gconf-editor is currently the last of the components under the GNOME team's control that is pending entering testing, and the last blocker for the 2.8 meta-packages to be candidates for sarge. If the s390 gconf-editor build is uploaded in time for today's dinstall, the transition will be complete tonight, and besides uploading a few new upstream releases from the GNOME 2.8.2 release, who knows, the GNOME team might get involved in some new fun...

Still, we are missing a new upstream GDM release, plus evolution 2.0, which got a new RC bug, trivially fixable. We're finding out if Takuo is able to do a quick upload, else we'll do a GNOME team NMU to speed things up.

The GNOME team is still getting big thanks on IRC and in some online forums. That's the stuff that makes us want to keep going. :)

NB: refrain of bitching in comments if that should read virii or viruses. We'll leave that to the language police in the various planets. ;)

After "lunch" on Friday, Steve Alexander found out that elmo's friends, the Conference cleaners, had silently attacked Canonical again. This time, we were missing one big paper with all our notes for the Rosetta presentation during the Mataró Sessions.

Without panicing, we decided to try to rethink the ideas as soon as possible so we could prepare the slides, but some problems in Launchpad had SteveA busy for most of the evening. mako, daf, Steve and I ended up working while we had dinner in a restaurant, thanks to Steve's micro-laptop, where we managed to get everything back in place.

Back at the hotel, daf and carlos kept hacking on Rosetta to iron out the last few bugs before the conference. While they did so, mako, LaMont and I had fun reading past entries of mako's and mika's blogs using the projector, some of which were just too good. I also read my mail and mailed my triathlon team mates without remembering my computer was still hooked up to the projector, allowing everyone to read my private stuff. Yay.

In the meanwhile, some of the guys from Madrid (telemaco, Ismael, etc.) rrived at Mataró, which left mako prepare his presentation with Ismael for a while. Carlos and daf kept banging on rosetta, and at around 2:30AM we started doing a demo exactly as we planned doing the day after. At 3:30AM or so, everything we needed to navigate through was working correctly, so we finally went to bed. No swimming on Friday either. Of course, no running on Saturday. I needed to sleep more than three hours...

Saturday was the big day at the Canonical meeting. Less people than I expected showed up, but there still was a good number of very active participants. While the presentations kept going, SteveA and I finished up our slides and decided which computer we would use for the Rosetta presentation. Given I had to type in a full translation of GNOME Hello, I really wanted a keyboard I felt comfortable with, so we discarded using daf's lappy and used mine, even if it was a bit slow.

After Mark's introduction, Steve and I started explaining what the problems currently are in the l10n world, and the presentation kept going smoothly until we had to demo Rosetta. After importing the template I found out some of the keys in the computer didn't respond. Uh-oh... not knowing what was going on with it, we quickly decided to switch to daf's laptop, and then the fun began... writing accents in a Welsh keyboard is a bit more complicated than usual because you have to use compose, so my typing efficiency decreased quite notably. Also, daf had his Japanese input method active, so every now and then I accidentally enabled it and started writing Japanese stuff, which amused the audience quite a bit. I managed to complete a few strings, and we decided to not finish the whole po as it would take too much time. We exported the file and I guess people could get the idea, but I really was not that happy about how things were going...

After the demo, SteveA and I finished what was left of the slides without using the slides, and luckily Steve remembered the bits, because at that point I was suffering some brain failures due to the lack of sleep and wouldn't have remembered much... People seemed quite interested in knowing more details about Rosetta and what it would be able to do in the future, and we got much feedback during the 20 minutes of debate.

After the presentation, I was interviewed by two girls from the Canal Blau of Vilanova i la Geltrú, which asked me the typical "why are you involved in Free Software" questions. We had a chat after that, as I told them part of my family descends from Sitges, where one of them lives.

I had a chat with Jordi Mas from Softcatalà during lunch, and then wondered if I should go to Barcelona to spend the evening. I decided to wait at the hotel until Sergio left to go back to València, and then back to my room for a long nap. I was woken up by the mobile phone, and didn't know where I was... found that mako was having pizza in his room, and got elmo and Kinnison to order one with me too. After completing Bad Dudes vs. Dragon Ninja again and showing people how President Ronnie goes to have some burgers with the bad dudes (Mark loved that, while elmo thought mako and I should be locked up in different rooms) we went out for some beers, where I met Juantomás and had a long chat, with him and his wife.

