which I completed in 1h53'31", very close to my 1h51'31" PB for a "half-marathon alone on trails" run.
Definitely not what is generally recommended, but, eh, this is how I was feeling...
Then, I could continue during the following weeks with my normal program made of 1 or 2 small runs (between 8 and 14km) at lunch time on weekdays, 1 small 10 or 11 km trail run (on this circuit or a longer variant), and a "long" run on Sunday mornings. The long run usually goes from 15km up to 25km.
Now, 6 weeks after the marathon, I'll run a 20km trail named Le Trail des Cerfs ("deer trail"), that starts in my neighbourhood and goes through my beloved Rambouillet Forest. I could have chosen the 35K or even 50K race, but preferred being less ambitious with the distance and rather focus and ahieving what will be my first official trail race. I don't expect much differences with the weekly runs, indeed, as most of these are already done in trails, fields and woods with very few paved roads. My goal for this race is 1h45.
Later on, I still have the plan to run a full marathon alone, probably on May 31st. This time, contrary to last year where I ran one by going through a very wide loop in Rambouillet Forest, I'll do 4 loops mostly in the city of Maurepas, on flat roads, with possible pit stops at my home (so, no camel back bag, but more focus on good performance). I plan to target 3h55 if the heat is not too high.
Further plans are less precise. I'm not sure I'll be able to run the Sur les Traces du Loup trail in La Ville aux Clercs, 2 hours away from our place. That would mean arranging a small week-end there with Elizabeth. Still, that really looks like a great race, with a start at 18:00 to enjoy the end of the day during the longest days of the year. Also, the race and the barbecue that follows appear to be very friendly events in a small French village.
On the other hand, the autumn marathon is decided. This will be the Seine et Eure Marathon, a "small (600 runners) marathon very close to our family house in the Seine valley. The race is one of the flatest marathons in France and has the reputation of leading to great times. That will be on October 17th. Target for me: 3h45 (unless I see changes happening util then).
posted at: 10:56 | path: /bubulle/planet-debian | permanent link to this entry
Today, I completed the backport of samba 3.4.7 to backports.org. We already had that version but it was desynced with the version in testing (2:3.4.7~dfsg-2 in testing while we had -1 in backports.org).
As the difference is noticeable (dropping the old smbfs package in favoir of cifs-utils, which I blogged about yesterday), I wanted to resync all this.
So a new 2:3.4.7~dfsg-2~bpo50+1 package (what a funky numbering scheme, isn't it?) was just uploaded today in backports.org.
The transition between smbfs and cifs-utils should be smooth (as soon as autobuilders complete the entire build which may be quite long, for samba).
I also completed a full resync of packages we have in experimental. Here, we do provide the bleeding edge samba version, namely 3.5.2. It's very likely that we don't provide samba 3.5 in squeeze, the option being conservative and keeping up with the 3.4 branch. So, for our users wanting the last samba version, having it in experimental is interesting.
Next step is building 3.5.2 backported packages. These will be provided "as is", for i386 and probably very soon amd64, on the pkg-samba team's unofficial repository. I'll be working on this tomorrow^Wtoday.
posted at: 09:21 | path: /bubulle/planet-debian | permanent link to this entry
Doing that, I just discovered that Ubuntu bug numbers are about to pass Debian bug numbers. At the moment, I write this blog entry, the highest bug in Launchpad, for Ubuntu, is #575161, while Debian's highest bug number is #580219.
Roughly speaking, Ubuntu highest bug number should be higher than Debian's in about 2 weeks.
Interesting.
posted at: 17:06 | path: /bubulle/planet-debian | permanent link to this entry
Salvo Tomaselli reported the Debian bug #580000 on Sunday May 2nd, against the kopete package.
Bug #570000 was reported as of February 15th 2010. We're now behind our "10,000 bugs in 2 months" pace by two weeks, so this confirms the slowdown in Debian's bug report rate.
I hope this doesn't confirm also the slowdown in Debian's development rate.
posted at: 22:57 | path: /bubulle/planet-debian | permanent link to this entry
posted at: 06:29 | path: /bubulle/planet-debian | permanent link to this entry
The situation for D-I translations is currently the following. It is not that brilliant, I have to say. And I limit details to "level 1".
