The 2010 Debian Project Leader election is over! (but the page is not yet updated Try here Thanks to Steve McIntyre for two long years of good work And, of course, congratulations to Stefano Zacchirioli! I, for one, welcome our new Italian overlord.

As I've mentioned here before, one of the differences between "Europe" and the US is just how cheap it is to start a company in the US. Before we go any further, I'll take a moment to add the standard "yes, I know that Europe is not one country" disclaimer, and specify that I'm mostly talking about continental Europe. Starting a company in the UK or Ireland isn't nearly as bad. In Oregon, I spent $55 to create DedaSys LLC. If I'd created it with one or more partners, I would have spent something on a lawyer in order to create a solid agreement, but that is of course a voluntary expenditure that we would pay for because it provided us with some value. In Italy, it costs thousands of Euros just to get started with a company; before you've made any money at all. And, while there are gradual improvements, it's still a bureaucratic process that pretty much must involve at least an accountant and a notary public. And you have to involve the very arbitrary number of 10,000 euros of capital in the company, supposedly there as a guarantee for the people you're doing business with. But 10,000 is not nearly enough to cover some kinds of problems you might cause, and way more than, say, a web startup with a couple of guys and their laptops need. My friend Salvatore says it's possible to sort of "get around" sinking the full 10K into your company, but in any case, the principal of "caveat emptor" is a more sensible one. At most, make a transparency requirement so that people dealing with companies can tell exactly how much reserves they have. During a recent bout of cold/flu, compliments of our daughter's nursery school, when I had some extra time on my hands, I decided to do something about this, however much it may be pissing into the wind. I set up a site: http://www.srlfacile.org (warning: Italian only) As well as a Google Group, and petition for people to sign, in an attempt to make a little bit of noise about the problem, here in Italy. While it's likely that the actual bureaucratic mechanisms are more smoothly oiled in other European countries, I have my doubts as to whether the actual amount of paperwork can compete with the very minimal page or two required by many US states. And in any case, the costs are still quite high, and while we all have different levels on the idea of the role of government, and ideal levels of taxation, I think we can agree that it's sensible to levy taxes only after a company has begun to make money! So - how about it? Anyone want to create sister initiatives in other countries in Europe where things aren't as simple and easy as they should be? Anyone care to tell of how this problem has been fixed elsewhere? I've heard tell that in Germany, there is now a simpler/cheaper way to create limited liability companies.

The DPL vote is over. I did not win, apparently. Let me start off by thanking the other candidates for running, and congratulating Stefano for winning. My own results aren't too bad; I beat 'none of the above' by a comfortable margin of almost 10 to 1, and ended up defeating Charles Plessy by a slightly smaller margin. My margin to NOTA was far better than during the previous DPL election that I participated in, though I did end up defeating more than half of the other candidates back then, which I didn't this time around. All in all, not enough to win, but enough to conclude that I might have a chance some other time. Margarita and Stefano were just better candidates this time around. My own vote was:

[ 2 ] Stefano Zacchiroli
[ 1 ] Wouter Verhelst
[ 4 ] Charles Plessy
[ 2 ] Margarita Manterola
[ 3 ] NOTA

If you run, you have to be serious about running, and support yourself. So I did. After that, I wasn't sure that Stefano would do a better job than Margarita (or vice versa), so I put them at the same spot. Finally, I became convinced during campaigning that Charles has a lot to learn still about what lives in the Debian community, and I am of the opinion that feeling the community is a vital requirement for a DPL; so that ruled him out. So why did I not win? Obviously because there were better candidates in the election, but the real question is: why is that the case? What did I do, or what did other candidates do, that meant they fared better than I? One possibility was given on IRC:

< svuorela> ol: and "I'm running because no other runs" is also
not a good way to get votes ...

Which is fair enough; but I did run with the intention of winning, and I did try to make that clear. I'm hoping that worked out okay, which would imply that there could be other reasons. I'd like to know about those. So if you voted, and did not place me on the top spot, I'd very much appreciate learning about your motivations for doing so especially so if you were one of the 39 people who would rather have redone the vote than see me win. Comments on this blog post are welcome, or by private e-mail if you'd rather not see them made public.

I tend not to be a huge Apple fanboy. They make nice stuff, but it's just too closed for my liking. I like to tinker. I had a PowerBook for a while, but I gave it to Sarah in favour of a Linux laptop. Sarah's been a happy Mac user for a number of years, and had an iPhone (until I gave her a Nexus One). For a "normal" user like her, a Mac is fine, especially if you want to embrace Apple's entire ecosystem. Anyway, the iPad. When it was announced, I sat up and took notice. Why? This seemed like something I might actually use as a casual computing device. I mean, I'm almost in the target market for Chrome OS these days. I spend most of my time in a web browser, and the rest of my time in a terminal window SSHing to another computer. I could leave this thing lying around on the coffee table in the living room, and instead of digging my phone out of my pocket to look something up, I could pick this up instead. It is also appealing because I found Microsoft's Surface to be pretty cool. The iPad is like a more affordable, portable, version of that. It also appeals to me as a computing platform for my parents. Their computing needs have simplified over the years, but they're still running Windows, largely because I've never had the time to try and foist Linux on them. Since I moved to the US, my visits back home have been too brief to do a proper migration. I think an iPad that supported user switching would be perfect. Mum and Dad could share it, and read their email and do their web browsing from anywhere in the house. Since the first generation iPad doesn't do user profiles and lacks a camera, I'll wait impatiently for the second generation one. I heard a rumour today that it would have a camera, and do user profile switching based on the face of whoever was in front of it. That would be pretty cool. I'd also be very interested in an Android tablet. I love Android's speech to text input support, and I could really see an Android tablet stuck to the wall in the kitchen, instead of a whiteboard on the fridge and a paper calendar. The WePad also sounds intriguing. So I'm not so bothered by the iPad's closed nature. I think for the set of users who have basic computing needs, and don't care about openness, it's very cool.

