Second generation package manager I have created many prototypes for advanced package manager features. Erich Schubert also made a tag cloud prototype. Recently, I wrote:

We have lots of interesting, usable metadata; we have algorithms; we have prototypes; we have ideas for lots of cool, implementable features. The question is, are we able to write applications that just combines what is needed from all this treasure to provide the right interface(s) for our base(s) of users?

Most of the back-end code is already implemented in Python, and I'm thinking of putting it together into a working prototype for a new concept of a file manager. This could be even implemented quickly, if someone joins me who is good at:

• Python;
• a Python GUI toolkit like pyGTK, wxPython or pyQT;
• Debian.

What follows is a sketch description of what I have in mind. The user

Debora is 34, working as a web designer in a small enterprise. She has a laptop running Debian which she uses both for work and at home. She likes Debian for the wide range of software, that allows her to be always able to access the best tools for the job, even in a highly changing environment like the one of web design. At home she enjoys the various tools for photo editing and management, and of course, having a husband and a 6 years old son, the selection of games.

The tasks

• Looking for new software
  • "Is there an application that can do X?"
• Looking at what is installed
  • "Let's see if this package does what I need"
  • "What was that nice package I installed yesterday?"
• Removing things that are not needed anymore
  • "Let's get rid of the cruft"

Looking for new software Use the smart tag search for searching: it allows a Google-like search interface, which afterwards allows categories to be used to semantically refine the results. This last step is conceptually very simple, as it maps to a sequence of "more like this" and "less like this". The search results can be sorted using popcon suggestions, that are very effective at showing software not in the system that many other people use. Looking at what is installed Let's see if this package does what I need Given an installed pacakge, show the contents of its bin/ directories and showing manpages as the details for the bin/ file: thanks to the Debian Policy, executables are all in /(usr/)?s?bin/ and they must have manpages. For packages that are tagged as interface::3d, interface::text-mode or interface::x11, allow to run them directly from the interface. Also show the contents of /usr/share/doc/pkgname and /usr/share/doc/suggested-and-recommended-pkgnames and allow to open documentation with the appropriate software. If there is an html directory which contains an index.html, only show that and not all the other HTML files. Find info documentation and allow to open it into a proper info browser. If there are manpages that don't have a corresponding executable, show them as extra documentation. What was that nice package I installed yesterday? List installed packages sorted by when was the last time they have been installed. Let's get rid of the cruft

• Of course, track auto-installed packages like aptitude and debfoster are doing.
• Also allow to list packages that have been auto-installed satisfying Recommends: lines.

There are also some "unusefulness" scores one can compute, but I haven't yet been playing with them yet:

• Popcon scores (show less popular installed packages first)
• List packages marked as <OLD> in the local popcon scan (and then, by increasing atimes): those are the packages that are installed but are not being used.
• Use negative popcon suggestions, that is, find out a way to compute the opposite as what is found here.

Task filter Have a look at the popcon suggestion page: there is a "Suggestion type" box. Of course the same idea can be used in a package manager to restrict it to a significant subset of the packages. More filters, as you can imagine, can be easily created. This is how it is implemented:

case "any": $filter = "'!role::shared-lib'"; break;
case "gui": $filter = "'!role::shared-lib && role::program && (interface::x11   interface::3d)'"; break;
case "devel": $filter = "'!role::shared-lib && devel::*'"; break;
case "game": $filter = "'!role::shared-lib && role::program && game::*'"; break;
default: $filter = ""; break;

The same filter can be gived to Debtags:

debtags search '!role::shared-lib'
debtags search '!role::shared-lib && role::program && (interface::x11   interface::3d)'
debtags search '!role::shared-lib && devel::*'
debtags search '!role::shared-lib && role::program && game::*'

Conclusion As I said:

There is so much really cool stuff to be written, just within reach.

So?

XSLT's principle of greatest surprise Ruby and Python both claim to adhere to the principle of least surprise, that is: things should work as the human coding in them expects to. But for today the rants about that are on hold. Still, in the last few days I made up my mind on something related: XSLT adheres to the principle of greatest surprise.