In very short: 33 languages are in good shape. *26 languages are in danger of being deactivated*
16 languages are complete:
ar bg cs de es fr ga it ja lt nb pt ru sk th vi
14 languages were complete before changes to sublevel 5 and only have 5 fuzzy strings to very seldomly used packages:
ast bn el eo et eu fi gu mr pl sl sq sv tr zh_TW
3 languages have a few strings to update in sublevels 1 to 3 but have active translators who will probably easy cope with them:
be nl zh_CN
da gl ko pa ro he hi hu ka pt_BR ml tl
8 languages were complete in lenny but had no update since then. These languages are at risk of being deactivated for Squeeze if they are not complete for sublevels 1 and 2: bs ca dz hr id km lv mk ne nn ta wo
Other languages are in various situations. Those marked "(P)" are not activated and have nearly no chance to be activated for Squeeze, the exception being Sinhala (si) which is making great progress. The 6 active languages listed below are at risk of being deactivated:
089% uk 1 087% ku 1 086% am 1 083% kk 083% si (P) 2 072% sr 1 069% cy 1 064% is (P) 1 043% ms (P) 1 034% kn (P) 1 025% br (P) 1 015% se 1 014% mg (P) 014% te (P) 2 012% fa (P) 1 009% xh (P) 1 008% ur (P) 1 002% hy (P) 1 000% lo (P)Squeeze release plans have not been made yet but I would say this is time to become aware of the situation and start hunting for translators for "endangered" languages.
posted at: 06:30 | path: /bubulle/planet-debian | permanent link to this entry
I use this occasion to congratulate Christelle Daunay, who broke the French national record for women for the second year in a row, bringing it to 2h24'22", a world-class performance, after finishing 3rd in the so famous New York City marathon.
posted at: 21:26 | path: /bubulle/planet-debian | permanent link to this entry
Upload time has been 3h49'29" (my watch says 27", eh!).
That was a good one, really (actually, I can compare to only 3 as it was my fourth marathon after Paris and Cologne 2009, plus the unofficial Rambouillet Forest marathon I ran alone 6 weeks after Paris 2009).
For marathoners around, here are the chronometric data. Please note that GPS watches tend to overestimate the distance, mostly because of the jerky way one runs in mass events, as well as geometric considerations related to the imprecision of GPS (imprecision may be lateral and thus the line recorded by the watch is often less straight than the real one). For instance, according to my watch, Paris marathon was 42.63km long, so an overestimate by nearly 500meters. As a consequence, the recorded speed of 5'23"/km should be corrected to 5'25". So, times below should be increased by 2 seconds.
Going back home was OK and I had no cramps and I still haven't any. This is also what one gets when preparing well. My only regret is that I didn't run over 30km since, indeed, Cologne Marathon! That made the difference in the last 12 kilometers, obviously. Having spent the winter on many half-marathons (either alone, or a few officials) clearly improved my "performance" up to 25km. For the next marathon (a marathoner always thinks about the next one), I need to have 1 or 2 long trainings a few weeks before.
Now, I have to find out which one will be the next one... and I wish good luck to my friend Noèl Köthe, who is preparing the Mainz marathon for May 9th as well as Dirk who's certainly preparing another one. Guys, rendez-vous is of course for Debconf10 in NYC. We won't run a marathon for sur, but how about an half one? :-)
posted at: 16:51 | path: /bubulle/planet-debian | permanent link to this entry
So, on Sunday will be my fourth one and my second Paris Marathon
.
So, after 4h10mins in Paris (april 2009), 4h12 in Rambouillet Forest (all alone in full autonomy mode) and 4h00'49" in Cologne, my target this time is 3h50.
Weather is announced to be perfect, my physical condition is AFAIK good, the motivation is very high....my only fear is that I registred in the "4h target" group (because, back in November, I wasn't expecting the results I got in February/March in half-marathons). As many people overestimate their target, I expect to be passing hundreds of runners during the first 5 kilometers, which is not the best way to start....but, hey, Paris streets are wide enough (except Place de la Bastille, my nightmare....I hope organizers solved the jam problem there).
Of course, I kinda hope seeing fellow Paris DD's waving giant Debian swirls, including our wanabee-DPL.....but let's stop dreaming. And running in a sunny day all across Paris and along river Seine, will be great again even without Zack. I'll also be missing Ralf and Noèl who ran it last year: Ralf is unfortunately suffering from knee problems and Noèl is preparing for Mainz marathon (target: 3h??).