This month is the ten year anniversary of when I started blogging with this post. After a strong start in 2000 thanks to advogato.org, I slowed down, and 2003 saw no blog posts at all. Since then I have kept going fairly steadily, to no set schedule, writing when I feel the need. In all, 984 posts over ten years, or almost two posts per week. 205000 words. 650 pages if it were all bound into a book. (Sounds like a lot maybe, but the software I've grown to run the blog is 3 times longer.)

It takes a lot to prepare for a big trip if you want to really enjoy it. This time, we re going to Japan, and we bought some stuff to stay connected there. First, there s the new camera: that s a Sony 230. We haven t made a photo trip to see all its capabilities yet, but it looks like an excellent toy so far. You ll see probably more photos on this blog in the future. And there s the new laptop: a Packard-Bell Dot-M/A. Theoretically it s called a netbook, but in practice it has everything a real laptop has. The reason I chose this model is that it features Radeon (X1270) graphics and a 64-bit processor, all in a 11,6" laptop which is one of the cheapest of all. Lots of power in one kilogram for a low price, although the drawback is more cells in the battery. Getting the Dot-M/A to work under Debian I first tried to install lenny on it, and while it worked nicely there are several problems with hardware support.

1. The CPU runs extremely slow; you would think it is an Atom. It takes no less than one minute to boot a minimal installation. This is a very strange issue.
2. Wi-Fi doesn t work, even after installing the firmware.
3. 2D works out of the box, but 3D doesn t: the kernel doesn t recognize the PCI ID.
4. Frequency scaling doesn t work, it always runs at full speed which eats battery at an impressive pace.

Upgrading to squeeze solved the first three issues in a blink. The CPU is now as fast as you can expect from an Athlon 64 @1,2 GHz, there s wifi and 3D. OTOH I was hit by a GStreamer bug when the useless snd_pcsp module was loaded why isn t this thing blacklisted by default? ACPI nightmare CPU frequency scaling is another story. I discovered that the BIOS for such Athlon L110 computers does not expose P-states in the ACPI DSDT table. Which means Linux cannot tell at which frequencies it is supposed to work. However, thanks to the awesome work from a guy named Krists Krilovs and the awesome tutorial from the Gentoo wiki I was able to:

• rip the DSDT table from the CPU that s the easy part;
• disassemble it using Intel tools still easy;
• debug the table, which is written in an arcane language no one understands, and using references to a tutorial that was probably excellent until it disappeared from the intarwebs;
• integrate the P-states for the Athlon L110, basing again on the work of Krists Krilovs who wrote the corresponding ASL code;
• build a custom kernel using it that s the easy, but long part, that s done on a faster computer.

After a reboot, I immediately noticed the fan slowing down. Under GNOME, the CPU was no less than 10 C cooler and the CPU frequency applet started to work. <hrule> The Debian kernel maintainers deliberately chose not to provide support for loading a DSDT table from the initrd. There are very good reasons for this, and anyway it shouldn t be necessary to hack something as awful as that to have power saving support. The question remains: how do we deal with this madness? There needs to be some kind of support out-of-the-box for the Athlon L110, which is otherwise a very nice beast. Could the powernow-k8 module set hard-coded defaults when it detects this CPU model? It would be better than the current situation. Other pieces of the toy Otherwise this laptop is very good hardware. Among other things, I enjoyed:

• the keyboard which is very comfortable for this size;
• the awesome screen, quite small, yet 1366 768 and very bright and sharp;
• the quite fast SATA disk (~60 MB/s);
• the card reader, which makes a useful replacement for a CD drive;
• the webcam that worked out of the box;
• power management that works just as expected too.

There is one minor annoyance: the integrated RealTek Ethernet card is only 100 Mbits/s. With all this performance otherwise, you would have expected Gigabit, but well, not everyone has GigE at home yet.

So you ve seen people talking about WebKit2, perhaps have seen someone claiming it drops support for Linux , and you ve been wondering what the hell that means for WebKitGTK+. Well, welcome to the preemptive Q&A section with WebKitGTK+ maintainers =D. Let s first explore some history so we can better understand what exactly is going on. What exactly is WebKit2? Currently, when we say WebKit we really mean one of the ports that are built on top of WebCore using the WebKit layer. WebCore is the part that does all of the hard Web-related work, WebKit an API layer that exposes WebCore functionality in a coherent way, so that the platform-specific ports can expose a public API layer for their applications to use, which is usually also called WebKit . This WebKit layer was designed by Apple to build the Mac, and Windows ports it maintains, and was later released as Free Software so that other ports, such as the GTK+, Qt, EFL ports could be built on top of it, instead of having to do all the heavy lifting from WebCore directly.

WebKit2 is nothing more than the second version of that interface, with a whole lot of changes on what you can expect from it, and on how it interacts with WebCore, and the platform-specific API and UI. First of all, the first WebKit was not API stable, and that interface was usually not made public by the various ports - they only exposed their platform-specific APIs. WebKit2 is being designed to provide a stable, cross-platform, C-based, non-blocking public API. This is huge. It will allow cross-platform code to be written without having to consider language, and port differences for basic functionality. The second big change is the API will be made fully non-blocking. Currently most things you do are asynchronous already, but some of them may be completed in a synchronous ways (like, loading a string into WebKit instead of an URI). This is important for responsiveness, and is also a very important need for what comes next: process splitting. WebKit2 will bring into WebKit proper the concept of splitting the UI process from the Web process, similar to what Chromium has. It also much more awesome than what Chrome has for a large number of reasons, including, but not limited to:

1. It s being contributed directly to the WebKit project, in a cross-platform way that lets ports such as WebKitGTK+ take advantage of it, instead of being shipped directly into Safari, like Google does with Chrome;
2. The process separation goes bellow the API layer, meaning that all complexity involved in managing the process separation is handled by the library, and hopefully none of it leaks to the application using it; that means that applications like Devhelp and Yelp will be able to take advantage of this without having to make their lives more complicated;

