Since the prolific-based 3.5" firewire+usb external HDD box I have is total crap (it often stops and restarts the disk, going readonly and losing data), I've been looking for a different drive.

Unfortunately, I havn't been successful so far. The cheap computer stores around here all seem to only have drives with (probabliy newer revisions of) this chipset... and I'd really like to avoid buying another prolific crap.

I updated yesterday to the latest firmware for the PL3507, and I also checked that the drive I put in the box doesn't have any errors. Didn't help. The drive still is very unreliable, in both firewire and USB mode.

Unfortunately, my laptop only has USB 1, otherwise I'd buy an USB-only box which supposedly uses a different chipset.

Watch this flash movie on youtube. Cool. I wish I had a camera useable for such things...

Balancing point homepage with some still images and link to ifilm (but I can't play the movie there... ifilm seems to be crap...)

Or you can just use a search engine to find a download link such as on Filefront.

The latest upgrade for localepurge actually killed all locales on my system. Here's a one-liner I used to restore them:

apt-get --reinstall install `dpkg -S /usr/share/locale/de | sed -e 's/:.*$//;s/,//g'`

Note that this isn't necessarily a complete list. But it should cover most.

Wenn man am Münchner Marienplatz von der SBahn zur UBahn geht, kommt man an einer Plakatwand nicht vorbei - gigantische Plakate, die im wesentlichen je eine URL enthalten, meistens etwas kurzes wie weg.de und neu.de...

Ich kann mich nicht dran erinnern, dass in den letzten Jahren jemals so massiv für verschiedene Internet-Dienste geworben wurde - oder mit so kurzen Namen.

Insbesondere Internet-Partnerbörsen scheinen momentan massiv Werbung zu machen (drei der fünf Plakate sind von solchen, glaube ich... dann noch eine Reise-Webseite und irgendwas mit Autos IIRC)

I have the impression that OpenOffice is still busy with cleaning up the code inherited (and e.g. porting to GTK2 and such things). Given the screenshots posted in the much respected Office UI blog by Jensen Harris, I think OpenOffice will run into trouble when Microsoft actually releases their new Office.

Microsoft seems to have a really neat and productive UI there - there is only one thing really wrong with it: it's completely different than anything else people are used to (and people have learned to use).

While there will obviously be migration issues, companies will probably delay switching to the new office to save licensing costs and such things, marketing people will be all crazy about the new features in there.

You might remember the research with the aggressive title "PowerPoint makes you stupid"... actually this will become even more true with the new Office, I fear.

It comes with default styles for common diagrams used in presentations, like a flow sequence or a circle sequence. While these aren't particularly smart things, they'll be presented even more prominently (because the gfx is so great), and the actual contents will probably be pushed back even further...

Anyway, these new styles look really hip, so people will be eager to use them. And use them. And use them again. And use them really big (eventually forgetting about the contents behind them)... and OpenOffice can't keep up with that so far.

Maybe OpenOffice needs a huge scale UI contest. Maybe with some abstraction scheme to be able to run different UIs on the same engine. OpenOffice in particular should try to become a technology leader, and not just try to copy Microsoft as much as possible... the UI changes in Impress are said to be pretty similar to what people expected when they think PowerPoint.

There are good reasons to stick with the established UI standards (even when that might mean 'cloning' the UI) since that keeps migration costs low. But I think it's a pity OpenOffice can't show many large technological improvements to put more pressure on Microsoft.

Ah, and yes, I guess many people at OOo are well aware of that, and like so many projects (commercial ones probably even more than FLOSS) they're just short on development resources...

This page uses the motion sensors present in some new harddrives in an interesting way.

Check out the flash/youtube video. It's too funny.

My laptops harddrive doesn't have motion sensors - but it has two very nice keys I use for desktop switching, right next to the cursor keys. Of all key combinations I've used so far for desktop switching this is the one I love most... Unfortunately, this is a very unusual keyboard layout, and I think even IBM/Lenovo doesn't have these keys on newer models any more. A shame.

Since many "office" people love to send "funny" powerpoints around via email (which is stupid, because email is ~40% slower than others) we need to provide them with an application to do this much easier.

Time to create the first pptcast. Or is ist pptfeed? pptlog?

Fortunately I'm not interested in reading them.

Count your sheep is a cute webcomic. You've got to check it out.

One. Dzzzz...