• XPath's data model is based on 4 data types: string, number, boolean, node-set. XSLT adds to it one more: document tree fragment (DTF). When you define a variable (or, better since XSLT is a functional programming language, when you bind a value to a name) you play with a value and a name. No matter the type of the value, that value is automatically casted to a DTF after the binding. Guess you can avoid it? Wrong. Guess you can cast it back to another data type when needed? Wrong, unless the desired data type is a string. The only other thing you can do with it is output to the final document.
  • Sure, you say, but you have the wonderful EXSLT extensions which standardized exsl:node-set() to cast back a DTF to a node-set (which probably has been previously automatically casted to a DTF by binding it to a variable name). Then, after crossing your fingers in the hope your XSLT processor implements the given extension, guess with me: since a node-set is automatically converted to a DTF with a dummy root node and one children for each node in the node-set (as it is according to XSLT specs), what do you expect exsl:node-set() on that node set to return? The original node-set? Wrong. You will have back a node-set with the dummy node as only member. ... enjoy tons of exsl:node-set($var)//* expressions in your code
• Being node-sets the only sequence-like data type in the language, what would you do in case you need to first create a list of strings and then pack them with a join-like function intermixing a textual separator? I personally would go for recursively creating a node-set of strings and then join them. Nice try, but wrong. Adjacent strings in a node-set are automatically joined during processing: bye-bye hope of joining them later. The solution? Wrapper elements around each string in the node-set ...
• So you have other wonderful extensions like the set module of EXSLT which implements set-like operations on node-sets. Having seen the following functions: set:difference(), set:intersection(), ... and knowing that for booleans there are true() and false(), wouldn't you expect to have set:empty-node-set()? I would, but unfortunately the problem is not I was unlucky in guessing the name, there is no such function in the language or in the extensions at all. Everyone in using hacks like /..
• AFAIK the most popular XSLT processor on Unix-like systems is xsltproc, which also implements several of the EXSLT extensions. Yay! Of the modules you see listed on the EXSLT web page, what's the one, out of heart, you think you can't live without. . . . . Regular Expressions!, precisely the one xsltproc is not implementing. I had to switch to 4xslt: nice and Pythonish, but not exactly as fast as a lighting bolt.

/me stops at 50% of my ~/Desktop/xslt-sucks.txt If only CDuce had a syntax I'm able to remember ...

Technology and science are, at times, part of some solution. But once unleashed in the world, they can also perpetuate problems and act as barriers. I just finished two articles that have addressed this conundrum in provocative ways and I recommend both to read just for the sake of reading as they are written stunningly well and if you teach a course related to science, technology, medicine and society, these two would make excellent introductions to the importance of critically dissecting science, technology, and medicine (and I think you could unpack them for hours and hours, which is very helpful if you are looking to fill up some time). One, Happy Meal is by the author of the immensely popular Omnivore’s Dilemma, Michael Pollan. In the article he starts with the simple suggestion of trashing most fast/junk/processed food in favor of whole foods, notably plants. But then he delves into the much more complicated topic of the science of nutrition, making a pretty sharp critique of the field, especially for 1) its reductionism 2) and the ways in which the food industry mobilizes nutrition data it is favor and often to the detriment of furthering the goals of real nutrition. After I posted the link on an IRC Debian channel, I had a pretty heated conversation with a number of Debian developers about the nature of science and how one goes about critique it (and mobilizing data to make counter-claims). As often is the case, we ended up arguing over the merit of his attack (though everyone agreed that his food suggestions were quite sound, and I find this disjuncture pretty interesting), as some felt like his attack on the science of nutrition went too far. I did not feel like his article was anti-science or alarmist but saw it as attacking a particular configuration of nutrition science (the reductionism, the inability to admit more complexity, and the turning away from common sense principles), which carry with it serious consequences in the so-called real world. He offers a critical edge that I think does not attack the principle of science generally (after all, he is using the techniques of skepticism and he marshals plenty of scientific data, like studies on omega 3s and 6 to make his point, though he does not only employ the rhetoric of science, which I think is key to understand his critique too) but just what is sees as a particularly problematic rendition of it that is especially problematic for nutrition (though I am sure many of his critiques are easily transfered to other domains). Here, for example, is an attack on reductionism:

If nutritional scientists know this, why do they do it anyway? Because a nutrient bias is built into the way science is done: scientists need individual variables they can isolate. Yet even the simplest food is a hopelessly complex thing to study, a virtual wilderness of chemical compounds, many of which exist in complex and dynamic relation to one another, and all of which together are in the process of changing from one state to another. So if you re a nutritional scientist, you do the only thing you can do, given the tools at your disposal: break the thing down into its component parts and study those one by one, even if that means ignoring complex interactions and contexts, as well as the fact that the whole may be more than, or just different from, the sum of its parts. This is what we mean by reductionist science.

