I woke up to a breakfast of pancakes, cooked by the wonderful riseorbleed. I don't know about you but waking up to a hot breakfast is heavenly. After some morning ablutions, we hailed a cab and along the way we chatted with the cabbie as I looked out the window. Luxor cabbies are really nice, every time I travelled with them, they had some interesting story to tell and big, broad smiles. Smiling is the way to fatter tips! The weather in San Francisco is supposed to be rather variable. You know how that is, rain one minute and then sun the next. Well, the entire time I was there, it was bright and sunny and gorgeous. San Francisco really put on her best to charm me and she really succeeded. I didn't think that postcard-perfect pictures were possible, but I saw them everywhere. I was happy to walk around without needing a winter coat. And everyone else smirked at how happy I was. Daniel seemed to have recovered from the previous night's revelry. His talk on delta was excellent because he's a very good public speaker. Which is good, because he gave two presentations at CodeCon. When I talked to him on Thursday night, he seemed a little embarassed at how successful delta had become since it such a simple piece of software. But people like GCC use it to fare down their bug reports to the lines of code that reproduce its bugs. Rob and Tristan sidled up to my bar and asked me if I could get them some juice. For their laptops. They were working on their presentation all the way until the last minute, which involved a drunken Tristan stumbling around on stage. Except that Tristan doesn't drink, so his imitation was quite poor. They were presenting Djinni, which is their implementation of a fast simulated-annealing algorithm that takes also pressure into account. Sadly, they didn't explain their algorithm but their talk was entertaining nevertheless. At lunch, Spider snuck me out of the building and she showed me around San Francisco. She pointed out the Eagle Tavern with a glimmer in her eye. I think she's trying to corrupt me. Or she's trying to pimp me out. Then we went to Stompers where she tried on some boots. We were stymied, however, by the fact that a very annoying woman and her husband were trying on every single pair of boots in the entire store. iGlance is a real-world video-conferencing and screen-sharing application that has some fairly good usability built into it. It also has a fairly sensible privacy model and does some NAT punching to boot. Although it's only got a Win32 port for now, it's free software so someone will fix it up. Sometime in the afternoon, the beautiful and fair wealhtheow came behind the bar and surprised me with a hug and a sandwich! I was happy with the random hug from a new friend, but the sandwich was wonderful. I had not eaten lunch on Friday so I was very grateful. She put a huge smile on my face. I munched on the sandwich while listening to the OASIS talk. This technology is a locality-aware server-selection resolver, which is pretty good for distributing servers around the world and finding the fastest one. They implement this as a DNS redirector so that lookups are transparent to normal Internet software. This made me feel a little bad for holdenk, whose SelfDirectedProject happens to be on a very similar topic. It does, however, validate his approach. maradydd's presentation about Query by Example was up next. She did a very good introduction to data-mining, which segued into her hack on PostgreSQL where you can specify examples of things you want to search for. That is not how SQL normally works, where you specify constraints. With QBE, you just specify things similiar to what you want, and things similar to what you don't want. I was chatting with Rob Pascual when I noticed Len lining up in the Q&A; queue. Suddenly, my ears perked up as I sensed something important was going to happen. So I pulled out my camera, stepped on some toes, and got into position. When it was his turn at the microphone, he started saying random stuff that wasn't question at all! Something cheesy like how Meredith had changed his life and how he wanted to be with her. Then he got up on stage and asked The Question. It was beautiful. The audience broke out into cheers and applause, except for the heartless bastards beside me. But who cares about them? This was such an awesome occurance that an interview was taped. After teardown, some people were going to head to a franchise Mexican restaurant. Spider insisted that I get real food, so we walked down to the Mission. We made a detour to Good Vibrations which is a San Francisco landmark if ever there was one. I swear that she's trying to corrupt me. Anyway, as we were walking out, I must have mentioned that I was from Canada. Because Cohen, a Torontonian, came out of nowhere and introduced himself to us. Well, what was I to do but to invite my fellow countryman to dinner? So we set off to find a Taqueria that served nopales. This failed since Californians don't seem to like eating cactus. Instead, we went to Pancho Villa where I ate enchiladas and salsa verde and guacamole and churros and I was so very happy! We bid farewell to Cohen after exchanging contact information and tried to find everyone else. Unfortunately, we couldn't reach anyone by cellphone, so we sat around dejected for a while. Then we realised that we could go to Annie's Social Club where an event was happening. This turned out to be a good and a bad idea. It was a good idea because we found everyone at CodeCon. It was a bad idea because Annie's Social Club double-booked us with bad karaoke! And then were quite rude when we complained. And then kicked out John Gilmore. So I just want to say that Annie's Social Club hates geeks and freedom. So Len hailed a black stretch limousine, we all climbed into it, invited some reporters along and moved the entire party to the XYZ bar. The XYZ staff were very awesome, accommodating an extra hundred patrons and getting an extra bartender. And they were very nice about it. Unfortunately, the bar was full, so I had to lead a bunch of introverts in the fine art of taking over an establishment. We walked in, started talking, I encouraged people to sit down in empty spots on the couches. They did a very admirable job of taking over the back. I got the stunningly adorable akashayi a seat and a drink, before I taked to the people sitting with us. They were a couple who had moved from Florida to San Francisco, and a man from New York who was quite the boor. The man in the couple knew the New Yorker and seemed quite embarassed about him. After twenty minutes of conversation, they excused themselves and bid us good night. This gave us plenty of seating space to talk about geeky things until the manager kindly kicked us out. I made sure Akasha got safely to her hostel and then took a taxi back to the guest room. It was quite the challenge to direct the cab driver, since I had lost the directions back, but I managed!