GNOME is all about usability. And it used to take this to a point making many people angry. Still I loved this direction, because it prevented me from wasting too much time on toying around with settings that don't really affect my productivity...

Today I've counted 37 entries in the settings menu. That is more than twice as many as should be there, IMHO... It makes me remember the settings chaos in kcontrol, where I often happen to know that the settings is somewhere and I just can't find where...

Granted, many of them aren't part of the GNOME core, for example two entries are Sun Java crap. Some policy tool I have no idea what it does, and a visually completely different dialog for the plugin. Or the art manager, a ruby application (why does it hide in settings?) to download themes from art.gnome.org.

There are tons of settings I've never even looked at, like the CD database server (probably defaults to freedb.org, and why would I want to change that?)

Yeah, I know... I should probably get some fresh Ubuntu install and look at the menu there, I bet it's only half as big as mine here...

Congratulations to Alex de Landgraaf. His proposal "Debtags, using AI classifiers for automating the tagging of Debian packages" made it into the list of projects that have been accepted into the Google Summer of Code.

There were others with a similar proposal, but Alex' was most convincing. The SELinux projects didn't make it into our slots - a pity. But I hope that we'll get SELinux into etch somehow anyway.

We would have liked to take twice as many projects; but that would mean some other organization gets a slot less. I think we got a good set of projects.

For my birthday party tonight, I baked a quiche lorraine [en.wikipedia.org]. Of french origin, this "cake" is quite popular in Europe. I used my mothers recipie, which is really easy and which I like a lot.

Heute abend ist meine Geburtstagsfeier, und als Snack habe ich gerade einen Quiche Lorraine [de.wikipedia.org] gebacken. Nach dem "Rezept" von meiner Mutter, das ist nicht besonders schwierig (einfacher als das von Chefkoch), und schmeckt mir sehr gut. Eine meiner Leibspeisen.

Quiche lorraine

First of all, you need dough, obviously. In my opinion, whole weat is best. Put some butter on the baking tray, then a thin layer of dough on it. Distribute small pieces of bacon on it along with thin rings of spring onions.

Mix some cream with some eggs, then add grinded cheese. Lots. Distribute over the tray. Then bake for approximately 30 minutes at ~180 C, until the cheese-cream-egg-mixture gets this lovely color.

Aguilegia [en.wikipedia.org] are pretty wildflowers. Unfortunately, some have become an endagered species, so maybe we should be proud about having so many in our garden. But we definitely are happy about their colorful contribution.

Akeleien [de.wikipedia.org] sind wunderschöne Wildblumen, die sich in unserem Garten anscheinend richtig wohl fühlen. Wir haben zahlreiche Sorten hier, vielleicht sogar welche, die als "gefährdet" eingestuft sind?

Aguilegia in passing-through sunlight

Unglaublich... FIFAworldcup Homepage, man beachte die Karte der WM-Spielorte...

München liegt ab jetzt in Garmisch. Und wenn man details sehen will, muss man mit der Maus über Österreich fahren. Hamburg ist auf halbem Weg in Kiel, und Leipzig in Zwickau.

Also liebe FIFA, sooo schwer ist es doch nun auch wieder nicht... habt ihr da einfach mit nem Fußball auf eine Torwand mit aufgemalter Deutschlandkarte geschossen, oder wie?

Der Sch(m)utzbund Deutschland (anklicken lohnt sich nicht) scheint sich auf braune Hass-Propaganda spezialisiert zu haben.

Deren "Slogans" beinhalten "Nein, Horst, du bist nicht Deutschland" - womit sie unseren Bundespräsidenten Horst Köhler angreifen. Wie man dem Bild entnehmen kann, scheinen sie es nicht zu mögen, dass er auch mal eine jüdische Synagoge besucht hat. Oh mein Gott, das habe ich mit der Schule auch... ist meine ganze Klasse (und vermutlich der Rest der Schule, wie wohl halb Bayern) auch nicht Deutschland?

Ebenso soll Gerald Asamoah "nicht Deutschland" sein.

Ich glaube sowohl Horst Köhler als auch Gerald Asamoah tun alleine mehr für Deutschland (und nicht nur für die BRD als Staatssystem) als alle diese rechten Spinner zusammen. Sie leben ein Deutschland ohne Hass, und das ist wesentlich schöner als das was die braunen versprechen versuchen...