And muses on the effects of this narrow path:

No one likes to admit that his or her best efforts at understanding and solving a problem have actually made the problem worse, but that s exactly what has happened in the case of nutritionism. Scientists operating with the best of intentions, using the best tools at their disposal, have taught us to look at food in a way that has diminished our pleasure in eating it while doing little or nothing to improve our health. Perhaps what we need now is a broader, less reductive view of what food is, one that is at once more ecological and cultural. What would happen, for example, if we were to start thinking about food as less of a thing and more of a relationship?

.
But you will see here that he reaches into scientific facts so that he is not throwing out the bathwater with the baby:

Eat less is the most unwelcome advice of all, but in fact the scientific case for eating a lot less than we currently do is compelling. Calorie restriction has repeatedly been shown to slow aging in animals, and many researchers (including Walter Willett, the Harvard epidemiologist) believe it offers the single strongest link between diet and cancer prevention. Food abundance is a problem, but culture has helped here, too, by promoting the idea of moderation. Once one of the longest-lived people on earth, the Okinawans practiced a principle they called Hara Hachi Bu : eat until you are 80 percent full. To make the eat less message a bit more palatable, consider that quality may have a bearing on quantity: I don t know about you, but the better the quality of the food I eat, the less of it I need to feel satisfied. All tomatoes are not created equal.

There is no doubt that the article is advocating, very strongly, a point and he is not being all impassioned and neutral as may happen with a purely scientific rendition of the the topic. This will no doubt cause debate and reaction, which I am sure is what he was aiming for. And whether you end up agreeing or not with his critique, it makes for one fascinating read. Along with this piece is another engrossing, somewhat older one piece by Carl Elliot: A New Way to be Mad that examines the ethical complexities of what seems to be a relatively new phenomenon: Body Integration Identity Disorder (which he may have helped fuel a little via his very article and it is this issue of looping, first discussed by the philosopher of science, Ian Hacking that Carl Elliot nicely captures). Since the article is not available for free, here is another one, also by Carl Elliot, that addresses the same topic Unlike Pollan, Elliot does not have any clear cut answers or a really trenchant critique. More than anything he serves up a tasty meal that will in all likelihood be new to many readers, and the flavors he adds are those that make us confront the many interesting moral quandaries that this case presents.

I reject the unconstitutional honorary membership allocated by debian.ch and I have not made the mistake that I made with DUS, of possibly accepting membership by conduct. You can tell debian.ch is nothing to do with me: its director can't even print my name right. Will debian.ch continue to be used for personal abuse? Will the next DPL actually defend the debian name against unethical enterprises? Update: To madduck, because you don't even have comments on your blog: Yes, the debian user database is wrong. I don't know why my gecos is different from my GPG key, most of my IDs and my AM report and I've asked for it to be corrected for years, as readers of debian lists may remember, but it's well-known how responsive admin@db.d.o can be sometimes. IIRC, Ian Jackson once refused to sign my key because my IDs and db.d.o disagree. Seeing as it was posted to another blog, I'll delete the similar comment, but that's fairly unusual for me.

(update: numerous typos removed) This year, attending the whole Solutions Linux event wasn't compatible with my paid work (for those who care, I'm since September head of the Personal and departmental computing unit in the Networks and Information Department of ONERA, the French Aerospace Lab). Despite this, I booked at least yesterday afternoon to visit the expo and more particularly the Debian booth. It is always a pleasure to meet again French developers and contributors as well as all these old dinosaurs which I learned electronic communications with 1200bps modems with, some time ago. The Debian booth was as crowded as usual. It was still featuring a Babelbox, built in one night by Pierre Habouzit and Julien Blache. Congratulations, you did an awful job here. I'm glad I saw all those good people. I was tempted to cite everybody and will finally refrain because I'm too afraid of forgetting one of you guys. For our international readers, this includes several of the people I may have had a few flamy discussions and even blog "wars" with in the recent past (I'm just missing you, Joss, being told that you couldn't attend that Wednesday). As usual, we all survived very well. The other "big" event was also the first formal meeting of the Debian France non-profit organisation. Everything is pretty young but the hidden creation work done by the founding members has been huge and the base is now here. Debian now officially exists in France. I'll do my best to play my part in this action and work with other members of the board to fit the organisation goals and bring Debian more prevalence in our country. I also had a very long talk with Bruno Cornec and Louis Bouchard from HP. We discussed about HP customers needs for official support for Debian on HP machines, and not only servers (ONERA has an internal project to deploy a few hundreds of workstations aimed at being the scientific platform for ONERA scientists and PhD students. We already have more than 100 of these, based on RHEL and guess what are my mid-term plans for them? :-)). We just need some official commitment from HP to support using Debian to unlock the internal fears. It sounds like HP could go this way if customers show enough interest. Of course, ONERA's project is pretty small but we are one of the key players in out domain in the country so that could be kind of interesting marketing for HP...and Debian..:-). Bdale, I need to talk with you soon.... Back to work now, for a few days. Then, I'll take 3 days off next week to attend the 3rd Free Software World Conference in Badajoz and present the results of the Extremadura i18n session we held in September, along with other people who held some of these sessions, namely Frans Pop (about the D-I session), Wookey (about the Embedded Debian session), Knut Yrvin (about the Debian-Edu session) and Martin Michlmayr (about the QA session). All people who I will be very glad to meet again.