Anyway, more news on Sunday afternoon....and let's enjoy a full rest tomorrow!
posted at: 19:58 | path: /bubulle/planet-debian | permanent link to this entry
Hopefully my samba co-maintainer will bring the right fix as he's been involved in solving the issue in Ubuntu..and is by definition more familiar with the LP infrastructure than me.
But, sure, that one made me mumble quite a bit this morning...
posted at: 07:06 | path: /bubulle/planet-debian | permanent link to this entry
That post was courtesy of the Smith Project that just hit its 3rd birthday on Thursday.
Free (virtual) camembert to people who spot the right number of English errors in this very post, of course.
posted at: 17:27 | path: /bubulle/planet-debian | permanent link to this entry
Ah, and I have to bring croissants at work this morning. Tradition. Eh.
In short, I'm turning 31 today. Easy one....this year..:-)
posted at: 05:53 | path: /bubulle/planet-debian | permanent link to this entry
This even was organized locally by a team of very motivated Thai Debian enthusiasts and contributors, such as Theppitak Karoonboonyanan (*the* Thai DD), Neutron Soutmun, Kitt Tientanopajai, and all those whose name I'm not remembering as of now (I hope they won't mind).
The even had kinda the structure of DebConfs, with a few days of "Debcamp" to begin. See the full schedule.
We were hosted in Khon Kaen University (KKU), one one the most famous universities in Thailand, a small "town in town" in a city with a few hundred thousand inhabitants (dunno exactly). Lodging was done in a hotel located inside the university. Interestingly, the hotel was also hosting youg students participating to "Summer Camps" (apparently training systems to get good school results) gving to all this a very young atmosphere.
The hacking lab and talks location was a 30-seat room in the university library, and meals were brought in there very efficiently, with the very specific way that Thai people have to transport each and every kind of meal (in small plastic bags closed by rubber).
I arrived only in the 3rd day because I had commitments at home that made it impossible to me to come for the first day. During these days, people have been very busy hacking and participating to the Bug Squashing Party. During that BSP, about 50 bugs have been touched, without about 15 or so closed.
Other non Thai attendees were Andrew Lee from TW, Paul Wise from AU, Daiki Ueno and Yukiharu Yabuki from JP. Organizers were expecting some attendees from neighbouring countries such as Laos, Vietnam or Cambodia. Unfortunately, none of them could come, including Anousak Soupavanh, leader of Lao free software localization efforts, who I was very impatient to meet. Transport difficulties, or visa problems, do not make things easy in that part of the world.
On Wednesday we had a "DayTrip" as it is common for such event. We went abotu 50 km away from Khon Kaen, to visit a nice place, close to a dam lake, and climbed a hill surrounded by a big temple and a giant Buddha statue. Then we had a wonderful lunch in a fish restaurant in the very specific Thai way to share stuff: everything is on the table and you pick your food here and there, at you rconvenience. Of course, local advice before trying apparently innocent food is always worth it because the fire might be hidden anywhere (for instance in that soja plate which I tried and that set my mouth as a burning hell for 20 minutes). The journey ended by a visit of a great temple in Khon Kaen and, very noticeably by a dinner in a very popular barbecue restaurant in "all you can eat" style for...100Bath (so, about 2.5 euros). Maybe only vegetarian people had more trouble enjoying the meal as it was mostly made of various meat (and sea food).
The talk days were very intense, at least in my opinion. Probably because I ended up giving four talks, some of them completely improvised (about IP-over-DNS, which I was using at the hotel and about which many wanted to learn a little bit more, and GPG keysigning processes). It turned out that the GPG talk was well received and, discussing with Paul later on, we agreed that such a talk, mostly meant to explain the DOs and DON'Ts For good GPG keys signing, could be a good idea even for Debconfs.
There were also a few talks about local initiatives and efforts to develop (and not only promote) free software. We have no recordings of these talks as we were infortunately missing some video recording installation (maybe next time, Thep) just like the miniConf that was happening in Panama at about the same time was having.
Due to local regulation on the university network, we had some limitations with Internet access (some firewalling that for instance was preventing SIP to work properly, which made a video-conference with a japanese user group fail, unfurtunately).