Ich glaube diese rechten Spinner haben in Wirklichkeit ein Problem damit, dass sie strohdumm sind, und daher keine besonders rosigen Zukunftsaussichten haben... und ich glaube dass die Bekämpfung der Arbeitslosigkeit auch ein Kampf gegen Rechts ist - beispielsweise sind die Braunen in Gegenden mit einer sehr hohen Arbeitslosigkeit aber sehr niedrigem Ausländeranteil besonders stark, während in Städten mit hohem Ausländeranteil aber gutem Arbeitsmarkt eben praktisch keine Neonazis sind. Die "Braunen" haben doch überhaupt keine Ahnung wovon sie da zu reden versuchen. Sie kennen sozusagen Deutschland gar nicht.

So I'm 26 now. Time flies... I'll be having my final exams starting in three weeks, and my diploma thesis should be done by the end of the year. Then I'm heading out looking for projects to earn money on. ;-)

Well, sounds like I'm looking forward to it - but I don't know. The last few years, I've been very uncertain about what I really want... almost like puberty... ;-) and on overall I've been rather unhappy the last years. I hope this year will become much better.

I'll be having a joined birthday party with a friend who happens to have the same birthday. We'll just go to a bar and have some drinks, though. I've noticed that I've mostly been inviting girls... well, I just happen to have much more close friends that are girls than boys... I never had a hang for beer-drinking and soccer watching I guess, but I like to meeting a sweet girl in a nice cafe... and many of my friends just aren't in munich, so they won't be able to come. Like the technies I work with over the internet.

So I've actually started looking at the new java license now. A couple of people already pointed out that while Java just entered Debian non-free, the license contains a couple of restrictions Debian can't conform to, so it will probably have to go again.

Unless of course Sun finally picks a sane license for Java sometime...

But I'd like to be... I'd love to bring that lame old line:

My sock's got your name on it

Not to be confused with:

My socket has got your name on it

Oh, if you want to play that game some time - instead of smacking people with clean socks, you can also just use something like saying "You're dead". Works just as well for virtually killing people. But is probably less fun. I've never been running around for days trying to smack someone with a sock.

Do you know what the most evil thing is about this variant of the game? You don't have a chance to stop him from trying to assassinate you again and again... in other game variants, you at least have a chance of defeating him instead.

Oh, and maybe Ted Walther misread the game instructions, and the reason he's been smacking people with his toy is that he's trying to assassinate them? Can someone hit him with a cluebat please and explain the rules of the game again?

To enable single-sign-on with an existing windows network, I've been using winbind and kerberos to login users.

Unfortunately, winbind doesn't have functionality to automatically add users to certain groups, e.g. audio, video, plugdev.

But I want to allow local users to access audio and USB drives on these machines.

I just found an easy way to do that:

1. Add auth optional pam_group.so to your /etc/pam.d/gdm file
2. Add *; :0 ;*;Al0000-2400;audio, video, plugdev to your /etc/security/group.conf file

This will give the first user logged in via the GDM login screen access to both USB sticks and audio playback. Using :* this should apply to all local users; a similar line with tty* should work for local ttys.

Note that this applies to any local user, not only to winbind users.

Other groups you might want to add: cdrom, scanner, dip, netdev

FIFA sucks. It's so much about commerce...

Don't buy their stuff. Buy your soccer balls at e.g. Transfair. These were bought at fair prices and not made by exploiting people in third world regions. Check out the 'Wilde Fußballkerle' design. Very sweet. Much better than this annoying worldcup logo just about everywhere.

But back to my original topic. These annoying 'official' worldcup songs. Dear FIFA, how about letting people decide which songs they like? I hate your choice... It makes me think of Titanic! And actually I hope that this happens to FIFA, that they go down like titanic because of some corruption scandal or so.

If you want to compare them, check out Voices from the Fifa World Cup at Amazon (with realplayer samples).

My current favourite is 1. FC Sommer, played by the Bananafishbones (last.fm), an indie band. They're cool. They do a festival and support new indie bands.

If you want to listen to an excerpt, point your browser to the 1. FC Sommer page, there is a realplayer stream available again, klick that audio button.

That song has some great wordplays in there, too bad that you can't translate them in all details into english. The refrain is "Der Sommer gehört dem, der den Weltmeister schafft" which translates to "The summer belongs to who wins the worldcup"; the world play is "Weltmeister schafft" (becomes the world champion) and "Weltmeisterschaft" (Worldcup). And there is also a nice song by Sportfreunde Stiller, who did a whole CD full of soccer songs.