During the last two months, the Debian Catalan localisation team has seen two of their major milestones completed, after some tough work. The group managed to release Sarge with a completely translated installer, but failed to provide a translation of the Install Manual, which was a regression for us, as it had been available in Woody. We didn't give up, and thanks to the stubborness of Miguel Gea and braindmg we slowly started translating the Sarge manual months after the release, so it could at least appear in the website. But progress was too slow, so we planned an online meeting for the beginnings of the year to try to complete the translation doing a whole lot of translations during a weekend. Soon after we learned that joeyh, fjp and the d-i team wanted to terminate Sarge d-i manual support in svn RSN, as they would start to make changes to the docs to reflect etch changes, but fortunately Frans agreed to give us a slightly extended deadline, if we agreed to finish the job by January 14th:

20:37 < CIA-7> debian-installer: fjp d-i * r33721
               /sarge/installer/doc/manual/ca/: Add Catalan translation as
               translators would like to finish it for Sarge and have promised
               to do so 14 Januari 2006 at the latest

I really didn't think we would be able to make 100% in just 10 days, but the team, assisted by a large number of minionsnew contributors that volunteered on #debian-catalan, managed to do it I believe a few days before the deadline. Guillem announced it on our list. Releasing etch with an up to date Catalan manual will be a lot easier, if we try to maintain it up to date as etch development carries on. The second milestone was yesterday's completion of the Catalan translation for all five levels of the etch debian-installer, after a year of no translation activity on that front, and months of braindmg bugging me to do my chunk of work. Although it's now showing a shiny 100%, there's a lot of review work to do, and polishing it should be our focus for the next release. I seem to have recovered some of my interest in translations. That's good, because GNOME 2.14 is around the corner, and it wouldn't be fair for Josep if he ends up doing all the job. :) Also, I've been quietly working with a group of people at Softcatal in a project that will probably have lots of press in our community. It's not public yet though, but I hope that it'll be announced in the next few weeks.

After blathijs and I today talked a little bit about blosxom plugins on the #blosxom IRC channel, I installed the listplugins plugin. Since I’m a perfectionist in some things, I had to configure it to link every plugin I use to it’s web page or source. While going through my plugin list, I noticed that there were three additional plugins I wrote myself and of which I thought I should share:

• acronyms works similar to and is losely based on Fletcher Penney’s autolinks but instead of setting links it marks configurable keywords as abbreviation or acronym and show their expansion when hovering over the keyword (all using standard XHTML).
• xml_ping_generic is based on xml_ping_weblogs and can ping an arbitrary number of URLs to be pinged with the weblog.com’s XML RPC ping API. By default it pings weblogs.com and technorati.com.
• date_rfc822 is nothing else than the 822-date command (which returns a date in RFC 822 conform format and is written in Perl, too) wrapped into a blosxom plugin. Work similar to date_fullname. I use it for including <pubDate> tags in the RDF.

