I guess one could summarize Easter as a certain prophet's big reboot?
Now, make it funky!
Yup, this Easter, I completely rebooted as well: after briefly visiting my relatives in some distant country that some morons insist I should call "homeland", I got back to my real homeland in the HELish land of singing demons for a full reboot: new job, new home, new girlfriend.
Then, my computing resources rebooted too.
Shake it to the Left! Shake it to the Right!
Reboot your life.
Life sometimes goes in a funny direction, like a full circle déjà vu: 20 years ago, I was studying Technology of Computerized Systems (a combination of electronics and programming). Now, 20 years later, I'm starting a dream job as the Business Development Manager of Artec Group and their core business is — you guessed it — exactly that. The really cool thing about Artec, from a Free Software advocate's point of view, is that they contribute firmware and drivers for their products to LinuxBIOS and the Linux kernel.
At this point, my initial tasks will focus on marketing and selling Artec's flagship OEM product, the ThinCan. What is the ThinCan? Actually, you're already familiar with one iteration of the product: the Linutop is a branded ThinCan, delivered with a custom hardware configuration, plus custom casing artwork designed by Frédéric Baille of Linutop SARL and loaded with Fred's favorite Xubuntu configuration. In the Linutop's case, their business model is to fill the market niche for a simple Internet surfing platform that fits 80% of average people's daily computing needs, so the hardware configuration they ordered reflects that.
Still, as several readers of this blog noticed, the "Linutop" could make a fantastic thin client and, sure enough, that was the main purpose behind the ThinCan's original design; it sells well in the Fortune-500 market as an RDP client running under Windows CE. However, until recently, there simply wasn't much demand for a contemporary X11 terminal solution based on the ThinCan. That is, until LTSP took off, thanks to the contributions of Edubuntu and similar educational Linux distributions. Sure enough, someone spotted the opportunity and contacted Artec to order a branded ThinCan iteration with Etherboot and PXE support, which they call the Linuterm.
By the way, for those who need a really cheap OEM thin client, I've got a great spring clearance offer for you:
Artec has about 500 pieces left of their older DBE60 ThinCan model, based on the AMD Geode SC2200, and we're selling them at 100 euro / piece, plus VAT and shipping.
The DBE60 is configured with three USB 1.1 ports, one parallel printer port, one 100baseT Ethernet port, one VGA port (up to 1024x768 @ 16bpp 60-85Hz or 1280x1024 @ 8bpp 60-75Hz) and one 1/8" stereo audio output jack. Its BIOS provides Etherboot support and its motherboard is populated with 64MB of RAM and 32MB of Stratoflash. It comes delivered with a European AC adapter. The Geode SC2200 is fully supported by the Linux kernel — with the sole exception of a missing ALSA snd-scx200, but this could easily be ported from AMD's deprecated OSS driver — and X.org support is provided by the "nsc" driver.
Minimum order size is 10 pieces. Contact me via my full name (unaccented, with one hyphen and one dot) at artecgroup.com quoting this special offer.
PS: someone was asking if that DBE60 special is also available in the aforementioned Fortune-500 configuration. It indeed is: add 16 euro / piece for the Windows CE 4.1 license with an RDP client.
Fred informs me that we now have our ordering system up and running so, for those of you who had been longing for their very own Linutop, do visit us and fill out the order form.
Please note that, at this stage, we are only able to ship within Europe. However, if anyone is interested in distributing the Linutop on other continents, we definitely want to hear from you.
I recently blogged about putting together the Linutop source code ISO. The resulting image has now been uploaded to our wiki. The ISO's content is divided in two sections:
Developers who are interested in producing customized OS images should download this ISO and consult the wiki for details on developing for the Linutop.
Dear Lazy Web:
I am developing a severe hatred of Gaim 2.0, because some of its dumb features hamper my productivity. Are there any simple configuration tricks that could fix the following issues:
Would you happen to know the answer to any of the above?