So just ignore FIFA and don't buy their branded stuff.

Apparently Sun has finally relicensed Java in a way that Debian can ship packages in it's non-free section. I guess they've been realizing that people started using other Java implementations or Mono instead, that were easier to install on Linux systems.

A couple of things were wrong with Java before that, making it a pain to install. The Debian Java people had gone a long way to fix that (e.g. by making the nice make-jpkg application that will convert the official Java into a nice Debian package you can install on your systems, but just cannot give anybody else). While installation wasn't hard with that tool anymore (download, run make-kpkg, install package), this still couldn't give you automatic upgrades. The free java implementations had huge benefits here, so does Mono.

Sun Java still has shortcomings with installation, though. First of all, it's in the frowned upon, second-class "non-free" part, secondly you need to agree to it's license on installation, making unattended installation (legally) impossible, I guess. (You can probably just preseed the debconf flag, though)

I've recently been recommending people to use Mono instead of Java if they only have these two choices. Mono apps are supposed to run on Windows as well, look a lot better than Java, are apparently more fun to write and easier to install. Now let's hope no patenting issues will be used by Microsoft to make Mono go away. (I wonder if that is one of the reasons why they absolutely do not support Mono meetings at their Dot-Niet gettogethers, because you could use that in a legal trial as them acknowledging Monos use of their stuff, because overall they are benefiting from Mono being .Niet on Linux and thus being a bit more platform independant...)

So those interested in my remote DNS update CGI (via nsupdate), you can download it from my playground for now. Note that as usualy it comes without any warranty; there might be still security issues in there (but you don't want to allow arbitrary users to access it anyway...) and so on, you know the drill.

The big plus of that app is that it doesn't modify zone files, but just loads the current zone via the DNS protocol (zone transfer), and modifications are sent back to your DNS server via DNS, too (nsupdate, just like dhcpd does). That way it works very well remotely, is quite secure and reliable (no parsing of zone files involved, no reloading of your name server).

While it's MIT-style licensed, I still appreciate improvements such as syntax checks for entered data (nsupdate does a bit, but not much).

Judging from the Google alerts I've been receiving recently, Google is heavily catching up with updating their index. The alerts I receive are approximately a week old pages (Planet Debian changes quickly, the blog posting just "found" by google is already gone for approximately a week, the blog posting is exactly ten days old.

In April, I received around 1 alert every 5 days; now it's one each day.

I think I've found a bug in the UML 2.0 OCL spec.

It's actually a well-known issue for C++, documented e.g. in the C++ FAQ.

It's not so obvious in Java, since Java didn't have Templates for a long time.

The UML 2.0 OCL spec says on page A-22, Definition A.27 (Type Hierarchy)

If t' ≤ t then Set(t') ≤ Set(t)

Any good C++ programmer should know that this doesn't work. From the C++ FAQ "Is a parking-lot-of-Car a kind-of parking-lot-of-Vehicle?":

"A Bag-of-Apple is not a kind-of Bag-of-Fruit." If a Bag-of-Apple could be passed as a Bag-of-Fruit, someone could put a Banana into the Bag, even though it is supposed to only contain Apples!

Note that the converse, Set(t) ≤ Set(t'), also doesn't hold. While you can put Apples into a Bag-of-Fruit just fine, you may still find a Banana in there.

This has an obvious reason: when passing data to the object it can be "downcasted", and when retrieving data it can be "downcasted", too. But it can never be upcasted.

This should occur in OCL, too. A "Set(Apple)" doesn't conform to "Set(Fruit)" by the usual interpretations that I can call "put(Banana)" on any "Set(Fruit)"!

Can anyone enlight me, if this really is a bug in the OCL spec, maybe has already been corrected, or if there is some magic in OCLs definition of "conforms to" or interfaces that solves this problem?

(Yes, I think there might be something in OCL that solves this problem, for example if this hierarchy is only used in condition checking, then in fact all operations are "get" operations... but I've just started reading on OCL, I don't know much about its expressiveness yet...)

[Update: Onne Gorter has some good thoughts on this. He basically notes that this is not a problem if you don't have any references to the "uncasted" list. For example if the cast is a function like in a functional language that actually returns a new set. I'll have to consider this for OCL, but given that OCL and UML are somewhat designed for Java - and Java suffers from this problem, too - I'm not so sure about it.]