Milton Country Park

This park near Cambridge is under threat, partly due to council mismanagement.

Climate change prompts Cabinet change in Canada - CNN.com

Carry on flying, says Blair - science will save the planet Special Reports Guardian Unlimited Politics

Weston and Worle News: Recycling Service Branded Rubbish

"A Weston resident has branded the town's recycling service rubbish and joined calls for it to be investigated.Carolyn Jones, who moved to Weston Village from south Gloucestershire 18 months ago,..." I've had to telephone and get Biffa to come back after they missed my collection every time since New Year. What's going on?

Linux Ecology-HOWTO

Nuclear Power?

Cycle Parking Farcility of the Month

CTC - Cyclist who refused to stay in the gutter wins

So you are allowed to cycle on the roads in Telford, despite what the police there think. More coverage:

Cyclist's conviction thrown out : Shropshire Star

BBC NEWS England Shropshire Cyclist wins road blocking appeal

Cycle-lanes - epetition reply

A relevant reminder from Tony Bliar's office that cyclists don't have to use useless cycle lanes.

Trains Swindon - Melksham - Westbury - SLASHED by the DfT and First

Campaign site about the government-enforced cuts to Great Western services in the Wiltshire/Somerset/Bristol area.

EDP24 - Dangerous driver in cyclist death jailed

BBC NEWS England Norfolk Tired motorist jailed over death

High-res plans of the TdF London stages

Friday was my birthday party, and is was a success. I got some nice presents even though I was unable to think up even one thing for my wish list beforehand. As an experiment I bought vegetarian bitterballs (a popular Dutch snack) to deep-fry, which were surprisingly well received. Stefan arrived at 03.00 so the party could then really commence. Saturday afternoon I spent with Matthijs tweaking some of our 1996 internet project, after which I went to visit Gerd-Jan's birthday party, who was born on the same day (and year) as myself, so I couldn't miss that. I was persuaded to join in on the current fad, poker, which left me with a whopping debt of five euros. Sunday marked the last day of the Winter Efteling season, which I visited with some Eftelist-moderators. I do not really like the WE, especially because it mostly lacks the green environment and real snow, but gains rain and cold, but it was interesting to study the Vogel Rok improvements. We closed the season off with some drinks in the Efteling Hotel bar, where I suddenly discovered it to be midnight. Not that I'm Cinderella, it rather meant that the last trains had already left. Jos and Ilona were kind enough to drop me off at Rotterdam Central station around 1:05, where I could take the 2:00 train. The night train takes this scenic route: Rotterdam, Delft, The Hague, Leiden, Schiphol Airport, Amsterdam, Utrecht; so it was 4:15 when I was home. Now for some rest.

I've just noticed that some debconf7 registrations close this week. Yes, this week - a full two months before national rail advance tickets go on sale. What's that about? [...] Jon Dowland replied

"MJ, the debconf7 travel page has far more flight information on it than anything else because it has been put together primarily by people who are figuring out international travel arrangements which will require flights."

International travel does not necessarily require flights. Intercontinental travel probably does, due to time constraints, or maybe even cross-continent, but I've seen DDs from nearby countries posting about booking their flights, and putting the deadlines before land tickets are even on sale is really shameful.