What WebKit2 is not? WebKit2 is NOT a rewrite of the whole WebKit stack. Webcore will continue mostly unchanged, and all ports currently building on top of it will keep working. It is also not a fork - the code lives in the same tree as the current version of WebKit, which will allow us to progressively move towards using this new, improved layer. WebKit2 is not Apple-only, and it is not dropping Linux support. Initial builds of the code that is being landed will likely show up building on Linux in the near future (specially because us porters are already eager to play with it). What happens to WebKitGTK+? In the near future, nothing special. We will continue working towards making it feature-complete, more stable, faster, and rocking on it as always. We will, though, start working out how we can best take advantage of WebKit2 in order to provide an even more awesome library for the G world. What this means is you can expect us to have a library that will provide a nice GTK+ widget, just like we have today, with a GObject-based API, like we have today, but that is built on top of this new WebKit2 infrastructure, taking advantage of the process-splitting, and the bigger focus on not blocking the UI thread. This should give us a platform that is more stable, and faster and more responsive than what we already have today. The API is bound to change, of course, but the WebKit2 version of WebKitGTK+ will be a separate, parallel-installable library, and we will keep supporting the WebKit1 version while we work on making the new one at least as good as the current one. This is long term we re talking here. We ll likely see WebKitGTK+ 1.4, and 1.6 come to life before we are satisfied enough with WebKitGTK+2. We hope this clears some of the doubts up, and lightens your hearts! The WebKitGTK+ maintainers.

The friend who lent me his Kottan ermittelt DVDs gave me the 1981 movie (cinema, not TV) Den T chtigen geh rt die Welt. I guess a warning is in order: if you liked the TV series, it's possible that you'll share my disappointment with this film. There are reports of a new movie to be made this year, and I also haven't seen the first 9 episodes of the TV series yet, so I'll probably report on Kottan again some time in the future. I was surprised but liked the film: There's a coherent plot and even a happy end. And it's till a film by David Lynch: Wild at Heart. Also by Lynch: the very early Eraserhead. Now this is Lynch (at least for me, since I started with Lost Highway and Mulholland Drive ), and I really like the hairstyle of Henry Spencer. Something completely different: Harold and Maude, or what happens when a bored high society youth falls in love with an 80 years old eccentric who shares his hobby of going to funerals of some random people.

snapshot.debian.org launched
with all Debian packages produced in the past 5 years
6.5 terabytes first Wikipedia database dump in 4.5 years
5.6 terabytes (32 gigabytes compressed) My reflex on seeing both of these was to think about putting them into git repositories. For snapshot.debian.org, injecting source packages into git repositories is easy with git-import-dsc (and it can also use pristine-tar to make the original tarballs accessible with a miminal overhead). I hope the snapshot.debian.org admins find time & space to do that, because being able to easily access git annotate data spanning 5 years for any package in the archive would be very useful. To produce a usable git repository from the Wikipedia dump would probably involve writing a custom git-fast-import that processed the huge xml dump, and chunked it up into individual files, and commits changing those files. Frankly, I do prefer wikis that store data in git in that format natively and don't have multi-month dump procedures. ;) How big would the git repos be? My SWAG is well under 1 terabyte for snapshot.debian.org, and between 30 and 300 gigabytes for the wikipedia data.

It s no surprise that information visualization and related fields have been active research topics in the past decade. The widespread availability of large, valuable data sets has created a need for improved tools and techniques for making sense of it. This is highly technical work, requiring expertise in design, mathematics, software and other fields, and there is a vibrant community of professionals and amateurs who experiment and critique the latest methods. In the geeky blogs that I follow, I see more and more amateur interest in visualization. Visualization is starting to take on political significance as well, as evidenced by the appointment of information design pioneer Edward Tufte to a US government advisory panel to help explain economic policy to the American public. A technology is confirmed to be truly and vitally relevant, though, when it enters the mainstream consciousness in the form of humor. In the case of visualization, geeks have been trading funny charts and graphs for many years, but this phenomenon seems to have finally crossed the chasm into popular Internet culture with the advent of the Venn diagram meme. Now, anyone with an elementary education and a drawing program can create their own chuckle-inducing infographics and enjoy their 15 minutes of Twitter fame. Likewise for the X as seen by Y two-dimensional grids, revealing how various (stereotyped) groups perceive each other. This is the future, becoming more evenly distributed by the minute.

Yesterdays NUUG presentation about Kerberos was inspiring, and reminded me about the need to start using Kerberos in Skolelinux. Setting up a Kerberos server seem to be straight forward, and if we get this in place a long time before the Squeeze version of Debian freezes, we have a chance to migrate Skolelinux away from NFSv3 for the home directories, and over to an architecture where the infrastructure do not have to trust IP addresses and machines, and instead can trust users and cryptographic keys instead. A challenge will be integration and administration. Is there a Kerberos implementation for Debian where one can control the administration access in Kerberos using LDAP groups? With it, the school administration will have to maintain access control using flat files on the main server, which give a huge potential for errors. A related question I would like to know is how well Kerberos and pam-ccreds (offline password check) work together. Anyone know? Next step will be to use Kerberos for access control in Lwat and Nagios. I have no idea how much work that will be to implement. We would also need to document how to integrate with Windows AD, as such shared network will require two Kerberos realms that need to cooperate to work properly. I believe a good start would be to start using Kerberos on the skolelinux.no machines, and this way get ourselves experience with configuration and integration. A natural starting point would be setting up ldap.skolelinux.no as the Kerberos server, and migrate the rest of the machines from PAM via LDAP to PAM via Kerberos one at the time. If you would like to contribute to get this working in Skolelinux, I recommend you to see the video recording from yesterdays NUUG presentation, and start using Kerberos at home. The video show show up in a few days.

My blog is written in English, even though I sometimes post content relevant locally. Anyway, I want this to be as widely known as possible. We are about to launch a website/community for my University's cyclists, BiciUNAMonos. Please note the site I'm linking is still very preliminary in several ways. But that's the least relevant for now. I am translating here a message written by fellow cyclist (and researcher in the Institute of Astronomy of my University) Sergio Mendoza.