Since I couldn't find an appropriate tool, I wrote it myself (to be put somewhere soon): a web frontend to update a bind nameserver without needing to go through the hassle of parsing zonefiles, reloading the nameserver and such.

How it works: it just uses the NSupdate functionality included with Bind. It's public key crypto, and all the code is already in bind and probably audited quite well. And the web frontend can run on the admin web box, properly firewalled off.

With that tool I can easily change the DNS zones on my SELinux boxes, without having to fight with allowing a web server on that box to read and write bind files and reloading the bind service. IMHO this solution is a lot cleaner: just tell the changes via UDP port 53 to the name server and let it do the actual job himself.

The "dynamic" zonefiles - those bind may write to - live in a separate dir (and separate SELinux file context) and with the extra SELinux protection, privilege escalation of bind is really unlikely.

(The web interface is really basic though - it doesn't do many syntax checks. So you should have sufficient DNS knowledge before using it, but thats a good idea anyway. There is a couple of stuff you can break with DNS, you know?)

Actually, all the Linux systems are very much alike. The libs are in /usr/lib, the bigger apps in /usr/bin, only the core system utilities in /bin and so on. Everything has it's place, and with the LSB that is pretty well defined. And it's not just the kernel and a text editor, this applies to most of your system. Including three office packages, a couple web browsers, a dozen IM applications and so on.

Windows however is very fragmented. Well, not really fragmented, because no two Windows systems are alike. On some systems it's "C:\Program Files", others use "C:\Programme", maybe a dozen other translations. I've seen apps using "C:\Progra~1" which works with both of above path names, at least on FAT, due to the old 8+3 filename restriction. So I don't know about NTFS.

This makes supporting Windows a PITA. You can test something on all your Windows boxes, and it just won't work on the first clients box you need to run it on. Happens all the time. Not to mention surprises with actually different Windows versions.

Thats what I love with OpenSource, Linux and Debian. You just "apt-get install lyx" and you end up with a complete, working LaTeX and LyX environment on your system within minutes. Completely hassle free. The amount of "third party software" shipped with Linux is it's biggest strength. Everything is there, and it'll just work and still not really make your system different from anybody elses. Even across distributions. I have an Ubuntu workstation at one institute at the university. The CS computer lab is all OpenSuSE currently. But I don't need to care. It works just like my system. I google how to change to page ordering when printing PDF slides 4-up, and it just works.

I envy you. I would have liked to attend, and I already had to pass last year. I had expected to be in the middle of my finals by now, but it'll actually be another month. But I have most of my dates now, at least.

Also I'm such a cheapskate, flying over the ocean to attend a conference is a bit out of my league. Guess I need to find a sponsor for such things. Just like I could use a build and test machine for my SELinux work and for backporting (I have four pbuilder setups on my laptop), maybe with an UML or Xen virtualization setup for easier and faster testing. Well, I may dream, okay? ;-)

My favourite DebCamp postings so far:
no nude swimming pool use (wear at least an Ipod!) SCNR.

As for the Google Summer Of Code: we've received lots of applications, I think somewhere in the 80s. Some pretty bogus (some don't even have a title), but also very cool ones. A dozen duplicate proposals, too. I think our "favourites" list is around 20, but I don't expect Google will sponsor this many, and we also need enough mentors.

I today witnessed yet another Windows presentation disaster.

Well, the presentation itself was good and well done (using latex-beamer).

But Windows continuously sabotaged the presentation. Every 5 Minutes, Windows announced that it's going to reboot soon, because it has installed some upgrades, and gave the presenter a 5 minute countdown that he could only postpone, but not disable until the presentation was over.

It was really annoying to have this stupid "Hey, I've installed a security upgrade, I'll reboot in 5 minutes to complete that installation" pop up again and again and again during the presentation...

check out this report of this 'nag until the user restarts' functionality in windows

Windows is Nagware. I wonder when Microsoft will add a function that pops up in the middle of any PowerPoint presentation if it thinks your copy is pirated and tell your audience you're a dumbass.

Look at this chart comparing the stock value of Google and Phoenix Sonnenstrom, a German solar power company, with good news every few months.

Yeah, just kidding. Still interesting to see them behave this similar on overall, and some contrary movements where Phoenix bobs up just as much as Google dips, and then meet again.