The event ended in a round table discussion about ideas to organize something bigger in the future. The local community in Thailand has apparently the energy, maybe ressources and local support to be able to organize a slightly bigger event as first try (somethign like an Asian DebConf or something similar, targeting mostly Asian contributors and about 50-100 people. Thailand seems to be a good target to host such event, with many things being relatively inexpensive (and not only beer!). And they even think about possibly hosting a Debconf at some time in the future (actually, Martin Krafft should also be credited for bringing this idea). That isn't as crazy as it seems and, provided that potential organizers start involving themselves in the current Debconfs, everything seems to be possible.
After all, if we look back to 2005, only one person (hello, Safir) was seriously thinking that Debconf could really happen in Bosnia and Herzegovina, right?
After this week (followed by 2.5 days of sightseeing in Bangkok for me, plus a small meeting today with local Thai Linux corporate users and IT company owners), I feel like the mood in Asia about Debian development is high and full of potential. The miniconf last year in Taiwan was already a good success, by establishing a good connection between people.....we need to keep that alive and, hopefully, there will be other miniconfs in this part of the world. And, well, if I can be there, I'll be there.
posted at: 19:41 | path: /bubulle/planet-debian | permanent link to this entry
Paris half-marathon was great. Freaking cold (-3°C with a strong North-East wind) but great weather with the sun in a blue sky.
I didn't see Ralf (and, Ralf, sorry but I was finally wearing an orange suite..:-)) nor Debian swirls....but, apparently, that wasn't missing for having a great race. Elisabeth cam along with me to Paris and I managed to see her (and she manage to spot me) twice: once in Bastille at km 10 and later on at Charenton at km 14
The interesting challenge now is what to set as goal for the marathon. Of course, I have to run below 4h (which I missed by 49 seconds in Cologne) but I apparently can do better. I think I'll be reasonable and set 3h50 and we'll see.
posted at: 15:26 | path: /bubulle/planet-debian | permanent link to this entry
This year, I'll run both, so tomorrow, I'll be running in Paris (and Bois de Vincennes) streets...along with 27,000 other people..., attempting to beat my 1h43'55" record set in my last (and as of now) only half-marathon, one month ago, in Bullion. That might be hard as running at one's own pace is easier in a run with 1,000 runners..:-)
Ralf Treinen told me he might show up around km 7 (place de l'Hôtel de Ville). I should be around there around 11h50 as the start is at 10h and I should probably cross the start line at about 10h15, then run 5mins/km.
So, if you're around eastern Paris and wake up "early" on Sundays, feel free to come and show me a nice Debian swirl so that I can run even faster (or less slow). I'll be wearing a black top with short sleeves and short running pants (not the easies to spot, I admit). Then try spotting Ralf and have a beer..:-).
posted at: 18:03 | path: /bubulle/planet-debian | permanent link to this entry
Last week-end, I booked the whole set of plane tickets for the Perrier family to fly to USA next summer:
posted at: 18:31 | path: /bubulle/planet-debian | permanent link to this entry
Ron Murray reported the Debian bug #570000 on Monday February 15th, against the gnome-keyring-daemon package.
Bug #560000 was reported as of December 8th 2009. We're behind our "10,000 bugs in 2 months" pace by one week, so that's a small drop in the bug reporting rate, this time.
I hope that people remember the Bug #600000 contest. Bets are closed now (the wiki page is still opened for modifications but I'll revert them if they are late bets).
Being 30,000 bugs away from #600000, it's time to image when that bug could happen. The current "2 months+1 week/10,000 bugs" says it would happen around September 10th. With the "2 months/10,000 bugs" that would be August 15th, just after DebConf 10.
The closest bet as of now seem to be either Miod Vallat or Kanru Chen. I probably lost already (I did bet June 30th) unless I imagine a very massive bug filing.
Too bad Matt Arnold, Thomas Preudhomme, Nicolau Werneck and Odd Henriksen, you already lost!
posted at: 08:42 | path: /bubulle/planet-debian | permanent link to this entry
JB began a 3 year cycle to graduate in electronics and industrial computing at Polytech'Paris-UPMC and is increasingly becoming more and more interested in Linux, programming and particularly low-level programming in embedded devices (as far as I can understand this stuff). As one can see, he's on his way to become a much clever geek than his dad.