"It could certainly use some more information for people travelling more locally. If you have some advice on how to get there from Europe or the UK without flying, please contribute it: the site is a wiki, you can sign up for an account and edit the page yourself."

As I've posted elsewhere, it's a mediawiki rather than a wiki, so once I get around to installing mvs, I can sign up and edit it. (Bring back patches in the mail...) However, I'm not sure how best to get to Edinburgh without flying yet, because most tickets aren't on sale yet, which is where I started, remember? David Moreno Garza replied

"in 3rd world countries, some things like technology, travelling or taking care of the environment are not a matter of choice, unfortunately. [...]"

Sure, if people are travelling from 3rd world countries to Edinburgh, then flying is probably the only practical way, but surely people from nearby countries can get to Scotland without dumping CO2 into our sky?

These are the current translation ratios for French and Czech in unstable, after 12 days of "blitz l10n NMU". Nothing happened yesterday, which explains why I didn't post. Today, a few packages uploads in unstable should have improved the stats for French. However, this is also the day where dpkg-cross introduced 3 strings without prior notification to translators. A good occasion to test my "Please interact with translators" new script..:-) Czech has improved a lot as a consequence of numerous package uploads after my strong activity during the week-end. Thanks again to all these responsive maintainers. So, now we're hunting tha last remaining parts for French:

• postfix, on its way by LaMont Jones
• s390-dasd, a D-I component, already completed in D-I SVN and which will be uploaded by Frans along with all other D-I updates
• conserver, which has indeed been uploaded today by J rgen H gg
• libpam-rsa, which I'll sponsor in a few minutes
• dpkg-cross, as explained above

Another package will soon appear, out from NEW: emdebian-tools, uploaded by Neil Williams, who took care to request for translation updates in debian-i18n. Thanks, Neil.

I had this planned for over a year, but for whatever reasons, I didn't get around to doing it until last week - Knowing that my Institute was moving to almost the opposite edge of Ciudad Universitaria, and would now be located at over 3Km from home (from around 300m that it was before), I decided to get a bike. And, after having it in my to-buy list, I bought it last Sunday (January 21), and Nadezhda got one for herself as well.
I had had a bike before, over 15 years ago, and we rode a bit when we were in Amsterdam, but that was back in 1996. Since then, I cannot recall using a bike - So, of course, I was quite shaky and afraid to begin with. I must admit, I envied Nadezhda, as she seems to be natural - While I was still trying to get the bike moving without crashing into something, she was literally making circles around me.
On Sunday, we went for a short ride in the main circuit of UNAM. On Monday night, we went most of the way to the gym/running lanes where we go to make our excercises - Yes, mostly uphill, and I managed! (she had already done that road earlier the same Monday). Thursday and Friday, I rolled early to do my excercise, and we basically crossed each other on the way. On Saturday, finally, we came all the way to the new Institute.
Sunday was amazing: We went with her family (two sisters, one brother and five pedalling nephews, plus one nephew in a basket in his mother's bike) along the Ciclopista. The Ciclopista goes on the path that used to be the Mexico-Cuernavaca railroad, and starting at Contreras, it goes uphill to cross the Ajusco/Chichinahutzin ridge. I cannot find a decent map, and I don't have a clue on what distance we rode, but it was just great - We crossed basically all of the Magdalena Contreras part of the Ciclopista and part of Tlalpan, until maybe 500m past the market at Calle 8. Of course, after the ride, we feasted with quesadillas at the market, plus some tacos of a delicious chicharr n prepared by Nadezhda's mother. It was a long ride anyway - I got quite tired, and the way back (fortunately, downhill) was still long. We had promised the four year old nephews to take them to the zoo that same day, although we counted on them being tired by that time - Well, a child's will is stronger than being tired, so we still went to Chapultepec to dive in a sea of people and look at some nice animals ;-)
...Finally, yesterday and today I have come by bike to the Institute. There are a couple of hard parts, and I still have not mastered the techniques for being an effective rider, but the way back is just delicious. Besides, yesterday I made 15 minutes on my way back, today it was 20 minutes coming here - almost the same time that going by car! Oh, and on the way back, I get to enjoy a beautiful view of my beloved volcanos, Popocat petl and Iztacc huatl. And, as Mauricio already pointed out, thanks to the light rains we had in the previous days... They just look amazing, covered in snow from their very bases.
Not much to add to it... But I'm delighted :)

MJ, in 3rd world countries, some things like technology, travelling or taking care of the environment are not a matter of choice, unfortunately. When some of the basic needs for humans to live are easily covered and people do not have to worry about, suchs as daily eating, it’s way more easy to develop a country and have people concerning about unimportant stuff. Even free software is a luxury matter on some countries! You know, human beings have an stupid custom acquired since they were born, most of them, and that’s the need for eating, how weaks! I think you get my point. Yours, an international traveller flying between the UK.