Regarding the fatal cycling accident in Ciudad Universitaria Yesterday night I got home, and was quite surprised to open my mailbox and find a considerable amount of messages under the subject Urgent: Fallen cyclist! I immediately noticed something was very wrong, and was surprised they were contacting me. When I decided to read the first such message, I realized what happened: Dr. Jorge Villanueva was run over by the Puma [internal University service] bus number 14, route 2, which was being driven over the speed limit. This happened last Sunday, April 11. On Sundays, all of Ciudad Universitaria becomes a cycling and running circuit. I was immediately brought back to the moment that, when I was leaving my office in the Astronomy Institute and heading towards Cerro del Agua avenue, a lady driving over the speed limit on the circuit that comes from Metro Universidad threw her car at me. I saw her at a few centimeters distance from her windshield... To the day I don't know how I pulled my bike away, as I was going to hit the windshield with my shoulder. Everything happened in slow motion, I just heard several cars honking at this imprudent lady. I didn't have a scratch, but got excessively close to getting a strong hit, and with the adrenaline raging I gave a kind caress with my closed fist to her brand new car. I had to free my adrenaline somehow... My legs were trembling when I tried to pedal. I told her, lady, I am excessively visible in my bike, why did you do this? , she didn't answer, she just looked despectively at me. This kind of accidents happen every day in the University and its main entries. Precisely today, a University professor threw his car at me entering the University via Cerro del Agua. When I reached him and confronted him, he told me completely sure of his words that I was invading the lane with my bike, and that it was my fault. Anybody would have thought a car-driving university professor is capable of reading the Metropolitant Transit Regulations that clearly states on its 1st article: The priority for using the public space, for the different transport modes will be according to the following priorization: I. Pedestrians; II. Cyclists; III. Users and providers of massive, collective or individual public transport service; IV. users of particular automotor transport; and V. users and providers of freight transport. We are now organizing to put up the first ghost bike in Ciudad Universitaria. This happens few days before the formal inauguration of BiciUNAMonos, the universitary association of urban cyclists; we will do this remembering our university mate. I see every day more people in the University hitting their bike pedals to get to their faculties, centers and research institutes, but what is striking is the lack of University infrastructure. The few cycleways we have are built to use a bycicle system that is open only until 16:00. It is good, but it is in no way useful for those of us who use the bike every day and at every hour, no matter the weather. The main problem lies at the communication points, when cyclists arrive from outside Ciudad Universitaria and have to take the regular circuits to reach the cycleways. The Transit Regulation, in the 1st article I quoted above, defines in an excellent way the priorities that must exist to allow human movility. In this regard we have to mention that:

1. The pedestrians must be enormously protected with walkways and semaphores to allow them to safely walk. In Ciudad Unviersitaria there are many places that don't even have a walkway or semaphores, and drivers do not respect the pedestrian's preference. Also, now in the central area of Ciudad Universitaria we have the Pumabus system on a confined lane, providing an excellent service to the pedestrians to move within the campus.
2. The cyclist comes next in this priorization, but there is nothing obvious for him. There are a couple of cyclist ways, usually invaded by pedestrians, and explicitly made to use the Bicipuma system. Cyclists need their own lane for circulation in the Universitary circuit; pedestrians have walkways, Pumab s has its confined lanes, but the cyclists still lack it. The third lane that remains is precisely for the IV and V entries in the Metropolitan Transport Regulations. They can perfectly share the remaining lane in the campus.

The last thing I wanted was to formally announce BiciUNAMonos reminding a fallen cyclist and dedicating a ghost bycicle to his memory in Ciudad Universtiaria, but we don't have a choice. We have to make it very clear to the University community that the bicycle is the best transport system, and that the drivers' imprudence inside Ciudad Universitaria must be strongly punished. Sergio Mendoza
President
http://www.biciunamonos.org

Coincidentally, today I also had a freightening experience. Taking one of my usual routes to work, going up from the sporting area in the Western half of the campus (just before where the map lists the highest point in my ride), just by Investigaciones Biom dicas, I heard a car hitting the breaks behind me and passing me (on my right side) quite narrowly. I was, as always, driving towards the center of the lane, to be able to respond to unforseen events (as this one, precisely). The Universitary authorities had the great(?) idea of placing round, small bumps before the Pumabus stations to avoid drivers parking in front of them. I have seen a strong accident caused by them, and was close to suffering one myself some months ago. I know that just after Biom dicas there are some such bumps, so I never ride too much to the right I don't know what was on this driver's head. Ciudad Universitaria has a 40 Km/h speed limit. Yes, it might seem too low - Circuits are wide and well planned. But there is a lot of cyclist and pedestrian transit, at all times, and I must recognize such a limit is well in place. I often take a longer route to my Institute, bordering the University along Av. Delf n Madrigal. People tell me I'm crazy, as this is an avenue that's often taken at high speed - Usually, cars pass me by between 60 and 80 Km/h. However, I even feel safer going that way than going inside the University. Contrary to the popular belief, Universitary people do not have better driving habits than the bulk of the population. And it's often much easier to drive along a way where the traffic even if it goes at three times my speed runs only in one direction, unlike the swarming of people going in every possible direction inside the main campus. Anyway, last Sunday we lost a University professor, we lost a cyclist. Some people lost a friend, a family member. We have to make this case well known, we have to speak with the authorities so they see the Bicipuma system is good, but not enough. Cycling is the only way to go for a city as complex as ours. A much needed first step is to allow proper vial connections linking the University and the many avenues that surround it. Another one is to make a campaign so the cyclists are not seen as Sergio regards as intruders in the streets The street belongs to us all. Cyclists should never drive on walkways or among pedestrians. Drivers need to learn to share the streets with us.

I finally have packaged up a very rough port of my epcp routine from Stata to R as part of a package unimaginatively called cnlmisc; you can download it here. In addition to the diagnostics that the Stata routine provides, the glm method includes a bunch of R-square-like measures from various sources (including Greene and Long). The only part I m sure works at the moment is the epcp for glm objects (including survey s and Zelig s wrappers thereof); the others that are coded (for polr and VGAM) are probably half-working or totally broken, and some wrappers aren t there yet at all. The error bounds suggested by Herron aren t there either. The print routines need a lot of work too; eventually it will have a nice toLatex() wrapper as well. But it beats having it sit on my hard drive gathering dust; plus I may eventually get motivated to write a JSS piece or something based on it. epcp for Stata is still available at my site. For more information on the measure, see Michael C. Herron (1999), Postestimation Uncertainty in Limited Dependent Variable Models Political Analysis 8(1): 83 98 or Moshe Ben-Akiva and Steven Lerman (1985), Discrete Choice Analysis, MIT Press.