It then became natural for him to consider attending Debconf even though he's not (yet) involved in Debian.
So, you folks will have to suffer having two bubulles instead of one. That should make a nice picture if Bdale and Elisabeth are there too..:-)
See you in NYC!
posted at: 21:07 | path: /bubulle/planet-debian | permanent link to this entry
We were then hit by a quite severe bug where files ACLs are messed up on Excel files after saving the files (upstream #4308, fixed upstream in 3.2.8).
I will probably consider fixing the official Lenny packages (there is no bug reported about this in Debian, though) along with a few other quite severe ACL-related bugs (Mathieu Parent, who maintains ctdb, pointed me to a few such bugs).
However, I couldn't wait for such a long time and backporting the fix for #4308 isn't completely straightforward. So, I decided to upgrade our server to the latest Samba 3.2 version, namely 3.2.15 (upgrading to 3.4 series could bring other kind of regressions and we have no time to properly test everything...users already have their guns handy).
This is where the Debian Samba packaging team unofficial repository becomes handy. It is meant as as service to users who want to run the very latest samba code and have the same packaging choices we have in supported Debian releases. We already had 3.2.14 there but I never took time to update to 3.2.15. This is done now.
Moreover, as our production server uses the amd64 architecture, these unofficial backports are now available for both i386 (my usual build machine) and amd64.
Using this repository is as easy as:
deb http://pkg-samba.alioth.debian.org/packages lenny-backports/ deb-src http://pkg-samba.alioth.debian.org/packages lenny-backports/
Please note that this APT source does not use GPG signing of Packages files. When I am less lazy, I'll try to at least sign it with my GPG key.
posted at: 11:44 | path: /bubulle/planet-debian | permanent link to this entry
They mostly seem to be related to my dual screen setup at work (no crashes at home) and/or VirtualBox use (I use VB at work to host my work Windows machine).
Of course, as expected, X.org 7.5 migrated to testing and now I'm screwed. I still have the crashes...and can't even easily downgrade..:-(
Apart from beating myself, would there be a chance that someone still has these packages somewhere? I probably need 7.4+4 packages.
posted at: 06:58 | path: /bubulle/planet-debian | permanent link to this entry
Apparently, from my cumulated experience, joking about funny translations is a very widespread game among german-speaking developers (of course, as everybody knows, us French do translate everissing and no single Frainche DiDi will make such fun abaouttel10n wouhorke, aha).
I'd say: more generally widespread among people .who jargonize about computing so often in their daily work that they often forget that not everybody, even in a country where people are as clever about English as German/Austrian/Swiss folks are (we all know that 100% population in these places speak a very perfect English) do understand what "downgrade" means.
Seriously, "funny" translations are our (translators) daily nightmare. We hate them because we know people hate us (or make jokes about our work) when they face them.
The example taken by Patrick is a nice case as "downgrade" is tricky for all of us. For instance, French translators have chosen "revenir à la version précédente" as "bring back to former version". This is about the only "right" way to say what's being done ("dégrader" is another option but that has a negative connotation). Of course, I often *hear* people (incl. me) say "downgrader" which is an horrible neologism....and of course doesn't say anything to people who are clueless in English.
English is, unfortunately, often a very compact language. This is our daily pain.
So, as German translation of "to downgrade", what do *you* propose, Patrick? "Downgradieren"?
Tricky, eh? Taken from APT for "downgrade":
That sounds funny to you? Fine. But as usual in free software, please report.....and patches welcome...
posted at: 18:52 | path: /bubulle/planet-debian | permanent link to this entry
After all, it's only 3.5 years that it's sitting here waiting for me..:-)
Looks like I'll have to do an upload for geneweb which I'll do happily as it will drop what I consider a quite bad hack.
posted at: 06:23 | path: /bubulle/planet-debian | permanent link to this entry
That release will unfortunately be the first to use a "~dfsg" suffix to upstream version. In short, upstream used to distribute RFC files in their tarball. They first removed them when we pointed them at this (because we had an RC bug by our zealous "non-free documentation" hunters). However, they were reinstated after a discussion in upstream lists where it was pointed that these RFC files were needed for Samba 4 sources (that are distributed along with Samba 3 though not used to build Samba 3 packages).
So, well, even though I don't feel really concerned by cutting hair in 4 about documentation files freeness, I fixed this by repackaging the upstream tarball without these files.