All plugins are published under the same open source license, they initially came with. In other news… I started using XFN, the XHTML Friends Network, at least the blogroll, and created accounts at Technorati and at yigg.de, a German Digg.com clone formerly respective yet still known as digg.de

Now playing: Battle Without Honor or Humanity — Hotei Tomayasu (from the Kill Bill Soundtrack)

I woke up on Friday morning and wondered what breakfast would be. Then I remembered that I was staying at the W and decided to forage for food elsewhere. I took a taxicab with spider88, enochsmiles, maradydd. It was a strange hybrid SUV cab that was painted all yellow, I would later learn that Yellow Cab drivers were scary and that I should avoid them. We got to StudioZ and discovered that we didn't know how to get inside. We figured that out eventually and took stock of the situation. I started moving tables about and setting up chairs, because I abhor standing about with my hands in my pockets. Then I helped hang banners from the rafters. After I did that, I tried to see if I could do anything else. Since CodeCon is run on the cheap so that anybody can attend, drinks like coffee are not included in the conference fee. And since the conference was being held in a night club, there was a convenient bar that would be a good place to setup a concessions stand. So I put things in order and stood behind the counter. I got to see people walk in the door and after registering with wealhtheow or Spider, they'd stand in bewilderment in front of my bar. It was a lot of fun to watch all sorts of people come through. And then I'd smile at them and offer to get them something to drink. We started Lance's talk a little late in order to let everyone settle in their seats. I met him the night before where he talked about his Apache module that has heuristics for detecting when a phisher is trying to clone your website. Or when your website has been cloned and the phisher is passing login requests through to you. This kind of thing is totally what financial institutions have to start doing. We got coffee from this place called Cafe Organica, which seems to have some mixed online reviews. Or rather, Jonathan did, in a big messenger bag filled with carafes. Since Jonathan doesn't actually drink coffee, it was left to me to decide whether it was any good. It seemed to be drip coffee that was thick, dark and strong. Very decent for something that was brewed hours ago. The customers didn't seem to mind, as we ran out of coffee every day. The next presentation was SiteAdvisor which takes an interesting tack on safely surfing the web. Instead of building security features into your browser, there's a central server that stores information about sites that you're visiting. But what was really interesting was how they got this information. They had an automated tool that went in pretending to be a real person, signed up for some e-mails, downloaded some programs, and examined what happened afterward. I missed the majority of the VidTorrent presentation, because I was busy chatting with people and offering them refreshments. Also, their demo didn't work because the wireless setup at the conference kept on going down. People were abusing the free Wireless network that Cliff was trying to keep up and running. Which is sad, because Cliff had basically volunteered to set that up on his own time. Tom came up after that to present Localhost, which is a unimaginatively named piece of software, but is interesting because it lets you browse a world-wide filesystem of torrents. What's interesting about his filesystem is that it uses a distributed hash table, so that even if there are files with the same name, you'll be able to distinguish the distinct copies. Which is useful if you want to tell whether you're downloading a legitimate and popular file, or some malicious file that someone's injected. The last talk of the day was Truman, which not only demonstrated an interesting piece of technology, but also quite educational. The tool itself is a specialised Linux distribution that can capture what happens inside itself and simulate a public connexion to the Internet. This is advantages, Joe tells us, because modern pieces of malware are smart enough to disable themselves when they detect sandboxed environments. After people filed out, we packed away everything into storage as StudioZ is a nightclub, so it was setting up for its nightclub crowd. Everyone else went next door to Loft11 where a party was being held, sponsored by bramcohen's awesome company I went inside and bumped into Graydon and we tried to go about finding fellow Canadians. This was not so successful on the crowded downstairs floor, so we went upstairs to chat and catch up. I met some people who were at the speaker's reception and successfully remembered their names. Then I met a guy who was at the night club just waiting for his buddy, and I explained CodeCon to him. He thought it was a great idea. Len Sassaman and Bram got up to give a little speech, which I tried to take photographs of. But failed. Instead, I noticed that they were on this awesome glass ceiling which made me want to dance. I didn't do that though, because my feet hurt from standing all day. I poked my head into a room upstairs that was all white, which was very cool. Some people were playing a game of Werewolf which finished with the villagers winning. I hung out with some people like Eric, Pete, and Kelsey who told the story of her pink Barbie backpack. I also met Chloe and her husband, whose name I have forgotten. But Chloe is awesome because she is now independently wealthy so that she can pursue her dream of studying the nitrogen cycle in soil using post-doc maths. Impressive! I also met akashayi who is not a professional computer programmer at all. She teaches martial arts by day, and slings code by night. She drove all the way from Los Angeles to attend CodeCon and was having the time of her life. I wistfully remember when I was that excited about computing, so just hanging out with her was awesome. She promised me a Kempo lesson if I ever made my way to L.A., which is an offer I intend to collect. I found Spider downstairs, sitting on a couch with Daniel. The night was getting long and we were totally exhausted, so we took a taxicab to the W. There, we went up to Enoch's room party where he treated us to $17 cheeseburgers. Now, I didn't think that cheeseburgers that expensive were worth it, but boy oh boy were those good. Spider and I took a cab to kineticphoenix and riseorbleed's place sometime past midnight and we dragged our luggage into their guest room. Which was really more of an awesome guest apartment. Hurray for knowing people in foreign cities! I don't remember much after that. I think I turned on a space heater before passing out.