On Sunday morning, kiko and I dragged Mark out of bed and went running to the promenade and the beach. Kiko and I went ahead of him at the end of the promenade, through a small trail at the side of the railway tracks, and around 1 km to the north of Mataró we found what apparently were two bunkers from the Spanish Civil war. After wondering why they might have been needed, we turned around and picked up Mark on our way back. The streets were already prepared for the start of Mataró's Half Marathon, which was going to happen that morning. Too bad I'm not fit enough for that...

At 13:00, a small group including lu, stub, Adi, Chris Halls and I took a train to Barcelona, and we walked the way from Arc de Triomf to the Sagrada Família, where we decided to go up to one of the towers. We met another Canonical group (elmo, daniels, jdub, thom, etc.) by chance around the building and we all started to climb the stairways. Soon enough, people afraid of heights (jdub, keybuk, Anthony Baxter, myself) started feeling unseasy up there as at some points the protection bars were nowhere to be found and I couldn't help imagining myself diving into the void accidentally. Also, the stupid doll that Daniel gave to me a few minutes before didn't help because it meant having one of my two hands occupied during the way down. Naughty Little Daniel!

I rushed to my uncle's house, and despite running quite hard for 15 minutes I arrived late as I first started running in slightly the wrong direction. I had a great lunch and at 6 left for Mataró again. In the train I found the Async guys, who had visited the Parc Güell, and I wondered where the rest were. I found out from thom that they probably were in the same train, but in the first wagon.

For a few hours, my head had been spinning for some reason, and I really felt quite sick. I could just think of go to have some sleep, and woke up at nearly 22:00 wondering if someone had not had dinner already. Luckily, the pyGTK BOF people hadn't, so we went to the town centre to look for some place. Apparently my face showed that I wasn't feeling very well, because soon enough Matt described the syptomps of his illness a few days before, and it sounded quite possible that it was the same thing, given that elmo had been over it. It was the Ubuntu virus! After an excellent dinner in an Italian restaurant, I found sjoerd and mako in my room, but we quickly went to bed until today. Again, no running or swimmingpool. kiko has promised me we'll go swimming today. I'm not so sure now, though!

Woah, I haven't had time at all to blog since I arrived here. The daily activity never stops, so it's going to be quite difficult to summarise what's been going on here for the last four days

First of all, it's a pleasure to be with all the Canonical people and the other few guests that have been around the conference since the first days. Mako and I have been spending quite some time just discussing stuff, both technical issues here and there, and the really important things in life like what MAME game is more crackful. I hadn't been laughing so much in a few months.

The conference is as interesting as Oxford was. Every day is packed with BOFs and discussion, with our daily dose of technology presentation early in the morning. Of special interest to me were the "language packs" BOF, which showed how hard it is to update translations on a Debian system without changing the package or recompiling anything; and the branding BOF, which also showed some problems in that direction: changing conffiles without recompiling.

Everyone misses the Oxford days for one main reason: some people, like James Blackwell, are starving, and others are just getting rid of a few extra kilograms. :) There was a proposal to look for the local jail to get some food there, as our "lunchpacks" are really not that attractive. Mark has been heard to refer to them as Death in a bag. You get the idea. Food at night is a lot better because we're free to go wherever we want to look for food, which is working very well.

As for what's my main focus for this conference, Rosetta, it's great to see how things are being polished day after day for tomorrow's debut. Rosetta has a big potential to become a tool that most of the l10n communities won't be able to do without, specially the ones that are starting now or will be starting in the near future. My main activity in the mornings is to make Rosetta blow up, and see how Carlos and Daf fix it. :) Then, I make it blow up again, and they come and clean up after me. It's a fun game. :) Fortunately, it's getting more and more difficult to make things crash lately, and hopefully everything will go smoothly during tomorrow's presentation I'm doing with Steve Alexander.