There is no clear reason, but for many months already, I have almost given up running Which is a shame, as I enjoy it so much. And yes, I can talk a lot about why not, but outside of one seemingly stupid point there is really no reason. And yes, I have taken on cycling, although on a very different fashion: Cycling is by far not as good as an exercise as running Yes, cycling for ~1h on average 3 times a week has kept my body from becoming sedentary again, but I do feel some increased clumsiness and a undeniable weight gain that I don't want to allow. I said there is one seemingly stupid point: Shoes. My running shoes are around two year old, and they are an important factor on me not running. Why? Because they started giving me blisters at ~5Km. And my shoe size is beyond what is considered as normal by Mexican standards European 44-45 (US 10 -11). Shoe-shopping is always frustrating and very off-putting for me, as instead of what most people do (go to one store, ask for their favorite n pairs on their size, test them on, pay, be happy), I follow almost the opposite process (go to n stores, ask for whatever they have on my size, test it on, pray, pay). Of course, for some people, entering many stores and checking what's offered at each is a thrill by itself. For me, that sounds as thrilling as visiting the dentist for an endodontic treatment. Oh, and to make things even better, I am quite low on cash. Scarcity economics explain that if a given good is hard to find, it will probably be more expensive. So, while many people in Mexico buy ~US$30 shoes, I often get to pay ~US$90. Fun, hah? Anyway, I had a short semi-vacation to Argentina. I decided to kill this excuse, and went shoe-shopping. Fearsome memories sprang in my mind when the first store said they didn't have anything right for me. But they pointed to a second store, one block away, where I found (I was even able to choose between several models!) a nice pair of Olimpikus. In 2008, I was routinely running slightly over 10Km. Today I went out to get back the feeling, and made a not-so-great-but-still-satisfying 4Km. And upon arriving back home, I read Christian's chronicle on his Paris Marathon run Well, only 10% of what my fellow DD achieved But I am definitively back on track. I expect to make some time to get a run, at least three times a week, trying to push my limit at least back to the nice spot I was at two years ago.

A while ago now, Sarah commented that one night when she's woken up in the middle of the night, that she'd observed me stop breathing for a while in my sleep. She wanted me to go see the sleep doctors that she'd seen (she was diagnosed with obstructive sleep apnea a couple of years ago). So I eventually packed myself off to the Stanford Sleep Disorders Clinic, which is at Stanford's shiny new outpatient centre in Redwood City. They took a look at me, and declared I had a crowded mouth, and a narrow airway, and asked me to do a sleep study. I did the sleep study in the middle of January. That was "interesting". I had a ton of wires glued to me, and needless to say, I didn't sleep particularly well, but apparently I slept enough for them to be able to diagnose me. A couple of weeks later, I got a letter with the results. I had a Respiratory Disturbance Index of 16.6. The letter from the clinic defined RDI as including "events of 10 seconds or more with cessation of airflow or discernable reduction in airflow associated with arousal or oxygen desaturation of 3% or more". I had one obstructive sleep apnea event, and 119 obstructive hypopnea events during the sleep study. At one point my O₂ saturation dropped to 89%, but it was very briefly. The overall average was 96%. A month ago I did another sleep study, this time while I was hooked up to a CPAP machine, and they fiddled with the pressure. I was fairly exhausted that night, and slept like a log. Based on the results of the second sleep study, I've been prescribed an auto PAP machine with a pressure ranging between 13 to 15 cm of water pressure (whatever that means). I got the machine, a ResMed S8 AutoSet II a week ago last Friday. It's slightly more advanced than the CPAP machine that Sarah uses, in that it'll adjust its pressure within the range it is set to, as it deems necessary throughout the night. It also backs off the pressure when you exhale, like Sarah's does. I'm not particularly thrilled to be sleeping with this thing on my face. The first night I used it, I slept fine. Something like from 10:30pm until 6am. Subsequently, I keep waking up at precisely 3:30am for some reason, and it's hard to get back to sleep with it on, so I take it off. A few nights, some sort of mask leakage alarm has gone off at some random point in the night. I haven't been in any state to try and diagnose what's going on when that happens, so I just take off the mask and turn the whole thing off. I was curious as to whether I'd notice some sort of life-changing difference between using it and not using it, like Sarah does, but so far, I can't say I've noticed any discernable difference. I also haven't gone a night without using it yet. Getting a properly fitting mask was a bit of a challenge. I spent ages at the durable medical equipment supplier trying on different sorts of masks. The one I've currently got still leaves a bit of a red mark across the bridge of my nose. I can take it back in the first 30 days, and get refitted, so I might still be doing that.

Upload accepted by organisers. 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.