I have a ThinkPad T40p, with an ATI Radeon inside. It has got a nice shiny TV-out plug on the side, and I have a short cable that ends in a composite video plug. My TV has a composite video socket on the front. When I press Fn-F7 (switch display), nothing appears on the TV. I'm expecting lots of pain to get this working: so far I am running fglrx and have the following line in xorg.conf:

Section "Device"
        Identifier  "ATI Graphics Adapter 0"
        Driver      "fglrx"
        Option "VideoOverlay" "on"
        Option "OpenGLOverlay" "off"
        Option "DesktopSetup" "clone"
EndSection

Anyone know the magic to make this actually work? NP: Dreaming Wide Awake, Lizz Wright

Musings on autoconf:
GNU autoconf (used the documented way) or "automake" is trying to do things in a way that did look apropriate in the 1970s.

But this is the best bit:
If you like best portability, you need to use my version of autoconf inside the schily makefile system and you need to use my smake instead of GNU make.
So my conclusion is that you did not understand portability.

Maybe it's just me, but I can't take J rg Schilling seriously when he smacks things down as being outdated and from the 1970s. Irony police and all that.

I bought a Belkin F6H500ukUNV from eBuyer (quickfind code 77131). At under £25 including P&P it seemed like a good deal and I've been meaning to get round to having a UPS to allow clean shutdown of my local machine during power problems. Belkin ship Linux and FreeBSD programs to talk to the UPS (though only over the serial interface - it has USB too). It doesn't like xterm it seems; prints Error opening terminal: xterm.. Seems fine with a terminal type of linux. It also puts all its files (config, help, binaries) in /etc/belkin. And it's binary only so I can't fix any of this brokenness. However, the lovely Network UPS Tools exists. Except Belkin have a number of UPS types, all incompatible with each other. After figuring out that I really, really needed to use the Belkin provided cable and not just a generic serial extender cable (grrrr. Why? If it's got a 9 pin D labelled RS232 why not make it a sensible port?) I got the Belkin software working and straced it to see what it did. A bit of looking at the nut source found several drivers that appeared to use the commands Belkin were sending (Q1, I & F). I tried powermust first, with the following /etc/nut/ups.conf:

[belkin]
	driver = powermust
	port = /dev/ttyM2

(I have a 4 port serial card, so /dev/ttyM2 is just a normal serial port, really.) And tada. It works. I can now get output like:

[noodles@pot ~]$ upsc belkin@localhost
battery.charge: 100.0
battery.voltage: 13.8
battery.voltage.nominal: 12.0
driver.name: powermust
driver.parameter.port: /dev/ttyM2
driver.version: 2.0.3
driver.version.internal: 1.1
input.voltage: 227.0
input.voltage.fault: 227.0
input.voltage.maximum: 234.0
input.voltage.minimum: 221.0
output.frequency: 50.0
output.voltage: 227.0
output.voltage.target.battery: 230.0
ups.delay.shutdown: 2
ups.delay.start: 3
ups.load: 0.0
ups.mfr: Mustek
ups.model: PowerMust
ups.serial: unknown
ups.status: OL

I don't know about doing shutdown yet, but ups.status correctly goes to OB (on battery) if I pull the power, which is probably the most important thing. :) At some point I'd like to move over to the USB interface; I have better things to do with my serial ports. Belkin don't appear to have done the sane USB HID power thing; instead it's some odd hid device that doesn't even have a Belkin manufacturer ID:

Bus 003 Device 015: ID 0001:0000 Fry's Electronics 
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               1.00
  bDeviceClass            0 (Defined at Interface level)
  bDeviceSubClass         0 
  bDeviceProtocol         0 
  bMaxPacketSize0         8
  idVendor           0x0001 Fry's Electronics
  idProduct          0x0000 
  bcdDevice            1.00
  iManufacturer           1 MEC
  iProduct                2 MEC0002
  iSerial                 0 
  bNumConfigurations      1
  Configuration Descriptor:
    bLength                 9
    bDescriptorType         2
    wTotalLength           34
    bNumInterfaces          1
    bConfigurationValue     1
    iConfiguration          0 
    bmAttributes         0x80
      (Bus Powered)
    MaxPower              100mA
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        0
      bAlternateSetting       0
      bNumEndpoints           1
      bInterfaceClass         3 Human Interface Devices
      bInterfaceSubClass      0 No Subclass
      bInterfaceProtocol      0 None
      iInterface              0 
        HID Device Descriptor:
          bLength                 9
          bDescriptorType        33
          bcdHID               1.00
          bCountryCode            0 Not supported
          bNumDescriptors         1
          bDescriptorType        34 Report
          wDescriptorLength     624
         Report Descriptors: 
           ** UNAVAILABLE **
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x81  EP 1 IN
        bmAttributes            3
          Transfer Type            Interrupt
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0008  1x 8 bytes
        bInterval              10
Device Status:     0x0000
  (Bus Powered)

The NUT energizerups driver appears to talk USB and use the Q1/F commands, so I think I'll try that later. Or it's back to usbsnoop and a Windows box I guess.

I logged into Last.fm this morning to look at my recommendations and among the more obscurer suggestions was Marmaduke Duke, who apparently are similar to Biffy Clyro, Hell Is For Heroes, Reuben and Million Dead, who are some of my favourite bands. So I decided to log into Amazon to look to see how much it was, and guess what was the music recommendation on the front page: I think someone is trying to tell me something. I'm now looking for some samples to listen to.

Currently four Japanese will attend Debconf6; Junichi Uekawa (you know him well), Nobuhiro Iwamatsu (SuperH guy), Yukiharu Yabuki (was working at OpenSource SIer company) and me. Today I reserved an air ticket of Nobuhiro and me. After Debconf, we'd like to visit Vancouver also.

5/12 17:15 NRT-16:40 YTO (staying at YTO[Toronto])
5/13 09:00 YTO-12:45 MEX
5/22 14:10 MEX-19:45 YTO (staying at YTO)
5/23 08:00 YTO-09:57 YVR (staying at YVR[Vancouver])
5/24 13:10 YVR-5/25 15:25 NRT

I’ve prepared amaroK 1.4-beta1 packages and uploaded them to experimental (a copy is available here). I recommend you to have a look at the changelog ; that way you’re more likely to discover the stuff that’s interesting to you. As an example: “Optionally update playcount for items played on iPod and submit them to last.fm and synchronize ratings between amaroK and iPod”. Note that you’ll need ruby installed to download lyrics now (feel free to reimplement the script in Python if you wish!), and that this version will rebuild your collection. If you can report bugs directly to upstream’s bugzilla, that’d be great (the packages are unstripped, btw). NP: Bergman, Enigmas in your hand

Item one: debbugs has moved from CVS to bzr. Item two: it’s well past time I came out in support of Mary Gardiner’s brief guide on pronouncing “bzr”. Item three: don’t you find the Sugababes’ latest pop ditty really quite catchy? Put them all together, and what do you get? Read the rest ...

As I told before, I’ve been working for a few days on a way to display japanese characters on a web page when you don’t have japanese fonts locally. That involved taking japanese characters from a font file, and turning it into a png image. So I just wrote a small and simple program using libpng and freetype. Which works quite well. But that’s not the point. During this “development”, I lost 2 or 3 days trying to understand what was wrong with my code that made it work properly from a shell command and fail halfway when wrapped in a php script. I still don’t know what’s wrong with php, but it works well with a perl cgi wrapper, and it still fails if I raise the php memory limit. Interestingly, other libpng tools (such as pngtopnm) fail in the same way. It wouldn’t surprise me if it was libpng’s fault. The first thing i did to understand why it’d fail, was to run the program under valgrind. And what I saw was not encouraging. A lot of “Conditional jump or move depends on uninitialised value(s)”. I tried different combinations of code, using png_write_image or png_write_row, with large or small buffers… It turns out libpng can’t write a png file with a width smaller than 148 pixels without those “Conditional jump or move depends on uninitialised value(s)”, which happen, as the stack trace shows, in the zlib, called by png_write_row, but i doubt it to be a zlib issue. You would think it would be a buffer size issue, but if you use a larger buffer for each line you call png_write_row for, you still get the errors. If anyone has a clue, please leave a comment. Update: Thanks Mark for your comment, though I don’t get how it can be safe to have such accesses to uninitialized memory. Deuce, the php issue can’t be a timeout issue, the program runs in fractions of a second.