Of course, I'm also keeping and eye on the Desktop team BOFs, but not as much as I'd like to, as the Launchpad stuff is downstairs. There was a good BOF two days ago about new aproaches to easy software management for Ubuntu. The ideas that were thrown out rocked, and I guess Hoary will be a great advance in that area.

Yesterday I found some time to upload HOWL packages for Jeff, while doing a double-round of alsa-lib (to be completed today with alsa-driver and -utils) and an update of ephy-extensions. In the meanwhile, GNOME 2.8 is getting better every day. According to Sebastien, we're just missing gconf-editor and evolution; gedit made it in yesterday, gnome-control-center, gnome-applets and gnome-utils will hopefully go in today, and this morning's evolution upload should have fixed the last two remaining RC bugs for the 2.0.x release.

Tonight we're hopefully going to the swimming pool, after a failed try yesterday. And tomorrow I *promise* I'll be down in the lobby at 7:15 to go running with lu, kiko and SteveA. I've failed to wake up for two days in row now and I really don't feel too proud about it... Staying up until 2AM mostly every day doesn't help much anyway.

I arrived a few hours ago at Mataró, after a 4h train trip. This was, for a change, one of my best travels in the last few years in terms of preparation and timing. I even managed to be at the railroad station in València around 45 minutes early. Of course, that didn't avoid me forgetting to buy the newspaper or a bottle of water, but other than that, everything went ok, unlike recently in Madrid. ;)

As I walked down the street looking for the hotel, I came across a small, lost group of Canonical people, including Mako, Rob and Matt, and Quique from Sindominio. Mako went back with me to the hotel, and we quickly went out again to find the main group. We didn't, as it seems phoning to an Aussie mobile phone using an American mobile phone through the shittiest Spanish telco doesn't work that well, so we ended all alone in the main streets of the town. We decided to have dinner in a restaurant specialised in ham and other pig products. The food was excellent, and we took our time to finish it all, while we talked about, ahem, our past experiences with alcohol and other substances.

Good news is that Mako will most probably spend a few days in València when the conference finishes. Kinda cool, he needs to see how our government spends money in great buildings, and he can have a look at the great LliureX office, otherwise known as the ZuleX. That stuff will probably give him ideas for a ton or two of blog entries.

Back at the hotel, I met fabbione, bob2, thom, Mithrandir, tbm, garnacho and a few others, but it wasn't long before I went up to my room, where I spent a while listening to Mako's stories until we decided 2AM was enough.

The Canonical meeting starts tomorrow. It's going to be fun.

After a very much needed long sleep, I wake up to discover that joeyh decided that another well known FAQ isn't what Debian really needs, so he put code to base-config that adds the first user to the plugdev group, if base-passwd added a static plugdev group.

Not long after, Kamion said this was acceptable and uploaded a new base passwd. When these two packages hit testing, the problem will be solved. Thanks to both!

In other news, I was a bit too optimistic on the amount of things that would hit testing last night. Unfortunately, a few missing s390 builds are also holding gnome-panel, gnome-applets and control-center, which I guess I can't call "minor bits" too easily. They will be in soon, hopefully today.

While we've been a bit quiet and haven't reported about the status of GNOME 2.8 as it entered unstable, things have been moving in the background at an incredible pace.

This has been the smoothest transition to testing since GNOME 2.0, and unless something unexpected happens today, most of the important GNOME bits will hit sarge during tonight's testing run. During the short life that these packages have had in unstable, we've had to deal with just two release critical bugs, a very unusual mark given how many source packages are involved in a full GNOME upload.

A few bits like gedit, gnome-utils or gnumeric will have to wait some more days before they can enter testing, due to missing builds or dependencies, but the core of the desktop is now ready.

In the end, we've decided to include GNOME Volume Manager by default, even if sarge's default kernel is 2.4. It wasn't trivial to take this decision, as there's quite a few factors that will make it not "Just Work".

gvm depends on HAL, which in turn depends on Linux 2.6. This means that Debian GNOME users will likely want to upgrade to kernel 2.6 if they aren't already using it to get a completely functional desktop, with superb automounting of cameras, USB memory sticks and removable media.

