Logically, Argentina won against Scotland tonight after a.....quite boring match. Scotland, as expected for them unfortunately, wasn't able to put speed in their game, the only way to beat Argentina this year. Even with that poor game (Lamont, their rear player, was a disaster), Scotland finished quite close to ARG, who didn't play brilliantly with many approximations (Hernandez being a bit arrogant, trying to kick too many drop goals) and a bit of luck. So, nothing really impressive in that match. Does that mean that South Africa has better chances to qualify for the final: who knows? Argentina can be unpredictable...but it seems they are a little bit burned out (they don't really have 30 top players and some of their key players appear tired). The semi-finals will feature France-England (about 50-50 chances from what we saw up to now) and South Africa-Argentina (60-40, in my opinion). Next week-end, Stade de France.

...but, really, Fiji would have deserved the victory. They did put the Stade Velodrome on fire this afternoon, pushing South-Africa to play....a quite bad rugby with few imagination and many errors. In the RSA team, only Steyn emerged and saved his team from sinking (37-20). RSA scored 5 tries mostly only because they have good four-wheel-driven "trucks". In the Fiji side, Sisa Koyamaibole and Mosese Rauluni lead their team to even the score 20-20 on 65th minute...unfortunately, losing in the last 10 minutes despite the support of the entire French crowd. The Springboks need to improve their play dramatically if they want to have a chance against Argentina (very probably unless Scotland realises a miracle). My conclusion today: the Three Nations from southern hemisphere (RSA, NZL, AUS) should really consider opening their annual tournament to other nations such as Samoa, Tonga, Fiji or Argentina. That would certainly force them to play another game than their usual game. Of course, that would mean being less attractive to players (just count the number of players from Samoa, Tonga and Fiji who play with New Zealand and Australia) but considerably widening rugby in the Pacific area.

RSA leads by 13-3 against Fiji. They played very slowly...and lost tons of balls during 1st half. IMHO, the Springboks should be scared. Another team was leading 13-3 yesterday evening...:-)

...and I like that one, as one may expect. With only 30% ball possession, France won against New Zealand. They transformed the very few occasions to score tries and finally, thanks to a magic inspiration by Fred Michalak, relayed by Yannick Jauzion, their last try made them win 20-18. The french defense was....close to heroic (179 tackles, whiles NZ tackled 36 times!), particularly at the end of the match, where the NZ bulldozer was trying to score one very last try. Not everything was perfect in this match (particularly a disaster in line-outs: I have never been fond of Rafael Ibanez, but it seems to be time for him to retire...), and only the courage made them win. I really wish I could have been in Cardiff to celebrate this. Champagne... So, rendez-vous to our English friends next week-end, in Stade de France. Given their match today, that will be though, again. Tomorrow, South-Africa-Fiji (yet another suprise?) and Argentina-Scotland. Eh, why not France-Argentina as final?

At Debconf, we (the i18n crowd) discussed about the NMU campaign we ran before the release of Etch (roughly from September 2006 to March 2007). We concluded that running a campaign slightly earlier in the release process would be likely to save us some work. Indeed, experience shows that the initial pings sent to maintainers often trigger uploads to fix the pending l10n issues. So, let's start thinking about it now. For those who don't remember, the rough process is described here. As one sees, it is deeply respectful for maintainers and takes great care to avoid unwanted NMUs. So, to begin, please check the list of the pending po-debconf l10n issues ranked by...a pretty complicated mathematical formula which aims to put packages which would most benefit from an upload in first position... In case one of your packages is listed near the top of that list, just consider uploading it soon. PS: of course, the mathematical formula includes a secret pondering scheme aimed at giving French translations more weight, as anyone would have guessed.

The English front players were the key of that though match. By dominating the Australian players, particularly in a few key scrums and mauls, they logically and brilliantly won that first 1/4th final. Despite a non optimal Johnny Wilkinson in kick play (maybe the only question for England in that game), England qualifies (12-10) while Australia will probably regret a few lost balls by their rear lanes in 2nd half time. Anyway, congratulations to both teams for this very intense first match of our rugby marathon for this week-end. Now, let's hoipe we'll have the same intensity tonight for the much awaited New Zealand-France in Cardiff....and of course, I hope for another victory for a northern hemisphere team.

Great play by the English team for first half of ENG-AUS. They would deserve leading the score...if only the Australians didn't have a magic inspiration and scored on try. Anyway, great rugby for rugby lovers in that first half-time. Go England and "Swing low, sweet chariot"....