I did choose to use "~dfsg" as suffix so that someone wanting to upgrade from our packages to some unofficial packages using the standard versioning (such as SerNet packages) can do it.
posted at: 06:25 | path: /bubulle/planet-debian | permanent link to this entry
Along the 15.4km running path, I took "some" pictures, so you can run virtually with me without even freezing.
Admittedly, most pics are slightly (or highly) fuzzy which proves that I did shoot the pictures *while running* (except a few!).
posted at: 11:02 | path: /bubulle/planet-debian | permanent link to this entry
At least one is scheduled, of course Paris. The highest probability as
of now if Berlin in September (probable enough for me to register as
soon as it becomes possible)...which would also leave a possibility to
run Donostia in late November if I manage to convince Elisabeth that
running 3 marathons in a year is not entirely crazy. I'll also try
again to run a solo marathon in late May in forêt de
Rambouillet (: I
still have to see some wallabies that are said to be living here).
Other planned runs include the Semi de Paris
(Paris half-marathon), which is twinned with the Paris marathon, 5
weeks before, as well as a much smaller one, the semi-marathon
des Lions in Bullion, in the Rambouillet forest, where I had a memorable
run on Dec 17th on a snowy day. That will make 3 mass runs in a
row, one per month, culminating to Paris marathon. Sounds like a good plan.
So, since I ran the Cologne marathon, my plans were mostly to stay
trained enough. For this, I went back to me regular schedule of one short
10km run on Saturday morning, then a longer one on Sunday morning,
plus hopefully small runs with colleagues, or alone, at lunch break.
The most interesting things to notice during this October-December
timeframe were then:
The goal right now is to both enjoy winter runs in snowy landscapes as
we are very lucky to have a real winter this year in Paris area (I'll
try to shoot some pictures tomorrow and post them on my
gallery)....and properly prepare the Paris marathon, where my goal is
to break the 4h barrier and go down to about 3h50 (some might say I'm
not ambitious enough).
A long term goal (probably 2011) is the well-known SaintéLyon
long distance night run, starting from my birth city, Saint-Étienne:
69km, departure at midnight, 1300m positive slope (1700m negative),
up to 850m altitude from 500m, 55 editions. This on the first week-end
of December, which in that season may mean "snow" (the 1990 edition
was stopped at the 2/3 of the run because of a 70cm snow fall, but 80
heroes completed the whole run anyway). That run will be the next
Grail, as it seems.:)
All this, plus many others cumulated about 375km in 3 months which is
fairly what I planned to do.
posted at: 14:38 | path: /bubulle/planet-debian | permanent link to this entry
It seems that several of these suffer from low interest from their respective maintainers, so that task will often involve checking whether the said font is still wished in Debian, then talk with the maintainer if there is still one to convince him|her to maintain the package collaboratively....or adopt the package when it is clearly abandoned.
As of now, I already imported 4 ttf-arphic-* packages in our SVN and I'm working to modernize them.
Dear readers, please feel free to point me to good candidates for this. Otherwise, I'll process the output of "apt-cache search -n ttf-" alphabetically...
posted at: 11:52 | path: /bubulle/planet-debian | permanent link to this entry
And, finally, in a very last rish, I ùade it and completed my remaining 541 mails to review in the bloody debian-user mailing list.
By the way, you could be interested in learning that I flagged all messages from the "Sponge Burning|A Republican!" giant thread (probably the biggest thread in Debian mailing lists ever) as spam. That's debatable but such huge crap is definitely not something I personnally think belonging to the Debian mailing list archives..:-)
To others: we still need a few other crazy minds to continue cleaning up spam from our lists archives (at least, debian-boot is in the way to become the first entirely spam-free mailing list on lists.debian.org).
posted at: 16:17 | path: /bubulle/planet-debian | permanent link to this entry
Well, sorry to disagree, but there's nearly nothing true in this, IMNSHO. I have never noticed that family name capitalization is more widespread in my own country (which I pretend to know in details, when it comes at its culture) than it is elsewhere. This is more personal habits than anything else. The only thing I can see where one is instructed to use capitals are manually-filled forms, where capitals are often the best way to avoid misreadings.