Additionally, we've chosen to enable pmount support in gnome-vfs for sarge, so hal won't be doing scary business with your /etc/fstab. The drawback is that for users to benefit of pmount, they need to be added to the plugdev group, or the won't be able to mount stuff automatically using gvm.

This sucks, but we have no way of handling this in a sane way, while pmount isn't part of Debian's base system or the plugdev group isn't added to base-passwd. People suggested that base-config added the user that is created when Debian is installed to the group, as it does with cdrom, audio and video, but this was rejected by joeyh for a good reason.

In short, we'll have to find a way of pointing our GNOME users that this stuff doesn't work because they lack permissions, but it's hard to do it in a non-intrusive way. I'm inclined to think that this will end up as a well known FAQ, just as the need to add people to the audio group so they can use their sound hardware has been in the past. But of course, the GNOME team would like this to Just Work.

Going back to the testing transition, it's been a fun one, although admittedly it wasn't as complicated as 2.2->2.4 or 2.4->2.6, but still, it's wonderful to see that everything went ok. Go, GNOME team!

(And now, Murphy will come out of his hiding place. He was probably waiting for this post...)

After quite a few unsuccessful attempts to find a PCI wireless card working only with free drivers (binary firmware admitted), I finally got lucky today, and I've been running a wireless link for the last 10 minutes.

When I posted my previous wireless blog entry, joeyh suggested me to just look for pcmcia cards, which are generally better supported, and use a PCI->pcmcia adapter. Given that I plan to buy a laptop RSN, this made sense, because the Airport Xtreme in Powerbooks isn't supported, so I'd need a pcmcia card anyway.

So yesterday I did my hopefully last visit to MediaMarkt and got a SMC2835W, as Sergio suggested. The store had the same card in two different packagings. Following sto's advice, I of course got the box that seemed oldest, to minimize chances of getting stupid new chipset revisions.

Today, my boss Pablo lend me his unused PCI->pcmcia adapter, a TI PCI1410, so all the pieces were in place. As soon as I plugged the card in my box and booted, Linux would instantly freeze as soon as it loaded yenta_socket.ko. Nooo, not again!

A few hours later, and after a few missleading Google searches, manty and dilinger suggested upgrading to 2.6.9, which carried quite some fixes that might be involved. Voilà! No more kernel lockups, and after some minutes of fighting pcmcia-cs, the card was recognised and running.

Finally, it's time to ifdown eth0. My flatmates will be glad to find no more network cable all over the living room when they come back on Sunday... "Thank manty, dilinger and joeyh", I will say. :)

Last week, all the team met with the trainer to talk about some of the season's goals, as always at the start of the season. The most charismatic members of the team expressed their discontent about how the team was being managed, and how our budget spent throughout the season.

The core of the problem is that for years, the University has had an agreement with the Valencian Triathlon Federation to provide a trainer an a project for every season. Roughly half of our ~15.000€ budget went to pay for this, although our trainer (currently one of our own team mates) didn't earn anything near 7000€. Somewhere, for some reason, much of our money was getting "lost" at the federation with no apparent gain for us. We presented an alternative management plan, with total transparency for the budget, everything done by ourselves, etc. which of course implied breaking this tie with the federation, and we called for a new meeting, yesterday, so people to vote to leave things as they were or to go for the new setup. We had started a revolution!

It's obvious the guys at the federation don't want to lose this cool income, so the first thing they did when they knew about our revolution was to phone one of us and warn them that if we went ahead, they would have to take action. What that means, I have no idea, but yesterday it became obvious they didn't just wait for the vote to happen...

So yesterday, at 9PM, we all walked into the meeting room, and let the trainer speak first. Maybe it was a huge coincidence, but he told us that at 6PM, the bosses at the Sports Service had called him and told him that due to a general budget cut from the local government, they had been forced to cut the money for the different sections of the club. This naturally had a big impact on us, as we are one of the most expensive teams in the uni: they have decided to discontinue the triathlon team, and will only support the triathlon activity, which means we can go to train to the University's sport fields and to the swimmingpool, but they won't pay for competitions or anything else, except for the few people that are eligible to attend to the Spanish Universitary Championship (which is a very small percentage, and mostly new people).