Dear Lazyweb, how the .... am I supposed to tell GNOME to use or not use an HTTP proxy for specific applications? Sample scenario:

• I want to use an HTTP proxy in Galeon (for anonymous browsing via Tor + Privoxy).
• I do not want to use an HTTP proxy in Epiphany (normal non-torified browsing).
• I do not want to use an HTTP proxy in Rhythmbox either. No need to tunnel all the Creative Commons podcasts and music I listen to through Tor.

Now guess what happens when I disable the HTTP proxy in Epiphany. It's disabled in Galeon, too. Enable it in Galeon, and Rhythmbox will use the proxy (thus slowing down huge downloads for no reason). Aargh. Is it really so hard to have per-application settings? I mean, this isn't exactly rocket-science, right? And yes, I do want to use all those applications at the same time. And no, I do not run a full GNOME desktop environment (I use IceWM, thanks), so I don't care about any GNOME-Desktop-Foo solutions — I just want each of those freaking applications to have their own settings.

Looks like our RC counter is heading into the wrong direction. Two years ago — when our RC counter was as high as now — it took approximately three months from there to the release. So…

I have become painfully aware of the fact that I cannot concentrate on anything for more than fifteen seconds before I am distracted by something. This ofcourse results in a severe decrease in productivity. I’ve also noticed this is a very common issue.
A list of things that distract me include IRC (Admit it, you too.), shiny objects, music containing lyrics, lack of sleep.
To do list; Cut down on IRC, spend one or two hours a day in a silent room with no shiny objects, sleep at normal hours.
I will lock myself in a work environment for one hour a day to do something 100% productive, and gradually increase the time as I get used to it.
How long will I be able to keep this up I wonder? PS: I have noticed that there are no decent laptop bags out there. I’m looking for something casual that can be used for everyday use as well as containing a laptop, that suits a 12″ laptop and is preferably not bigger than that, and that most importantly does not scream “HEY I HAVE A LAPTOP” (You’re just asking to get robbed). Inform me if you know of any.

I was recently asked to help someone understand the OpenSource movement, and the effects it might have on companies such as Oracle or SAP. I talked a bit that these companies can benefit from being independant of big vendors like Microsoft, while still not having to do everything themselves. But also on how opensource alternatives to their products will threaten their business in the long run and force them to be more innovative and constantly improve their product. I then talked about the 'viral' aspects in OpenSource; not about that GPL licensing bullshit (just read the licenses of stuff you want to use, and don't use them if you don't like the license...; this virality is highly overrated due to Microsofts FUD), but a very different kind of virality: Let's say you are supposed to evaluate different CMS solutions. So you install one or another that sounds promising. Most likely you'll install an opensource database like MySQL or Postgresql. And when you've decided upon a solution, you are quite likely to go with this database, and say to yourself: "if MySQL doesn't scale up well enough for our needs, we can still but an Oracle license". This would be probably different if you had had installed Oracle in the first place. Say a demo or so. But the demo versions of the CMS you tried already made you comfortable in using MySQL; it's easy to install (most likely included with your system anyway), and it just works. But would you have tried a CMS in first place that actuall required you to setup Oracle? Anyway, this "social" virality brought me to another point: developers and administrators. When you see numbers on the market share of Microsoft, they're still huge. Some 95% of users are using Microsoft as their primary operation system (I'm discounting the fact that a huge share of DSL modems here is running Linux, so actually many people are Linux users, and I'm also discounting the use of www.google.com, which would make 99.9% of people Linux users...). But these are the average users. But they aren't on their own. The software choice of the "average user" is influenced by a couple of factors, including:

• if anything goes wrong, whom can I call for help?
• company (administrative, security) policies
• word of mouth and (software) recommendations