I though my main point was clear: I don't like name capitalization. I just see it as a practical way to silently tell people "*this* is my family name and *that* is my first name". Actually, this is exactly what several Asian contributors are doing on a daily basis and, from what I've been told, for that exact purpose. So, my "proposal" is just returning them the favor and give them a chance to call me "Dear Christian" and call you "Dear Wouter". Let's say this is a way to show them some consideration. This is somewhat ugly but unless someone has a better proposal, I'll stick with it.
And, on that matter, well, I think that Wolf's proposal to do the opposite would probably lead to weird results. Though let's admit that Gunnar has a point: there's often more than just "first" and "family" names in many places (including mine as my full name is "Christian Jean Antoine Perrier") and that make things pretty more complicated. Maybe we should all call ourselves "dude" or "pal", after all..:-)...but that again is "western" style habits. In some places, that would obviously be considered very rude...even in our friendly free software community.
I just had the case 10 minutes ago, by the way. Had to answer by mail to the Khmer translator of D-I whose name is "Khoem Sokhem". And it just took me quite a while to dig into my archives and find out that I should use "Dear Sokhem" (plus another 10 minutes to confirm that he is a "he".... as I previously worked for months with the former translator before learning that "he" was indeed a "she"...quite a shame for someone who pretends to have some awareness about gender considerations in free software, sin't it?).
Anyway, thanks for this "conversation" and Happy New Year to all my friends around. At least for those who celebrate a new year in the upcoming hours..:-)
posted at: 08:56 | path: /bubulle/planet-debian | permanent link to this entry
As many of my readers may guess, being involved in internationalization means interacting with people of different cultures.
And, when it comes at names, this is just a nightmare..:-)
Most so-called "westerners" (western of what, by the way?) are used to the common tradition we share to use people's "first" name....at first...followed by what is most often the "family" name (the one we usually get from our parents).
As a consequence, it is fairly easy for me to guess that, interacting with someone named "Barack Obama", I can use "Dear Barack" if I'm in position to use his first name and "Mr Obama" if I need to show some respect|distance|whatever.
Also, as most "western" names are quite familiar in my ears (international culture, blah blah), I can easily guess what is a first name and what is a family name (though, here, "Barack" is probably not the best example). Gender may become a little bit trickier in several cases, but, thankfully, in English, "Dear" does not change with gender (but the third person does whic sometimes leads to problems).
The problem arises when interacting with many other culturally different people such as my friends in Japan, China, India, Africa, etc. Here, dammit, the habits might vary a lot and things get harder.
Should I call my friend "Kenshi Muto" as "Dear Kenshi", "Kenshi-san", "Muto-san" or whatever? From my readings, I see that most Japanese people do put their family name at first. So, I should then call him "Dear Muto" if I'm very close to him (oh, maybe it's 'her', by the way, who knows?)...or maybe "Kenshi-sensei" because he (oh...or she) is a respected figure in the Japanese FLOSS community...or whatever. Damn!
Thankfully, I know Kenshi for years, we're good friends and I know that he's using the "western" way to write his name. I also know he's a man : after 5 or 6 Debconfs, you know everything about people!
But, of course, it becomes harder when it comes at people I never had contacts before.....and I don't want to appear as impolite, or silly so, then we jump into my daily "nightmare". Using Wikipedia for hints about Chinese, Indian (the many ones) or even Icelandic or Indigenous peoples of the Americas helps a lot, still. I can only encourage my various friends in free software to do the same. We probably all deeply appreciate when someone shows some knowledge of our culture or at least tries to do his|her best to understand our culture. That is probably also part of the mutual enrichment we get in our free software activities.
As a tribute for this, I now write my own name as "Christian PERRIER" in my email headers, so that people have a slightly better clue that "Perrier" is my family name and that someone who wants to be familiar with me should call me "Dear Christian".
I can probably ncourage my friends all around the world to do the same. For once, using capitals is not about 'shouting' but more about helping other people to figure things out.
Next blog writing about the interesting challenge we all face in free software meetings and conferences: shake hands, kiss, hug or just wave? :-)
posted at: 09:12 | path: /bubulle/planet-debian | permanent link to this entry
Apparently, after TeXLive packages changes, this one will not be the only one discovered by Lucas automated archive rebuilds.
Anyway, Merry Christmas to those of you who celebrate it.
posted at: 09:26 | path: /bubulle/planet-debian | permanent link to this entry