How much does it says about a company to have higher value growth than Google (in a 2 year period) without being publicly hyped like Google was? I guess it was a very good choice to invest into solar energy.

Because I'm currently seeing a forgotten logout screen, and have witnessed that a dozen of times now, with both GNOME and KDE (and Windows I think, too; but I rarely witness anyone using Windows...):

Please make the logout dialog automatically logout.

People tend to click on "Logout", and forget about the usual "do you really want to logout, or maybe shutdown, or maybe reboot, or maybe ..." dialog pretty much any operating system has.

I've never heard of anyone forgetting to pick "cancel" in that dialog, so I think you could safely default to the "logout" option after a timeout of 60 seconds or so. Just make it not leave computers unprotected (a screensaver will NOT activate!) for a long time just because people forget to logout twice.

When you play this song backwards, do you get "I don't want a friend, I just need a lover"?

That's how I feel every so often. I have so many close dear friends, I'd trade them in 10:1 for my true love...

I'm heading towards my final exams - around one month to go - but I'm caring more about not having to wake up alone than about my finals. :-( I wish I would just be happy with the friends I have. That I go dancing with. That I chat with every day. That I can snuggle against. And then actually prepare for my finals.

iRomance - ich weiss nicht, ob ich das jetzt als "sarkastisch", "witzig", "real" oder "gemein" bezeichnen soll. Jedenfalls: sehenswert, 1a Qualität - und legal: CreativeCommons Lizenz!

Hilarious movie on YouTube: the evolution of dance by Judson Laippy. Excellent performance.

(If you don't like YouTube because of Flash, you can grab an as much proprietary .wmv at cartoonland)

It's worth the 6 minutes of playback time. He doesn't live up to the title - the theme "evolution of dance" is lost quickly, and becomes a general mocking of all famous dance moves, though.

Yesterday, I went to the Munich Bladenight... 20000 people (excellent weather) on a 20km track through the city filling major roads. (picture gallery, though mostly of the gettogether before the start)

I didn't blade much recently - I went to only one Bladenight last year, and that probably was only a 15km track, since I was in the US and Canada until mid August. So it probably was a bad idea to start with this 20km long track... at least my feet think so - they're still aching. Interestingly, it was worst when walking after the bladenight. Everything was okay while blading... And biking was much better than walking afterwards - you don't have your full weight on you feet when biking.

Anyway, if the weather is good next week, I'll be going again. The north route due next week is just 15km.

Wie in der Süddeutschen Zeitung zu lesen war, haben sich auf der Bürgerversammlung 95% gegen den Transrapid ausgesprochen.

Trotzdem soll er gebaut werden. Die Bürgerversammlung sei "nicht repräsentativ" (und unser Oberbürgermeister hat auch keine Ahnung, was die Leute wollen, oder wie?)...

Es ist schrecklich, wie arrogant die Bayerische Staatsregierung hier mal wieder "regiert". Warum wählen die Leute nur immer wieder die Amigo-CSU? Denn das ist die einzige Erklärung, die ich finde: die CSU hat andere Gründe, den Transrapid zu bauen. z.B. weil vom Bund dafür Millionen kommen, die sie so "offiziell" an irgend eine Baufirma leiten können (die natürlich rein zufällig einem Bekannten gehört). Für mich riecht das schon sehr nach Korruption... der Bau der Arroganz-Arena lässt grüßen.

P.S.: jede unserer Trambahnlinien transportiert mehr Fahrgäste als der Transrapid soll. Wie wärs statt dessen mit der Trambahn durch den Englischen Garten? Und die Express-S-Bahn-Lösung, die auch vom OB Ude bevorzugt wird, würde auch schon einen deutlichen Zeitgewinn bringen, allerdings zu viel geringeren Kosten.

I'd just like to point out that using Debian is not sufficient for sending us your application. Apparently some projects were submitted with minimal wording changes to Debian, Python, Postgres, Gentoo and Apache.

If you send us your project, it should have some extra relationship to our project... it should be something our users or our community needs and that is somehow related to Debian. There is nothing wrong with being useable by other projects as well, but the relationship to Debian needs to be a bit more than just "Debian is my favourite distro."

For example the SELinux policy module dependency handling is obviously something all distributions would like to have to ease installation. But when integrated with some nice magic of autoselecting policy modules based on the Debian packages installed: go for it.