MJ, each French NPO is free to establish its own regulation rules, which may include a differential voting system if they want to. That should of course respect the French law (for instance, no discrimination based on sex, religion or any other genetic factor) but, as far as I know, nothing prevents deciding that "personnes morales" (ie non physical people) have a higher or lower vote than others. However, this is certainly not common, but many NPO simply don't have non physical members anyway (some don't authorize non physical people to be members). For those who care, Debian France (which is a NPO, but not regulated by 1901 law, but by the special NPO law for Alsace and Lorraine, inherited from the 1810-1918 years) admits physical and "moral" members with an equal voting right. The only special case is a rule that allows Debian developers to have a veto right on any decision of the organisation. See that official regulation rules of Debian France for details (French only). PS: Debian France is currently inactive, due to a longstanding dispute with the French Tax collecting organisation (Direction des Imp ts) who initially refused to consider us as a non profit organisation...because "we sell an operating system" (doh!).

So, now we know what the 1/4 finals will look like:

• Australia-England, Sat 6th, 13:00 UTC, Marseille
• New Zealand-France, Sat 6th, 19:00 UTC, Cardiff
• South Africa-Fiji, Sun 7th, 13:00 UTC, Marseille
• Argentina-Scotland, Sun 7th, 19:00 UTC, Paris Stade de France

France wons quite easily against Georgia (64-7). Nearly perfect match for the French team, except maybe in mauls where cleaning was not always perfect and when setting the ball back in play ("remise en jeu", I don't have the exact English technical term for this). Georgia was playing 4 days after their win against Namibia, which is pretty short for recovering. Lionel Beauxis, for his first selection in RWC, was elected MVP of this match, after a brilliant kick play. Argentina had to either win against Ireland, or lose by less than 7 points. They very brilliantly won (30-15), scoring 2 tries while Ireland scored 2 tries as well, but much less penalties. The Irish couldn't prevent Argentinians th play their game (sloooooooow but efficient). The key to win against Argentina is playing fast with rear lanes and prevent them to slow down the game, which the French failed to do in Paris. As a consequence, there is a great probability that the four major nations of southern hemisphere reach the semi finals:

• Australia had a much more complicated pool than New Zealand but they played all their matches very strongly, without much weakness. I don't really see England beating them, given the current status of their team. But, well, who knows? Noone woud have bet on England in 2003.
• New Zealand seemed invicible in pool play, but their opponents were pretty weak. The weakness of NZ seems to be their defense and maybe the lack of hard fights in pool play. France has still many weaknesses and only a perfect match can lead them to victory. We all know that the French rugby team is always playing its top rugby when not supposed to win (the NZ team probably remembers Oct 31st 1999 in Twickenham, 31-43).
• South Africa does not probably have the most brilliant game but is certainly the hardest team to beat, as they very rarely show weaknesses, contrary to Australia and New Zealand which sometimes play "the wrong side". I really don't see the Springboks fail against Fiji, though the opposition will be really interesting.
• Argentina is probably much too strong for Scotland, which I really don't see winning. Will I be proven wrong?

Rendez-vous samedi Marseille! PS: South Africa-USA will be the last pool match tonight. I'll probably skip it anyway.

Yesterday, New Zealand crushed Romania (85-8), just like they crushed all other teams they faced up to now. Of course, my only hope is that they stop crushing teams...as France is very likely to be the next one. Indeed, the only question about New Zealand is about their defensive skills as all teams they faced scored one try against them. Particularly, their wing players seem to be pretty "weak" in defense ("weak" in terms of New Zealand...:-)). I haven't seen anything of Australia-Canada (37-6). Apparently, the Australians were pretty quiet (6 tries, still!). Very unfortunately, I haven't seen THE match between Fiji and Wales (38-34). The Fiji team created an "half surprise" by qualifying for 1/4th finals. All reports I read talk about a wonderful match, with great indecision, a full commitment by both teams, and 9 tries. Sigh. I screwed programming my VCR and we were strongly disappointed this morning. Finally, Scotland managed to qualify after beating Italy 18-16. Up to the very last minute, Italy hoped to qualify but Chris Paterson scoring 6 penalties out of 6 attempted converted all occasions the Scots had to score points. I'm slightly disappointed for the Italians who played great games all along this World Cup and would have deserved to qualify and meet Argentina, France or Ireland. Today, the final 2 teams for the 1/4th finals will be determined. Argentina has great chances to finish first (they only need one bonus point) and meet Scotland. France needs to win against Georgia. Scoring the bonus point would secure the qualification for them. Ireland has little chances to qualify as they need to beat Argentina without Argentina to score a bonus point, therefore with more than 7 points difference. The most probable ticket is Argentina finishing first and France 2nd (thus meeting New Zealand).