But if you look at the computer geeks around you, what are they using? A huge share of the people I know are using Linux. Not everyone is using Linux as their main desktop system, but many are. Others are using Linux just for their development stuff, some only on servers. It doesn't really matter, Linux is huge with the geeks. When I look at the sofware my mom has installed on her Windows system (which also dual-boots Linux, and interestingly is the primary choice for watching a movie - Windows doesn't playback sound with some videos...): half of it is opensource. There is OpenOffice.org, Gimp, Gaim, Firefox, Thunderbird, Inkscape, LyX, Mplayer and a couple others. He favourite game? PySol. My dad already has switched to Linux as his main operating system. So why did they switch to so much opensource software? Because it's what I recommended them, and where I can help them. My mom needed a software to record something for her blog (yes, she's "podcasting" now) - of course I recommended her an opensource application, audacity. My mom is giving private lessons in math and physics; she's very successful at that. When one of her students got herself a laptop, she of course was given all kinds of (opensource!) software. For her highschool thesis she needed a vector graphics program - she ended up with inkscape. Now if my parents have some problem with Windows, who are they going to ask? I didn't manage to solve the video playback issues, my mom by herself switched to Linux for watching movies. It didn't work with Windows, I couldn't help her - she switched to something where she knews she'd get more support (and where the problem by chance didn't occur). Now let's have a look at the university. Our computer pool for the students is all Linux. There are a few windows systems for legacy purposes, I never found out if they are accessible to regular students; I only know they're supposed to exist. When doing some assignment, you're very often required to use some opensource software. For a thesis, you'll very likely be building upon some opensource system. The mathematics institute is still on solaris, but recently Gnome was installed there. Physics and electrical engineering also have mostly Linux systems, I've heard. For software development, it's pretty much a "develop on Linux, run anywhere", whereas Windows-only developers usually need to port their applications (or trust on Wine being able to run them good enough on Linux). So more and more people start developing on Linux, and get to love it's great development environment (after all, Linux was pretty much completely done by software developers...). And that is how Microsoft might already have lost the desktop war: developers, technical students and administrators (who usually also love their Linux boxes). They're only a small share of the userbase, but they're influential. In fact, other users rely on them being able to eventually fix their systems for them. But since these are using Windows much less already, they will be less capable of assisting others with Windows. They'll be putting Linux systems everywhere, and keep on encouraging them to make the switch. Of course it will still take some years, but you can expect more and more companies (and cities) to switch over, like the city of Munich. Windows support staff on the other hand might even become more expensive. I don't know of any numbers on the MCSE programme (Minesweeper Consultant, Solitaire Expert) - does anyone have numbers on "graduates" for that?

Today Robert Millan announced http://goodbye-microsoft.com/, which makes loading and booting d-i from Windows easy. These screenshots show how it works, and I like this one especially: The Windows program is even compilable from Debian, so this can be wrapped up into a Debian package and go into the official archive eventually. Gorgeous.

---

Update: slashdotted -- screenshot mirror here, front page mirror here

This addition to my Bad Technology page is plumbing. I hate plumbing. It is really inconvenient if it goes wrong, plumbers are pretty rare these days and most of my attempts to fix anything to do with plumbing end up wet and/or messy and/or damaging the house. I was just starting to understand the sealed system at the old house and then we moved to a house with an open vent system with loft tanks and inverted loops and draincocks and stuff. Humanity has invented the internet and other great stuff, but it seems we can't invent a heating system that is reliable, affordable, environmentally-friendly, understandable, hackable and doesn't fail messily. On what may have been the coldest day since we moved here, the heating stopped. Testing the controls, looking at the system design and working through a troubleshooting list suggested that the sensors were working and the boiler was working but the valve that sends hot water to the tank or heating was just clicking and sending nothing to the heating. I tried calling our usual plumber. Voicemail. So, I popped the cover off the valve. In addition to the wet/messy/damaging worry, that also involves electrics. Water and electicity - an ideal combination(!) There were some springs, a round bit and a worrying number of wires (live, for all I know). Also dust. A lot of dust. Thick lumps of dust. Grab the air gun, blow out the dust. Set the controls to try the heating. Still no joy. The round bit seemed to be twitching and making the ticking noise. Tapped it with the screwdriver handle. Nothing. Hit it. It moved. Pump starts, boiler starts, sound of heating pipes expanding as the hot water enters them again. I think that's my first successful fix to the plumbing in three months. Be cautious, but give it a go - I guess that's the same tactic that works for my successful computer hacks.

I suggested in a previous blog entry that conferences should provide computers that speakers can use for their presentations. The reason for this is that getting one computer working with the beamer in each room is an easy task, while getting the laptop of every speaker to work is much more difficult.