I've been asked about details on the Debtags SoC ideas. Here are some mails for you to read:

• Tasksel replacement via debtags?
• Bayesian tagger experiments
• Central database wishlist
• Managing the authorative copy of the DB

Another suggestion for a Google SoC project: semi-automatic removal of unused software.

Installing extra software is very very easy with Debian. However, you'll just fill up your harddrive, and actually keep lots of software around you actually do not use.

It would be nice to have a tool (maybe functioning similar to popcon?) that monitors (e.g. via atime) which applications you actually use, and suggest others for deinstallation. Of course handling dependencies right (i.e. if you chose to install the GNOME meta package, it shouldn't suggest to remove standard parts of GNOME even if you don't use them).

Well, I don't know what others think, but I would like to see that done, and I can imagine this being an interesting SoC project. So go for it!

Google alerts, which are email notification on new sites being added to the google index for a certain query, broke these days. Seems like they just "upgraded" their software.

The alerts are still sent, however they list web pages that don't match the query. For example, my alert on "layered-subversion svn" just turned up some generic hits with "svn" "subversion" and "layered" in them.

A pity. I really liked this function, but it's not too useful anymore now. I hope they fix that again.

[Update May/5: todays alert mails had only correct results.]

It's also a shame that there are some restrictions with google on combining e.g. site: and link: matches. Especially with negative prefix... For example I'd like to see all links to my homepage, but not those from my blog. You can do that with Altavista, MSN search, for example. Exalead also gives much more valueable search results than Google...

I've been investigating alternative system inits from time to time the last few years. With the advent of bootchart, there was quite a hype around improving boot times with parallelization and such. Recently this topic has manifested itself in the initscripts-ng group on alioth, and some to-be Google Summer of Code projects.

I've been using a modified version of minit for a long time, just lately I was to lazy and switched back to regular init. Since I use suspend2, I don't really reboot my system often anyway.

However, minit has more features than being parallel and small. It offers service monitoring and respawning. Which brings me to something missing totally from current linux init systems: service state tracking.

Right now, when you start a service, e.g. the web server, you pretty much hope it will never die. Agreed, apache does a good job here. Others may or may not. Mysql uses a really hackish shell script to respawn. And then there are of course tools like monit that check your services and respawn them when needed.

There are a couple of things wrong with starting services via SysVInit style scripts from the shell. For example, in order to start a service you need root rights. And then this will "leak" file descriptors from your shell to the daemon (thats why some apps need to be started with "nohup"). Which also causes extra errors with SELinux, for example. And this is just one of the differences from services being started by manually running their init script from the script being run by runlevel changes.

Therefore, I think we should move to a new init that can offer full state tracking for services, and is actually used for all services (and not just initial start and getty respawning...)

Here's a fairly extensive state diagram I came up with:

Some features I'd like to have in a new init:

• state tracking with interim states
• dependency management with virtual services (e.g. webserver)
• service monitoring and respawning
• control fifos for starting/stopping services through init
• init monitoring socket for e.g. DBUS connector and service failure notifications

Sorry, no comments on my blog, intentionally. Please use the initscripts-ng mailing list for discussion, thats what the list is for!

Sometime tonight, your clock (date "+%H:%m:%S %d.%m.%y") will read

01:02:03 04.05.06

"Irgendwann" heute Nacht zeigte die Uhr den schönen Wert

01:02:03 04.05.06

If you are interested in doing a SELinux related project during the Google Summer of Code, you can submit it to Debian, if it's somehow Debian related.

It would be nice to get some extra tools (e.g. dependency handling, module autodetection) via the Google Summer of Code.

If you're interested in the Google Summer of Code: the application phase is open. Submit your proposals now.

Some notes:

• Make sure to pick an appropriate organization. We've received some requests at Debian that would fit much better to X.org or GNOME
• Wrap your text to ~76 chars. Some of the proposals were a PITA to read because they had too long lines, and since the proposal is in a scrolling textbox I always have to scroll down to the end of the box, then scroll right, then scroll back up to the line I was just reading. (Yes, this is an UI issue with Google, but believe me, you'll benefit from a well written and easy to read proposal, because more people are going to read through it and eventually rate you.

So far, I havn't seen an application to any of my suggested projects (LaySVN, SELinux, Debtags AI Tagger, Debtags database). So if you're interested, go for it!