Best Regards,
Mr.Linutop
A few weeks ago, the Linutop team received a request for its source code. We had already anticipated the GPL source offer clause in our development plan, so it was just a matter of myself getting around producing a source code ISO image. Piece of cake, right?
Almost. You see, Linutop includes a custom kernel package and several separate wireless module packages, so the the following command would not work as expected:
apt-get --download-only --ignore-missing source $(dpkg --get-selections | cut -f 1)
Why is that? Because APT ignores the --ignore-missing option whenever executing the source command. Instead, Roland Mas suggested that I used this simple Bourne loop:
for p in $(dpkg --get-selections | cut -f 1) ; do apt-get --download-only source $p ; done
Done this way, the source code of each package is fetched individually and, if any package's source is not available, APT exits but the Bourne loop moves on to the next package in the list. Nifty, isn't it?
Linutop will makes its first public appearance tomorrow at Solutions Linux in Paris. French developers who are interested in getting their hands on one of our 50 development units should call up Fred at 0685868576 and schedule an appointment. Others who just wanna see the Linutop in action are welcome to drop by booth D 16 during the exhibition.
Meanwhile, rumor has it that someone else might be dropping by Montréal for the February meeting of FACIL with a mysterious Free Software -powered aluminum box...
I previously blogged about UTF-8 Migration Tool, a GTK2 wizard that helps users upgrade their legacy locale and recode text files to their UTF-8 equivalent. After merging a number of patches from Nicolas François to fix various parsing issues, I am now happy to report that the package has entered Debian's unstable distribution.
Plenty of migration cases remain untested, though, and the Debian i18n team particularly welcomes testing for locales that utilize a non-default legacy encoding or that contain an @modifier part.
If no major issue is found, the goal would be to release this package with Etch to help users migrate their system to UTF-8.
Kenshi sent the following GPG-signed message to the Debian CUPS mailing list today:
Hi folks,As you know, there are many bugs about CUPS including Release-critical and I haven't an enough time to fix them.
My primary motivation was to internationalize/localize it. Now it is already implemented by upstream.
Unfortunately CUPS is too complex for me to maintain. I haven't a time to track bugs, I haven't an enough knowledge about Postscript or printer specific codes, and I haven't any testbed of user's printers.
If you'd like to become primary CUPS maintainer instead of me, I pass a baton happily :) Of course I'll support you such as uploading.
Thanks,
Interested parties should contact kmuto directly to manifest their intention.
One recurrent theme in discussions about Copyrights and Digital Rights Management is that the Media Industry allegedly represents the interests of the Artists. Yet, recent actions in Canada and elsewhere prove that Artists who benefited from the wider exposure of their creations via the Internet in fact oppose laws that would preserve and further entrench the monopoly of the Industry.
Finland's case is no different: as reported by Jaakko Kuivalainen in the Tekijänoikeuslaki blog [in Finnish], a number of Finnish Artist Coalitions are currently opposing the transfer of copyright issues from the Ministry of Education (who also handles cultural affairs) to the Ministry of Commerce, because they fear that the interests of the Media Industry would suddenly have precedence over those of the Artists and small Independent Cultural Producers.
Yet another proof that, no, Media Moguls do not speak on behalf of the Artists.
From the Debian Bug #374997 department:
* Package name : utf8-migration-tool Version : 0.4 Upstream Authors: Tollef Fog Heen, Martin-Éric Racine * URL : http://q-funk.iki.fi/debian/pool/u/utf8-migration-tool/ * License : GPL Programming Lang: Python, GTK2+ Description : Debian UTF-8 migration wizard This wizard upgrades legacy system locales to their UTF-8 equivalent. It also informs users whenever files in their home directory still utilize legacy encodings.This started as an Ubuntu tool to enable easy migration to UTF-8 for both locale settings and user file encodings. Tollef says that since Ubuntu has been UTF-8 by default for a few releases already, they are not likely to further develop it and invited me to take over development, so I have.
I have found this tool very useful to help me locate remaining files in my home directory that are still in a legacy encoding and to check system files for UTF-8 locales utilization.