The summary is that we've been kicked out, not only the revolution group but also a lot of the new people. Realising we had nothing to do in the new situation, we of course didn't even try to discuss or vote the new plan, and just started discussing what our options are: creating a new club, trying to find some club that might want us in, etc. All of these are going to be a lot more expensive, of course, but oh well... someday this had to finish. If we create a new club, the name will probably be "Komando", which was the codename of the conspiracy group that prepared the revolution. Long live the Komando! :)

And while all of this happened, I've been basically idling, triathlon-wise. All of these conflicts inside the team haven't helped my lack of motivation, so I find myself entering December and not having gone to the swimming pool a single day and just one short cycling training session. The only thing I'm doing is go running with the new people twice a week, because it's fun to know them a bit more. Hopefully if the new team starts I can get started again...

We've spent the last two days carefully selecting what packages we should build and upload in an order that makes life easier for autobuilders.

Yesterday, the first few libraries were uploaded, as well as the user-guide. Some people had problems because a few packages didn't make it into incoming on time so some bits were uninstallable due to gnome-keyring depending on new atk.

Today, we've continued with important libs like libgnome/libgnomeui, libbonobo, eel and gnome-vfs. We've also uploaded a few apps now: bug-buddy and nautilus have hit sid and incoming respectively. As nautilus didn't make it to unstable by a few minutes, you should be careful about doing dist-upgrades today. apt will probably want to remove nautilus entirely, so I suggest you don't do it. :) That, or you pick libeel, libnautilus and nautilus from incoming, which will also work.

Sadly, in the process of building some of these packages, pbuilder left an active bind-mount of my local Debian mirror and I, trying to get rid of the bogus unclean build dir, recursively deleted the mirror entirely. Of course, this has slowed me down a lot, to the point that I had to ask Sjoerd to sponsor libgnomecanvas for me while I pick up the bits.

Maybe I'm a bit too optimistic, but I hope that maybe tomorrow we'll be able to finish all the builds, and we'll just have to wait for the autobuilders to do their work. At first everything went smooth, but they are already stuck in dep-wait failures due to libgnome, nautilus and others not having all the build-deps in place. Hopefully all the buildd's will retry soon...

A final note, remember not to happily dist-upgrade your unstable box today. Before that, check that nautilus isn't in your "to be removed list...

Ouch, this time it wasn't so easy. I went to the dentist again to continue with the wisdom teeth extractions.

First news was that he now wanted to extract the two in the right side, while I thought he'd go one by one. One would probably be quite straight forward, like the other day. The other one would probably need surgery.

As soon as the anesthesia was applied, the doctor extracted the upper tooth very easily, in a matter of seconds. Cool, it was a lot less aggressive than the other day.

The one in the bottom would be more difficult though, as it wasn't completely out. Instead of cutting through the flesh, he tried to cut half of the tooth so he'd be able to manually extract the rest. Urgh. He started applying evil tools on top of the poor thing, which made quite terrible noises, and worse, a smell of something burning that impressed me quite a bit. I guess I was way too nervous because they had to repeatedly ask me to calm down. As soon as I relaxed a bit, some weird noise or a bit of pain would put my body on defense mode again... and this went on for more than 30 minutes I think. In the end it was out "cleanly", and I didn't need any sewing, which is good.

When it was over, I noticed I was still trembling a bit and my legs were humid with sweat (and believe me, it's quite strange to have me sweating). After so much anesthesia, I could barely speak, because I couldn't articulate any sound which involves using your tongue...

Eating is a nice exercise now. I can't open my mouth much, and I must avoid using the right side of my jaw entirely. During the dinner, some bits of food would get "lost" inside my mouth, due to my total insensibility, and I had to look for them a few times using my finger. It was a great spectacle. As I write this, the anesthesia is completely over and I'm starting to feel deep pain in the two craters. The one in the bottom part is quite huge and still bleeds. The taste of blood gets a bit boring after so many hours...

One more time, and all this shit will be over.