Nicolas Fran ois just announced that we (shadow package maintenance team in Debian) are taking over the maintenance of shadow upstream. Shadow is the source package that builds the passwd and login packages. login provides.....login, while passwd of provides 'passwd' but also several low-level user/group maintenance utilities (useradd, userdel, etc.....not to be confused with higher level utilities we have in Debian such as adduser). Since March 2007, Tomasz Klockzek (insert some specific Polish letter here and there), the former upstream author, nearly vanished, probably too busy with his RL and professionnal activities. Shadow is in a pretty good state, thanks to Tomasz great work and the great collaboration we had last years with him. So, it sounded quite logical topropose taking the maintenance over. Therefore, we proposed this to maintainers from Gentoo and RedHat. While none of them really want to maintain the software, they're OK to collaborate if we're taking the "lead". The shadow "team" in Debian is currently made of Nicolas, Christine Spang and me. We might need more help, particularly from people involved in PAM-related stuff and skilled in C programming and security. As of now, I don't expect us to do more than trying to merge in patches used in Debian/Gentoo/RedHat and maintain the code without adding features. The main contact point for shadow maintenance is the pkg-shadow-devel mailing list on Alioth.

This project that sounded like a joke when a French person announced on April 1st that he plans to review English in some texts of Debian apckages, is now well going on. Contrary to what the status page shows, this is far from being a single-person project. I receive invaluable help from a few people, with a very special mention to Justin B Rye and MJ Ray. Both of them (and a few other people, including the package maintainers) carefully review my rewrites for the considerable amount of Frenglish I introduce. One of the funniest part is reading both of them sarcastically commenting on the "dialect" we're using as we decided to go for US-style spelling and writing style (by far the most used in Debian) while MJ and Justin are definitely citizens of the UK (and very clever in GB English). But both "play the game" ('jouer le jeu' as we way in French) very constructively. I think that the improvement for packages at the end of the process (a very long one as it includes translation updates for debconf templates) is noticeable and nearly always much appreciated by package maintainers. We are indeed only limited by time. I have decided to begin with packages that use debconf (a few hundreds) and I cannot start more than 1 package per day (the initial review is very time-consuming: about 20 minutes for one package and up to 1 hour for some of them). So, of course, new contributors will be welcomed (even those speaking dialects!).

My week-end's TODO list: continue the reviews for the Smith project (resumed when coming back from holidays: there are 19 reviews under work as of now), work on a few DDTP translations, work on my possible participation to FOSS.in (I already registered....hope I'll be able to get some funding, there are really interesting things to do over there), and work on the samba package (upstream's 3.2.0pre1 released yesterday). As usual, too much things to fit in 2 days...especially with a lot of interesting rugby to see..:-)

Martin Michlmayr reported Debian's 444444th bug yesterday, for those who like nice-looking numbers. This bug was reported against a Debian Installer package and closed a few hours later by an upload by Joey Hess. As one sees, even if the D-I team is still a little bit jerky after Frans Pop resignation as Debian developer and D-I release manager, the maintenance of the code is still happening. J r my Bobbio and Otavio Salvador, our new release managers, are still learning and building their knowledge to be able to release a beta version in the next months. New interesting mark in Debian's BTS is bug number 500000. It took 10 months (309 days) to go from 400000 to 444444. That means 695 days at the same rate to reach bug #500000 which means one can expect bug #500000 to be reached on Oct 18th 2008. But we all know that the bug report rate is not linear in Debian....:-) Update: thanks Frans for correcting my jerky English: "resignation" is better than "resignment".

I use a rather simple CRM114-based setup to handle spam in my incoming mail. Nothing really funky here as I essentially used examples from the Debian package's examples/ directory. The only personal touch is using a two-step method for this. All mails are received on my home server and delivered to the "bubulle" user there. A non aggressive CRM114 setup runs there for this user. Mails identified as spam are sorted in a specific folder on the server and are theoretically examined regularly. Indeed, this is something I can't do anymore, given the number of identified spams (about 1000 per day, most of them in Debian mailing lists and Debian BTS). After a quick look, this spam is re-teached as being spam to CR114, then stored as backup in a giant mail folder, which allows me to retrieve a possible false positive later. All other mails are left in /var/mail/bubulle waiting for my laptop to grab them. This gives very few false positives, but still lets a lot of spam slip through. This is why I also run CRM114 on my laptop when mail is grabbed from my home server (I use fetchmail between my laptop and my home server, wherever I am). There, spams are again identified (more aggressively than what's done on my home server) by CRM114. The identified spams are sorted in a folder which I routinely investigate carefully for false positives (there are several a day, but the numbers are low because of the prefiltering done on my home server). Spams are teached as spam to CRM114 and, of course, ham is teached as "ham". I recently added the following to my mutt setup so that SPAM/UNSURE/HAM mails are easy to identify in mutt's index:

spam "X-Spam-Status: Yes"                     "90+/SA"
spam "X-CRM114-Status: SPAM"                     "90+/CRM"
spam "X-CRM114-Status: UNSURE"                     "40+/CRM"
set spam_separator=", "
color index red       default    ~H90    # Spam
color index cyan       default    ~H40    # Unsure

With this, messages identified as "Spam" are shown in red in the index, "unsure" messages are showed in cyan and regular messages have their normal colors. This is really helpful as I can spot messages that are "unsure" and very quickly tell CRM114 that they're ham or spam, with:

macro index \es ""s/home/bubulle/Mail/Administration_systeme/Spams\ncrmlearn --learnspam\n" "crm114 learn as spam"
macro index \eh "crmlearn --learnnonspam\n" "crm114 learn as ham"
macro pager \es ""s/home/bubulle/Mail/Administration_systeme/Spams\ncrmlearn --learnspam\n" "crm114 learn as spam"
macro pager \eh "crmlearn --learnnonspam\n" "crm114 learn as ham"

with the "crmlearn" script being:

#!/bin/sh
 grep -a -v "^X-CRM114"   \
 $HOME/.crm114/mailfilter.crm -u $HOME/.crm114/ $1 --force   \
 grep -a "^X-CRM114"

Shortly speaking, for any cyan message I see in mutt, I hit "ESC-h" if I consider it to be ham or "ESC-s" if I consider it to be spam. Finally, all mails identified as spam are grouped in a dedicated mail folder (/home/bubulle/Mail/Administration_systeme/Spams above) which is sent daily to my home server, where it is used to teach CRM114 about spam, then deleted on my laptop. All this makes spam handling pretty easy with a not so complicated setup (the trickiest part is the nightly laptop->home server synchronization).

Great match tonight between England and Tonga. I already blogged several times that England is not among my favourite teams but tonight they played very well, with a very inspired Johnny Wilkinson (except for kick goals). Their 1st try is a complete magic inspiration by him with a penalty played quickly directly in the arms of Sackey. Wow. Tonga began the match by challenging the English in physical play and some of ou neighbours probably feel like they had a big train passing over them. As long as Tonga could play wide game with their rear lines, they were close to England and even lead the score for a while after their 1st try. The physical challenge was finally won by England, which scored 4 tries and really played as they should have to win their ticket to meet Australia in 1/4th finals in Marseille. Tomorrow will feature two virtual 1/8 finals: Wales-Fiji and Scotland-Italy (in St-Etienne!). Yet again a long day for rugby fans. I bet for Wales and Italy....we'll see if I'm correct. New Zealand-Romania and Australia-Canada will be much more anecdotic as they have no real importance for qualification. I think I'll skip those two..:-)

Sunday was a quiet day for the top teams. Neither Australia nor New Zealand were in danger against Fidji (55-12) or Scotland (44-0). I'm still impressed by the australian team. They didn't show much weaknesses against Fidji, despite the strong physical contest bringed by the Fidjian forward players. The australian rear lanes were impressive, particularly Matt Giteau, scoring 27 points and designed as MVP for Australia-Fidji. The All Blacks crushed a B team from Scotland. The scottish staff decided that winning against New Zealand is nearly impossible and therefore they choosed to have their key players resting, which is a wise choice, but not a gift for the Murrayfield supporters. As a consequence, New Zealand quitely won the match and therefore qualifies for 1/4th finals (and they'll win their pool unless Romania plays with 30 players against them....). No match for yesterday and two matches today. Unfortunately, Canada-Japan (12-12) in Bordeaux was broadcasted too early for me and I haven't seen it. For Japan, this is the end of 13 consecutively lost matches in RWC for Japan, so kind of victory. I've seen parts of Romania-Portugal. I wished Portugal wins that one as they're clearly the most courageous team in the RWC and certainly, among the non top teams, the most promising one. The Portuguese lost 10-14 even though they lead 10-7 for a while. Quite a deception for me but surely these players will never forget the support of the French crowd (most of these being the strong portuguese community we have in our country). Tomorrow will feature Georgia-Namibia and Samoa-USA. I'll try to watch the latter, which is played in St-Etienne, my welcoming home town (ask the Scots supporter and team!).