Given how Etch is going to be the first Debian release with UTF-8 locales by default, I figure that it could be a useful migration tool for others as well.
Since I'm currently paring down my involvement in Free Software, I never got around fixing the GTK annoyance reported by Denis Barbier in response to the ITP and thus never uploaded the package to the NEW queue so, if anybody is interested in picking up this package's maintenance, please do so by responding to the above bug.
This week, following the announce of the freeze, I upgraded most of my home cluster to Etch. The upgrade went surprisingly smoothly, although I had a bit of a scare when the kernel upgrade pulled in a bunch of dependencies that resulted in the forced removal of all previous kernels. This minor scare asides, kudos to everyone who made this new Debian release possible!
My only disappointment about Etch is that we're stuck with Firefox 1.5.0.7 instead of 2.0.1 or Iceweasel. Then again, speaking of Iceweasel: Forget it! Breakages it introduces have been marked as pending or patched for a number of weeks and yet there's still no updated package. At this rate, better wait until Etch r1 or Lenny to introduce it, rather than release Etch with a browser fork that has not undergone several months of thorough testing!
This got me thinking about life after Etch, or more specifically, about what I'm gonna do with the packages I maintain. The Dunc Tank debacle and similar other unfortunate events that overtook Debian this year have prompted me to reevaluate my motivation. Another unrelated issue is, I've been working in software development for more than 10 years and my guts are telling me that it's time to move on so, while my computers will still run on Debian or Ubuntu, I don't intend on doing any more development per-se.
Thus, I have already taken the initiative of soliciting co-maintainers for most of my packages. The Debian GNOME team agreed to be added for Planner, which is nice given the sheer number of bugs that are regularly found against this package. CUPS-PDF, Gaim-IRChelper and NumlockX haven't seen a meaningful bug in ages and can be rebuilt using bin-NMU magic, so they're pretty much taken care of. This leaves the issue of the Estonian and Russian dictionaries (plus a large patch in my personal repository that would upgrade Aigars' Latvian dictionary to generate the Aspell wordlist from his MySpell wordlist) open. For these, native speakers that develop for Debian or Ubuntu would be welcome to join me.
What next? I have a couple of interesting offers, one on the Business side of computer hardware manufacturing in Estonia and another one as the manager of a western company's Latvian office. The tricky part in either case would be the immigration process: despite close to 9 years of living right here in Europe, I am still perceived as foreign labor whose hiring must be thoroughly justified (in the case of Estonia, I even need to request an explicit exemption to the immigration quotas), whereas if I had been granted EC Long-Term Resident status or Finnish citizenship, I could just sign the contract and notify the local authorities of my change of address. Instead, 9 years later, I'm stuck at the passport control queues for non-EU citizens with its disgustingly suspicious border guards. Sigh!
Thus, while I indeed have shares in Linutop (yes, we are finally shipping - stay tuned for details) that might eventually turn into gold, I also have close to zero motivation left in my personal life, because lingering European bureaucrazy prevents my talent from reaching its customers - turning all my attempts at having a career into a neverending series of bitter disappointments. Is it any wonder that I'm so adamant about getting out of the Rat Race to return to more gratifying creative activities like composing and photography, then?
Actually, there is a second disappointment about Etch: we are stuck with Evince 0.4.0, at a time where upstream just released 0.7.0 with strong recommendations that everyone upgrade to that. Heck, even Ubuntu has 0.6.0 already. I find the Evince situation way more annoying than the Firefox one, because, from a user's perspective, Evince 0.4.0 fails to display several PDF documents that more recent releases handle just fine.
Yours truly appears this week in computer magazine Arvutimaailm [Estonian] and in the technology insert of newspaper Eesti Ekspress [Estonian], in feature articles about the Linutop. Both articles are signed by Elver Loho of the IT-Neeger team.
Daniel Jacobowitz writes about problems he's having getting proper ALSA support for his desktop computer. Boy, does that sound familiar!