I have again worked on the automatic package build, test and publishing process for clamav-data, which runs twice an hour on a volatile.debian.net host. Details of the process are described in VUA 2-2 on debian-volatile-announce The “backup” copy on http://people.debian.org/~zugschlus/clamav-data now has accurate mtimes in the directory, and actually the last three published packages are available from here. The rsync process was previously broken in a way that only the current package and two older ones from middle of July (from before the automatisms were in place) were available there. The publish process now barfs if the specially hacked katie which is used to move the new clamav-data package into volatile.debian.net without a maintainer signature leaves the publication directory (which is to be considered part of queue/unchecked for katie’s environment) in a non-empty stage. This will hopefully catch more katie errors than the parsing of katie’s output, which is of course still being done. After a new package has been built and sucessfully published, the package build directory is cleaned up. All files belonging to changes files older than 60 days are deleted to keep the disk from filling up badly. But that’s an internal matter only which you shouldn’t be noticing at all. Additionally, after the last changes, I have hiked up the verbosity level of the package build process for a while so that I can notice when the new process dies or encounters new challenges. That shouldn’t be noticeable outside my personal mailbox though. One issue that is still left is that the automatic clean up of the Pool doesn’t seem to work for source packages. Binary packages that are not referenced by any Packages file are supposed to be removed after 15 days, but the source packages seem to stay around. I cannot do anything about that, and am targetting aba on this issue.

* helix wonders if erich ever gets tired of complaining Customer: I wish to complain about this developer what I purchased not half an hour ago from this very channel.
Vendor: Oh yes, the, uh, the German Blue...What's,uh...What's wrong with it?
Customer: I'll tell you what's wrong with it, my lad. 'E's complaining, that's what's wrong with it!
Vendor: No, no, 'e's uh,...he's discussing
Customer: Look, matey, I know a slasher when I see one, and I'm looking at one right now.
Vendor: No no he's not complaining, he's, he's discussing'! Remarkable dev, the German Blue, idn'it, ay? Beautiful coat!
Customer: The coat don't enter into it. It's bitching.
Vendor: Nononono, no, no! 'E's discussing
Customer: All right then, if he's discussin', I'll debate with him! (shouting at the computer) 'Ello, Mister Perly Python! I've got a lovely fresh babel fish for you if you show...
(owner hits enter) Vendor: There, he replied!
Customer: No, he didn't, that was you spoofing an email!
Vendor: I never!!
Customer: Yes, you did!
Vendor: I never, never did anything...
[...]
Customer: 'E's not chattin'! 'E's vitriolic! This dev is no conversation! He has ceased to design! 'E's obsolete and gone to meet 'is makefile! 'E's a nagger! Bereft of constructivity, 'e breaks in pieces! If you hadn't set an autoresponder 'e'd not be replying to mails! 'Is patches are now 'istory! 'E's off the keyring! 'E's monologing. 'E's bitching, THIS IS AN EX-FLOSSDEV!! You get the idea. ;-) Apologies to one of the greatest comedians. No, Erinn, I'm not just complaining all the time. I actually worked quite a lot on OSS these days. But in some projects it feels much like you are the only one working there, probably a bit too much... just don't take my harsh critique too seriously, I'm actually a nice guy, just sometimes too involved with things. (But no, I'm not retiring or so. But facing exams soon.) On an unrelated note, I today uploaded the selinux-basics package to unstable, it contains some small niceties in setting up a SELinux system on Debian. But it's still far from completed, easy or documented.