After the stunning match between RSA and Tonga, we watched an also very good England-Samoa (44-22). Even if their play was still far from perfect (lack of imagintation, mostly), the English players succeeded quite easily against Samoa who never gave up. Johny Wilkinson's come back is a success, imho, and England will then play against Tonga for qualification on Thursday. The winner of that match will end up 2nd of that pool (assuming that RSA wins over USA) and should logically play against Australia (while RSA will play against Wales or Fidji, this assuming that Australia wins over Fidji today). Argentina had to win against Namibia with the bonus point. They easily made it (63-3). As usual in such situations, Namibia resisted about 30 minutes (and were even leading 3-0 for a few minutes). However, the imprecision of their game (a disastrous kick play) left dozens of balls to the Argentinians who then could easily score lots of tries. The Argentinian were, at the end, obviously sparing their energy for the upcoming fight against Ireland next Sunday. In this pool everything can happen (even Argentina being eliminated, which is however not very probable). From what they showed, the Argentinians can definitely win over Ireland, which would then give them the first rank and an easier 1.4th final against Scotland or Italy. Two matches today: Australia-Fidji and Scotland-New Zealand. Australia and New Zealand have a strong advantage and I don't really expect any surprise here.

If you hadn't the occasion to see RSA-TON, just look around for it, find if it is broadcasted at a later time and watch it. The Tonga team just had 10 minutes of weakness, with South-Africa scoring three tries. But they never gave up, scored two tries and finally ended up 25-30, winning the defensive bonus point and really worrying South-Africa. For sure, the repeated failures of RSA on penalties also helped for this scenario, but Tonga became the stars of this pool and Finau Maka is the star of Tonga..:-) The Bollaert stadium in Lens is more used to the fights of the local soccer team than high level rugby but they supported the Tonga team the very same way they support the "Sang et Or". I think that the Tonga players will remember that one. England can *really* worry.

Thursday featured Wales-Japan which we briefly watched. Easy win for the Welsh who begin to secure down their qualification for the 1/4th finals. They'll just need a victory over Fidji on Sept 29th to do so (but Fidji would qualify by winning so that match will be one of those to watch during last round). Nothing really interesting to say about WAL-JPN (72-10). The difference in the front lines was too obvious for Japan to get any chance to win. However, at least during the first half of the first half-time, the were even leading after scoring one of the best tries as of now in the tournament. Yesterday was a key day for the French team. Victory against Ireland was mandatory. They made it (25-3, two tries) which is by far the most important. That wasn't definitely an easy match but noone was hoping it to be one. The french game is still far from being perfect (sometimes strange kick play and weird decisions by the rear players, sometimes slow assistance to players in rucks and sometimes stupid fouls giving their opponents opportunities to kick penalties (or expelled players as Traille was at the end of the game). On the other hand, some parts of the game were good quality: scrums (two balls won), line-outs (4 or 5 balls won on opponents' throw: great Julien Bonnaire there) and the magic inspiration of Elissalde and Michalak on the two tries as well as Clerc scoring these tries. There is still room for progress for the French, definitely (the same play against New Zealand, Australia or South Africa would be a loss). They're however on a good way. Next step is winning over Georgia, hopefully scoring a bonus point (that may be harder than expected as Georgia showed good defense) and hope that the Irish team will win over Argentina. Again, this week-end will be busy:

• Saturday:
  • South-Africa-Tonga: both teams won their first two matches. Tonga is hoping to qualify, still, so while they probably can't beat RSA, they can hope scoring the defensive bonus point.
  • England-Samoa: England has to win to keep their chances to qualify. This is Johnny Wilkinson's come back and they should do it. We can expect good support by the French audience for the Samoans and very interesting match to watch
  • Argentina-Namibia: I already explained why I think that Argentina is really able to reach the semi-finals. This should be a quite easy match for them with one goal: score the offensive bonus point to secure down their 1st rank
• Sunday:
  • Australia-Fidji: Fidji didn't show a very strong play as of now against Japan and Canada, so I think that Australia will have a quite easy win
  • Scotland-New Zealand: I really fear for the Scots that they are crushed by the All Blacks and they will probably more think about the upcoming match against Italy and probably don't hope that much for this one.