I myself have three computers that are partially or totally unsupported on ALSA:
Partially supported by snd-powermac, except that some controls are inverted and other controls just plain don't work. Probably something as simple as the wrong offset in the bitmap used to select the control on which to act. Bug filed in the ALSA BTS. No reaction.
Anyhow, sound support for Apple products is being migrated to snd-aop, which was written from scratch using a modular architecture that enables easily adding support for new sound chips. Nice idea, except that no effort is being put towards supporting pre-G4 products...
Back when Ubuntu appeared on the Linux scene, Mark Shuttleworth agreed to pay a 500$ bounty to whoever made ALSA work on all PowerMac variants. Nobody took it, as far as I know. Those who are interested in taking on this challenge might wanna check my wiki page for our summary of the ALSA PowerMac situation.
Recently supported, thanks to someone on the PA-RISC Linux mailing list having a go at coding it. Never merged by upstream ALSA, though. It's been a while since I gave it a try but, last time, only output worked.
There is no snd-scx200 driver at all. Bug filed in the ALSA BTS. The issue of sound support on early Geode products was also mentioned on the Linux kernel mailing list, during the big OSS cleanup discussion so, at least, some people are aware of the issue.
It so happens that the Geode SC is used in a number of popular embedded boards, such as the Soekris and yet, nobody tried porting the old OSS drivers available on AMD's Geode driver source page. Odd.
As we can see, there's plenty of ALSA support remaining to be coded, ported or cleaned up, and some people are willing to pay for the results. Who's gonna grab it?
I promised last week that I would be telling more on October 1st, about the mysterious Project X that I had been sporadically blogging about over the last few months. Better late than never, so here's the story:
I met Morocco-based French entrepreneur Laurent Bervas via his blog in January 2006, while looking for countries nearby EU with a promising ICT market, because of growing obstacles in getting work in my field, due to my not being an EU national.
By the time the situation had reached the ridiculous extreme that friends were continuously loosing juicy recruitment bonuses, simply because their employer's HR department had issues with hiring non-EU nationals — despite EU policies that clearly favor foreigners married to an EU national — I figured that I ought to do something creative and see which of the countries bordering with EU might offer interesting opportunities.
Laurent was blogging about all the exciting opportunities he kept on seeing, since he relocated to Morocco to conquer the high-end Real Estate market. One day, he wrote about Morocco becoming a rising figure in the outsourcing business for the Francophone market, which prompted me to contact him. After a couple of weeks of discussion, he cautiously dropped his idea for a niche market he had spotted: a simple Linux-based set top box to surf the Internet. He then asked if I would be interested in developing it and in launching a startup with him to promote it. I gladly accepted.
That's when Laurent decided to introduce me to an old friend of his, Frédéric Baille, who spent several years at Compaq, culminating with a role in the team that launched the iPaq. Fred's help was instrumental in defining the market positioning of our product.
Laurent and I discussed the product features at length over IM and immediately agreed that the OS would be based on Free Software. That left the hardware part unresolved... or so did Laurent think.
Years ago, I briefly worked in the management team of an Estonian data security startup. This gave me the opportunity to network with several movers and shakers of the Estonian ICT. Among them was a promising ASIC and industrial design startup that produced really cool Thin Client hardware. The answer to our hardware needs, I figured, could easily be fulfilled by them.
My idea of contracting the hardware to Estonia was initially received with skepticism, so I made a simple proposal: why don't we hold the inception meeting in Tallinn and meet with the potential hardware supplier at the same time? Fred and Laurent accepted, so we spent part of April 2006 there, with me acting both as their friendly tourist guide and as the middleman for the introductions.
Fred and Laurent immediately fell in love with Tallinn's designer bars, omnipresent WiFi network and blond-a-plenty. Most of all, their visit at our potential hardware supplier gave them the answers they needed: Estonia is IT.