• km 0: I'm in the 4h target sas. Thankfully, I arrived early so I can go to the front line of it, so I'll start about 100m behind a 3h45 pacemaker.
• km 1: 5:12 (Champs-Elys es)
• km 2: 5:24 (Champs-Elys es, Concorde)
• km 3: 5:06 (Rue de Rivoli...not as jammed as I though, hence the time)
• km 4: 5:11 (Rue de Rivoli, Chatelet)
• km 5: 5:07 (Rue de Rivoli, Hotel de Ville. Sadly, I lost my water bottle even though it is meant to be very secure in the trail belt I use. Too bad for the energetic drinks. I won't be in full autonomy as I expected: I hate the mess around Food/Drinks stands)
• km 6: 5:30 (Bastille: slowdown at the beginning of Fbg St-Antoine, but no real jamm like last year)
• km 7: 5:02 (Rue du Faubourg St-Antoine)
• km 8: 5:18 (Rue de Reuilly)
• km 9: 5:10 (Avenue Daumesnil: great time despite the difficult turns at place F lix Ebou ....and the climbing to Porte Dor e. Here I reach the 3h45 pacemaker and pass him to avoid the jam around him)
• km 10: 5:20 (time to slow down a bit as my current time is slightly frightening: I'm around a 3h43 pace. Also, avenue Daumesnil to Vincennes castle is going up!). km 10 is reached in 53:26 according to official recordings, so 5'20". Not that bad.
• km 11: 5:07 (Chateau de Vincennes. I fail to really slowdown....which is probably the bad decision I'll regret later on)
• km 12: 5:12 (Parc Floral. The pace maker stopped for peeing. I'm away ahead of him, dammit)
• km 13: 5:13 (Route de la Pyramide. Good constant pace. Too fast, still)
• km 14: 5:02 (Hippodrome de Vincennes. Dammit, will I ever slow down?)
• km 15: 5:41 (Hippodrome, again. I had to pee..:-)...The pacemaker passes me again and is now running 100m ahead of me...and gazillion of other runners)
• km 16: 5:22 (Avenue de Gravelle. The hardest part of the half-marathon, but this time we're going *down* and this is a good moment to relax. Finally, I'm now using the right pace)
• km 17: 4:58 (Avenue de Gravelle again. The time seems too fast but this is really an easy part and despite the high pace, I'm relaxing. Nothing is hurting as of now. The pacemaker is now only 30m ahead of me)
• km 18: 5:10 (approaching Porte de Charenton and its pavement I dislike a lot....so I'm enjoying the good macadam for now....pacemaker 10m ahead)
• km 19: 5:13 (Porte de Charenton done. Great ambiance, one of the best all along the road, with Paris firemen animating the neighbourhood. Very motivating.....and I now that next kilometer will be goind down!)
• km 20: 5:05 (Rue de Charenton. Another hard point of the half-marathon, but here we're going down....then up! Still I'm feeling very good)
• km 21: 5:24 (Avenue Daumesnil. I'm about 300m behind the half-marathon mark when my watch beeps for km21. Now I came back to my planned pace....il tait temps!)
• Half-marathon official mark is reached in 1h51'25". Prior to my two official half-marathons this winter, this was my personal best on that distance! 5'17"/km, that's however faster than what I planned. That would make 3h42'50". Now I really suspect that I'll regret it!
• km 22: 5:10 (Reaching Bastille. Still (too) fast. I try to calculate how I'll meet Zack and Ralf)
• km 23: 5:20 (reaching the Seine river banks. Crowded place and not much room left laterally by spectators. Elizabeth was there but she didn't manage to see me as she was thinking I was way behind: the SMS service that sends the time at km20 is based....on official time and no real time! I passed the 3h45 flag by 100m and am now seeing another 3h45 flag 100m ahead of me. I'm definitely too fast)
• km 24: 5:20 (Quai Marie. I saw Zack and Ralf and could have a hi-five with them. Ralf should have a pic. I wasn't able to talk to them as I don't like ppl who suddenly stop to hug friends...but Ralf organized things very well and was very precise in his indications)
• km 25: 5:39 (Down to Seine bank expressway. The food/drink before km25 wasn't good and I slowed down to eat one of my gels too...not the first one. Legs begin to hurt and I now feel that I'll have hard times in expressway tunnels)
• km 26: 5:03 (very long tunnel on the expressway. One of the hardest points in the road. It's hot and noisy inside and the moral is not very high. Also, I know there's a steep climbing at the exit. The watch time is probably wrong)
• km 27: 5:55 (Concorde. I told you! Despite the imprecision of time, I know I'm slowing down. The pace maker is now 50m ahead of me again as he passed me in the tunnel. Also, this is a difficult part as it features other expressway tunnels where we go down then up)
• km 28: 4:50 (Grand Palais. Energy seems to be coming back though I really fear that won't last for very long. Pace maker is closer, though. For Dirk: this is the traditional place where the Chicago Marathon, Paris sister marathon is advertizing on the metro bridge)
• km 29: 5:19 (Tour Eiffel. I always fear this part as it is likely that the many tourists who cross the Seine river to go to Eiffel Tower have to croos the street we're running on. Anyway, it's still going well, but the 30km wall is approaching)
• km 30: 5:25 (Maison de Radio-France. Hmm, I'm not feeling very well but apparently I could keep the pace. The pace maker is about 50m ahead of me). I'm reaching km 30 in 2h39'47" for a 5'19" pace.
• km 31: 5:34 (Rue Mirabeau. I'm fearing the rue Molitor slow climbing. 11km left, it's time to use my trick: imagining that I'm running my usual Saturday tour in Maurepas Forest.....so let's say, I'm leaving home and forget that it's about 2h45 that I'm running already)
• km 32: 5:23 (rue Molitor and Porte d'Auteuil. After all, that street isn't climbing that much! Still, I'm now fearing the loops in Bois de Boulogne where there are less spectators and encouragements)
• km 33: 5:37 (Stade Jean-Bouin. Uh oh, now I feel that I'm slowing down *and* can't go faster. First signs of possible cramps. It's time to motivate myself to continue drinking enough as there are still 9km to run)
• km 34: 5:28 (Avenue de la Porte d'Auteuil. That completes the first and smallest loop in Bois de Boulogne...but the longest one is to come, plus the (last) climbing on the Avenue des Fortifications. "My" pace maker is now way ahead of me and I can't see home anymore. I'm on my own with a few other thousand of people around. What's good is that I'm still running faster than most of them (people really overestimate their targets!)
• km 35: 5:56 (Avenue des Fortifications. OUCH. The turn at Porte d'Auteuil followed by the small climbing was hard. Now I know that I'll have hard times in the next kilometers)
• km 36: 5:35 (going around Hippodrome d'Auteuil . Let's go for the most demotivating part of the road: the one where we go the *opposite* way of our destination! And we still have 6 kilometers to run. This is the place where, last years, we were seeing people who were 3 kilometers ahead of us. Not this time as the loop in Bois de Boulogne changed)
• km 37: 5:34 (Bois de Boulogne alleys. I'm happy because, despite of the very hard legs, I'm keeping a pace below enough of the 4h pace....and, even though 3h45 is out of question, I have hopes for 3h50...which was my target)
• km 38: 5:41 (Begin of the very long All e de la Reine Marguerite, for 1.5km. My only thought is now to avoid going under the 5:41 pace)
• km 39: 5:35 (interesting to see this now as I perfectly remember that my moral wasn't very high at that time...but I apparently managed to keep that constant pace)
• km 40: 5:41 (All e de Longchamp: the one where we see Porte Maillot tower very far away and another very long alley, for 1km. But, hey, this is the magical 40km mark! Moral is climbing up. If I manage to not get cramps until km 41, that will be fine!)
• km 41: 6:10 (the hardest kilometer of this run, very obviously. I had to stop to get water and running again is now really difficult. People around are also slowing down a lot as the weather, even if not very hot, is sunny and around the 18 C mark....after a cold winter....thankfully I had worse in Thailand so I don't suffer that much)
• km 42: 6:02 (Porte Dauphine is approaching, it now feels like home, people are excited and spectators are back and very noisy...it's now OK)
• km 42.195: 5:22 (my watch counted more than 500meters over 42km, remember....crossing Porte Dauphine is magic, the noise is incredible and I can....nearly...fly. So, trying to sprint enough to be sure reaching the 3h50 mark)