It seems that my idea has been rejected by almost everyone who read it, so I'll document some tips for getting a laptop working.

 SZ:    Pixels          Physical       Refresh
*0   1400 x 1050   ( 474mm x 356mm )  *50
4    640 x 480    ( 474mm x 356mm )   50
5    800 x 600    ( 474mm x 356mm )   50
6   1024 x 768    ( 474mm x 356mm )   50
8   1280 x 960    ( 474mm x 356mm )   50
9   1280 x 1024   ( 474mm x 356mm )   50

Firstly there is the command xrandr which can be used to change the resolution without logging out. Above are the most useful lines produced by running xrandr with no options on my Thinkpad T41p. The left column is the index to the list of resolutions. For example I run xrandr -s 9 to use mode 1280x1024 and xrandr -s 0 to use mode 1400x1050. This takes much less time than editing an X config file!

The next thing to note is that my Thinkpad has a refresh rate of 50Hz, apparently most beamers expect at least 60Hz, this explains why I have had ongoing problems in getting my Thinkpad to correctly work for presentations for the entire time that I have owned it. If you own such a Thinkpad then I recommend that you just bring another laptop to do your presentation on the assumption that the display possibly won't work and probably won't work properly! I had developed this habit anyway after repeated problems in getting my Thinkpad working (occurring on a number of occasions in several countries). It's good to now know the reason for this (thanks Keith).

When setting the resolution there are often tweaks that can be used. For example in my talk for the Debian Miniconf of LCA 2007 I used mode 800x600 (I think - Keith set it up and I didn't look closely after verifying that things basically worked). Even though the beamer didn't have good support for a low refresh rate it worked when the resolution was low enough. Fortunately the xrandr program allows changing resolution fast enough that all 13 resolutions could be attempted in about a minute.

The support for better display detection and configuration is steadily improving. Hopefully this year the problems will be solved (which means that for the Debian and RHEL releases in 2008 the problem will be solved).

A possible work-around is to use Xephyr (the replacement for Xnest). In a previous blog entry I described how to get Xephyr going for use by Xen images. It seems to be a common symptom of display synchronization problems that the edges of the screen will be clipped. The most common work-around for this is to not use the full-screen mode of OpenOffice - which means that instead of having a small amount of text clipped there is a large amount of OpenOffice menus etc on the screen. As Xephyr accepts any resolution it should not be difficult to arrange for it to use 98% of the screen space and then run the presentation full-screen in the Xephyr window. This will be particularly useful for programs such as MagicPoint (my favorite presentation program) which don't support a windowed mode of operation.

If you have any other suggestions on how to solve or work around display problems with laptops then please leave comments.

(Stop the tests for now, it seems I made a mistake. Will fix it tonight and make an update.)

I have been working recently on an update to the ppp-udeb package (responsible for configuring PPPoE networking in D-I and the target system).

I think I have finished fixing the last bad bits from the ppp-udeb package[*], but currently I am unable to do a test of a complete installation due to bug #407702 in qemu. Still, I am confident it should just work.

So I am asking if there are people interested in testing the image I have prepared with the new ppp-udeb package. The image can be downloaded from:



http://users.alioth.debian.org/~eddyp-guest/ppp-2.4.4rel-5/mini-with-ppp-udeb-20070124.iso



(Security concerned people can check the signature and md5sums and go on, if they trust me :-) )

How to test

In order to test, just boot the image (no boot parameters are needed) and try to do a basic installation via PPPoE.

What should be (or would be nice to be :-) ) checked:

• in D-I environment
• bad/good login should be detected (good logins are silent)
• transitory problems when trying to do the initial connection (cable unplugged or PPPoE server down) should be detected
• in the installed system:
• target system boots fine with the PPPoE connection set up correctly
• hosts file and host name are set correctly
• permissions are set correctly on the network configuration files and the ppp configuration files (an output of "ls -lR /etc/ppp" in the target would be appreciated)
• loopback interface is configured correctly

Please respond as soon as possible (eddy.petrisor at_sign gmail.com).

[*] I did some tests myself but I can't test everything due to bug #407702 in qemu; some screen shots I made in qemu can be found at http://users.alioth.debian.org/~eddyp-guest/ppp-2.4.4rel-5/screen%20shots/


Update: thanks to some people that did the test, I found out that I made some wrong assumptions in my code. I will make another image tonight and make another call. This will hopefully fix the issue.