The hardware we selected is built around the AMD Geode LX, a nifty Pentium-compatible single-chip computer. It comes equipped with VGA output and four USB 2.0 ports, along with high-quality audio via 1/8" input and output jacks, plus a built-in encryption engine. We added 100baseT Ethernet connectivity (with provision for other networking options) to that, to complete the platform.
For the software part, we used ingredients from Debian's Etch release, focusing on a configuration that would enable easy access to the most popular applications for home and small business use: Instant Messenger (Gaim), Multimedia player (Totem), PDF reader (Evince), Web browser (Firefox), word processor (AbiWord).
The result is this:
The editorial team at IT-Neeger (an A-level Estonian ICT blog) immediately scooped the story [in Estonian] and it produced such a strong response that the traditional Estonian media contacted the IT-Neeger team to write a feature article (stay tuned for details). The local Slashdot, Minut.EE, also reported [in Estonian] on the IT-Neeger article.
Thus, myself and Laurent Bervas spent the better part of this week being blogged, interviewed and podcasted by a variety of Estonian and French publications, which is how I only got around blogging this bit tonight.
Contrary to our competitors, we fully intend on keeping this platform open, so our initial market will actually be Free Software developers with a penchant for innovation. As such, our first 50 units are exclusively intended for interested developers who want to improve upon our initial idea and participate in defining the final product. Inquiries are welcome!
Q: What do you get if you combine
The answer on October 1st 2006.
For a long time, one of my Debian packages, cups-pdf, could not work on Ubuntu, because Ubuntu runs its CUPS daemon using a low-privilege user. I'm happy to report that this issue is finally resolved: following the release of CUPS 1.2, a simple solution was discovered by Andre Klitzing.
However, I only noticed this recently, after registering myself on Launchpad to respond to Ubuntu bugs affecting my Debian packages. Sigh! Perhaps notifying me earlier, via the Debian BTS, would have been a good idea?
Anyhow, after checking that the solution wouldn't break anything on the Debian side, I committed the fix to cups-pdf 2.4.1-2. Enjoy!
My results for this meme:
history | awk '{print $2}' | awk 'BEGIN {FS="|"} {print $1}' | sort | uniq -c | sort -nr | head -10
73 sudo
67 update-web
53 ssh
51 nano
44 ls
25 dir
21 upgrade-daily.sh
17 cd
13 requestsync
13 more
Among the more unusual sightings, update-web is a Bourne shell script to update my APT repository and synchronize my homepage, and upgrade-daily.sh is another one to upgrade my whole cluster in a semi-automated way.
Meanwhile, requestsync is a really neat Python script by Martin Pitt to request the synchronisation of Debian packages into Ubuntu, automatically fetching Debian changelogs and other relevant information to justify the request.
During Debconf5, I voiced the opinion that all Debian packages with a priority Standard or higher, as well as packages with a high Popularity Contest rating and development packages with a priority Optional or Extra that many people rely upon for development work, ought to be team-maintained as a matter of Policy. I justified this by pointing out that Debian's Social Contract puts our users among our key priorities, which implies that keeping the packages that our users value the most in excellent shape ought to be the embodiment of this value, and that the best way of meeting this goal is to enforce a policy of team maintenance and easy NMU for those packages.
While my own packages are rather modest contributions to Debian's pool and few of them rate anywhere close to the top in popularity, I honestly feel like practicing what I preach is the best example I could give. Another reason is that several business ideas that I'm working on are monopolizing my time, which means that I'm not as available as I would like to be to participate in Debian. I however see this situation as only temporary, which is why I have not orphaned my packages. Instead, I am soliciting Debian Developers to come as co-maintainers. Interested parties should check my QA page for packages that might interest them and contact me for details. Thanks!
PS: in case this wasn't obvious, the whole point of enforcing team maintenance and allowing NMU on non-RC but annoying bugs is to avoid unresponsive lone wolf maintainers who let tons of bug reports pile up on their packages without fixing them or even acknowledging them. Ubuntu avoids this by allowing anyone on their development team to chip in, while Debian still clings on to each maintainer's absolute reign over their packages, which leaves users of neglected packages completely in the dust.