It seems that I have become quite a gadget-whore lately. I'm spending all my money buying one gadget after the other, no end in sight... Anyways, I finally got one of those fabulous Motorola A780, a Linux-based smartphone. Getting one turned out to be way more complicated (and expensive!) than what I thought it would be. In the beginning, all sounded quite good: my cell phone contract with o2 is two years old soon, so I can get a new (cheap) cell phone. Stupidly enough, o2 doesn't offer the A780 in their shops and there seems to be no way to order one either (they do offer other Motorola phones, though). After asking the same questions in different o2 shops multiple times (and almost giving up), I accidentally saw the A780 in the local Saturn (a German electronics store). And indeed, they sell the phone, and they can even prolong my contract with o2 (there's a dedicated o2 employee working in the Saturn store), so that I can profit in the form of a cheaper phone. Or at least that's the theory... In practive, however, I have a student-contract (saves me some bucks) which has the stupid "feature" that it can only be prolonged in o2 shops. Guess what, the Saturn guys cannot give me the A780 as they can't prolong my contract, and the o2 shop simply doesn't have the A780 at all. Argh! After grumbling, asking around, googling, and even more grumbling, I finally decided to do the following: I got a new "dummy" o2 contract in the Saturn (yes, I'll have to pay that for 2 years) which enables me to get the A780 and to get it cheaper. I'll keep using my old contract and my old SIM card for simplicity and leave the new one untouched. If you take into account the money I'll spend on the new contract it doesn't save me too much money, but at least it's distributed across two years... I'll terminate my (new) o2 contract tomorrow, to make sure I don't forget about it (I don't want to have it any longer than the 2 years I'm forced to live with)... Stupid, stupid world we live in. Nobody should be required to perform such "hacks" in order to get the phone he/she wants... Enough ranting now, here's some juicy details about the phone:

• ARM CPU (at 400 MHz, I think)
• 48 MB RAM, 48 MB on-chip flash memory, 256 MB Transflash card
• 240x320 touchscreen
• 1280x1024 built-in digital camera (can record videos, too!)
• USB (device, host, OTG)
• Bluetooth
• GPS (+ it comes with CDs with maps of Europe)
• Plays MP3s and videos
• Can send SMS, MMS, emails, and has an IM client
• and lots more great stuff...

For details on the hardware see this wiki page. The only thing which I'm missing is WLAN, but once USB host support works (the hardware does support it), you can easily use a WLAN USB dongle... I'm pretty sure I'll be having lots of fun with this thing, and I'll quite probably be contributing to the OpenEZX project, which was started by Harald Welte (of gpl-violations.org fame) and tries to create the first 100% Free Software GSM-phone using the Motorola A780 and similar phones. Judging from these blog posts by Harald, running your own 2.6 kernel on the phone is not too unrealistic anymore, and telnetting into the phone (via USBnet) seems to work fine already... Expect more spammingblog posts about the A780 in future...

On Heise’s security site HeiSec, Microsoft is advertising (in German) with a Flash animation of a rolling, black horse approaching the reader. Then suddenly a red gate closes and a text apprears:

Trojans must stay out.

If we translate this back to ancient greek history, it would say:

Microsofties must stay out

since Trojans were the inhabitants of Troy (German: Troja) and in the horse were the Greek aggressors. So I strongly agree. ;-) I really hate it, if people just reverse the meaning of something by abbreviating it (here by turning the adjective into a noun). And then not noticing it. The term Trojan Horse in computing is just one (unfortunately) often seen example… But no wonder that Microsoft doesn’t care about such things. They care about so less (e.g. stable software, secure operating systems, users, administrators, trust, etc.) except keeping their monopoly, making money and making even more money.

As of a few hours ago, I'm now running our xorg 7.0 packages on my main system. I'm ironing out a few niggling issues with the wonderful help of Zephaniah E. Hull before uploading the remainder of what I've got to experimental. There's still a few minor things left to package, but there's now a solid base that you can actually run and upgrade pretty cleanly from testing and unstable. It's still not for the faint of heart, but things are looking good for a quick move to unstable from experimental. Daniel Stone's previous work is a great base since he's worked out many of the issues beforehand.

As an aside, I know people are hungry to play with Xgl and compiz, and rest assured I'll get them to you ASAP. I haven't even tried to compile them myself yet, let alone run them, but I'll be working on it as soon as the 7.0 packages are in satisfactory shape.

…you are saying Iran or Somalia are less important countries? Does the importance matter here? Are you minimizing them because of a strong reason or because you can’t drink on them since they are not important countries? Why not visit them? If I had enough funds, I’d do it. Hey, DPL, do you approve travel expenses for me to visit Iran or Somalia?