As a conclusion, even though I can improve the pace and avoid going too fast in the first 25km, I'm very happy as I avoided nearly collapsing as I did last year (with sometimes over 7'00/km and a few walks). 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 No l K the, 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? :-)

I ve just upgraded my Thinkpad (which I use for most of my work) to Debian/testing with KDE4. Improvements Kde 3.5 (from Debian/Lenny) didn t properly display the applets in a vertical task bar. I want a vertical task bar because my screen resolution is 1680*1050 and I find that a less rectangular screen workspace is best for my usage patterns. In my previous post about my Thinkpad T61 I described how the sound controls weren t working [1]. These problems were fixed as part of the upgrade, KDE just does the right thing. Now when I press the buttons to increase or decrease the volume the ALSA settings are changed and a small window is briefly displayed in the center of the screen to show the new volume. Sounds are now made when I plug or unplug the power cable, this was configured in KDE 3.5 but just didn t work. Problems If I have a maximised Konqueror window and I use the middle mouse button to open a link in a new window then the new window will also be maximised. Previously when I did that the new window was not maximised. What sometimes happens is that I want to open several links from a web page in different windows, so if I can open them in non-maximised windows then I can click the title-bar or the bottom status-bar of the parent window to get it in the foreground again. Probably an ideal solution to this use-case would be to configure the middle mouse button to open a new window in the background or minimised. I can t figure out how to implement accelerator keys for window controls. In particular I like to use ALT-F9 to minimise a window (CUA89 standard). The upgrade from KDE 3.5 to KDE 4 lost this and I can t get it back. I want to have an icon on my panel to launch a Konqueror session. I don t want a large amount of space taken up for a launcher for several different Konqueror options, I just want a regular Konqueror for web browsing available at a single click. There didn t seem to be an option for this. KDE 3.5 has an option in the add widgets to toolbar dialogue to add icons for applications. I have just discovered that in KDE 4 the only way to do this is to go through the menu structure and then click the secondary mouse button. Having two ways to do something is often a good thing, particularly when the other way is the way that was most obvious in the previous version! It was annoying that the font choices for my Konsole session were lost on the KDE 4 upgrade, it s not a complex setting. Also the option to resize a Konsole session to a common size (such as 80*25) seems to have been lost. I had to spend at least 30 minutes configuring kmail to get it to display mail in much the same manner as it used to. You have to use the Select View Appearance (Theme) icon off at the right of the Search box and select Classic and then go to Select Aggregation Mode (immediately to the left) to select Flat Date View . I m happy for KDE 4 to default to new exciting things when run the first time, but when upgrading from KDE 3.5 it should try to act like KDE 3.5. I decided to use Kopete for Jabber just to preempt the GNOME people adding Mono support to Pidgin. I had to install the libqca2-plugin-ossl and qca-tls packages to enable SSL connection, missing either of those gives you an incomprehensible error condition that even strace doesn t clarify much. Given that it s generally agreed that sending passwords unencrypted over the Internet is a bad idea and that it s a configuration option in Jabber servers to reject non-SSL connections it seems to me that the Kopoete package should depend on the packages that are needed for SSL support. Failing that it would be good to have Kopete offer big visible warnings when you don t have them. I use the KDE 2 theme and the right side of the title bar of each window is a strange dappled pattern. Not sure why and I have more important problems to fix. Parts of KDE crash too often. I ll start filing bug reports soon. The management of the Desktop folder has changed. In previous versions of KDE the directory ~/Desktop had it s contents displayed in iconic form on the root window. Now by default it doesn t do that. It is possible to change it, but this is one of those things where the default in the case of an upgrade should be to act like previous versions. The way to enable the previous functionality is to go to the desktop settings (click the secondary mouse button on the background, select Desktop Settings and then under Desktop Activity change the Type: to the value Folder View and then specify the directory below. The facility to have different background colors or pictures for each of the virtual desktops seems to have been removed either that or the KDE configuration system doesn t have enough functionality to let me discover how to configure it. When the panel that I have on the left of the screen crashes everything that was next to the panel gets dragged to the left, this includes extending the width of maximised windows. Then when the panel starts again (which if lucky happens automatically) it pushes things back and if icons had been moved left it just obscures them. When using Konqueror to browse a directory full of pictures it doesn t generate thumbnail icons. When I middle-click on an icon for a picture it is opened with Konqueror not the image viewer that was used in KDE 3.5. The image viewer from KDE 3.5 had less options and therefore more screen space was used for the picture. Also the Konqueror window that is opened for this has a navigator panel at the left which I can t permanently remove. When I use Konqueror my common action is to perform a Google search and then use the middle button to open a search result in a new window. Most of my Google searches return pages that have more than one screen-full of data so shortly after opening a window with a search result I press PgDn to see the next page. That press of PgDn for some reason takes me back to the Google search. It seems that when a web page is opened in a new window the keyboard focus will be in the URL entry field, and pressing PgDn in that field takes you to the previous web page. This combination is really annoying for me. Conclusion Getting the sound working correctly is a great feature! Lots of little things are fancier and generally the upgrade is a benefit. The lack of thumbnails when displaying a folder of JPG files is really annoying though. The time taken to configure things is also annoying, I support four relatives who are just users so that probably means at least an hour of configuration work and training for each one so KDE 4 is going to cost me at least half a day because of this.

• [1] http://etbe.coker.com.au/2010/03/16/thinkpad-t61/

Earlier today, I wrote about building a computer with Jacob, our 3.5-year-old, and setting him up with a Linux shell. We did that this evening, and wow he loves it. While the Debian Installer was running, he kept begging to type, so I taught him how to hit Alt-F2 and fired up cat for him. That was a lot of fun. But even more fun was had once the system was set up. I installed bsdgames and taught him how to use worm. worm is a simple snake-like game where you use the arrow keys to eat the numbers. That was a big hit, as Jacob likes numbers right now. He watched me play it a time or two, then tried it himself. Of course he crashed into the wall pretty quickly, which exits the game. I taught him how to type worm at the computer, then press Enter to start it again. Suffice it to say he now knows how to spell worm very well. Yes, that s right: Jacob s first ever Unix command was . worm. He d play the game, and cackle if he managed to eat a number. If he crashed into a wall, he d laugh much harder and run over to the other side of the room. Much as worm was a hit, the Linux shell was even more fun. He sometimes has a problem with the keyboard repeat, and one time typed worrrrrrrrrrrrrrrrrrm . I tried to pronounce that for him, which he thought was hilarious. He was about to backspace to fix it, when I asked, Jacob, what will happen if you press Enter without fixing it? He looked at me with this look of wonder and excitement, as if to say, Hey, I never thought of that. Let s see! And a second later, he pressed Enter. The result, of course, was: -bash: worrrrrrrrrrrrrrrrrrm: command not found Dad, what did it do? I read the text back, and told him it means that the computer doesn t know what worrrrrrrrrrrrrrrrrrm means. Much laughter. At that point, it became a game. He d bang at random letters, and finally press Enter. I d read what it said. Pretty soon he was recognizing the word bash , and I heard one time, Dad, it said BASH again!!! Sometimes if he d get semicolons at the right place, he d get two or three bashes . That was always an exciting surprise. He had more fun at the command line than he did with worm, and I think at least half of it was because the shell was called bash. He took somewhat of an interest in the hardware part earlier in the evening, though not quite as much. He was interested in opening up other computers to take parts out of them, but bored quickly. The fact that Terah was cooking supper probably had something to do with that. He really enjoyed the motherboard (and learned that word), and especially the CPU fan. He loved to spin it with his finger. He thought it interesting that there would be a fan inside his computer. When it came time to assign a hostname, I told Jacob he could name his computer. Initially he was confused. Terah suggested he could name it kitty , but he didn t go for it. After a minute s thought, he said, I will name it Grandma Marla. Confusion from us did he really understand what he was saying? You want to name your computer Grandma Marla? Yep. That will be silly! Sure you don t want to name it Thomas? That would be silly! No. I will name my computer Grandma Marla. " OK then. My DNS now has an entry for grandma-marla. I had wondered what he would come up with. You never know with a 3-year-old! It was a lot of fun to see that sense of wonder and experimentation at work. I remember it from the TRS-80 and DOS machine, when I would just try random things to see what they would do. It is lots of fun to watch it in Jacob too, and hear the laughter as he discovers something amusing. We let Jacob stay up 2 hours past his bedtime to enjoy all the excitement. Tomorrow the computer moves to his room. Should be loads of excitement then too.

apt-xapian-index now comes with a query tool I've just uploaded a new version of apt-xapian-index to unstable. Now it comes with a little query tool called axi-cache. You can search this way:

axi-cache search foo bar baz facet::tag sec:section

In fact, you can use most of the things described here. You can then say axi-cache more to get more results, or axi-cache again to retry a search, or axi-cache again wibble wabble to add keywords to the last search. This allows to start with a search and tweak it. In order to work it needs to save the last search so again or more can amend it. Searches are saved in ~/.cache/axi-cache.state. You can search tags instead of packages by adding --tags. It will suggest extra terms for the search, and also suggest extra tags. It can even correct spelling mistakes in the query terms once the index has been rebuilt with this new version of update-apt-xapian-index. I need to thank Carl Worth who, with notmuch, reminded me that if I just build a nice interface on top of Xapian's query parser I go quite a long way towards making a Xapian database extremely useful indeed. axi-cache also integrates with bash-completion so that tab completion is context-sensitive to the command line being typed:

$ axi-cache search image pro
probability      process          processors       programmability  provides         
problem          processing       production       pronounced       proving  
$ axi-cache search kernel pro
problems     processor    production   proved       provided     
processing   processors   programming  provide      provides     

Thanks to David Paleino who wrote the bash completion script. Just for reference, this is the command line help:

$ axi-cache help
Usage: axi-cache [options] command [args]
Query the Apt Xapian index.
Commands:
  axi-cache help            show a summary of commands
  axi-cache search [terms]  start a new search
  axi-cache again [query]   repeat the last search, possibly adding query terms
  axi-cache more [count]    show more terms from the last search
Options:
  --version             show program's version number and exit
  -h, --help            show this help message and exit
  -s SORT, --sort=SORT  sort by the given value, as listed in /var/lib/apt-
            xapian-index/values
  --tags                show matching tags, rather than packages
  --tabcomplete=TYPE    suggest words for tab completion of the current
            command line (type is 'plain' or 'partial')

If you install the package for the first time, you may need to rebuild the index by running update-apt-xapian-index as root before using axi-cache.

...you want to run many. Everybody told me so. Which is why I was warned when I ran my first one year ago in Paris, I knew I would run other. 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 No l who ran it last year: Ralf is unfortunately suffering from knee problems and No l is preparing for Mainz marathon (target: 3h??). Anyway, more news on Sunday afternoon....and let's enjoy a full rest tomorrow!