Tuesday, September 20. 2005
Wen kann man haftbar machen, wenn ohne Warnung der Strom im Haus abgeschaltet wird, und dadurch aus irgendeinem Grund die Grafikkarte eines Rechners stirbt? Zuletzt passiert bei mir, heute, ca. 8:00. Und ich bin sauer.
Sunday, September 18. 2005
As I mentioned yesterday, I'm working on wrapper around Unix sockets. Well, just in case somebody else also wants to use the __gnu_cxx::stdio_filebuf GNU extension for marrying file descriptors or C FILE*s with C++ iostreams, one word of warning: don't try to create one iostream for both directions. Instead, create everything separated, from the FILE* to the stdio_filebuf to the istream and ostream, otherwise one of the two channels (in my case the out channel) gets messed up. Here's a little code snippet to demonstrate that:
FILE * ifh = fdopen(fd,"r");
FILE * ofh = fdopen(fd,"w");
__gnu_cxx::stdio_filebuf * ibuf = new __gnu_cxx::stdio_filebuf(ifh,std::ios_base::in,1);
__gnu_cxx::stdio_filebuf * obuf = new __gnu_cxx::stdio_filebuf(ofh,std::ios_base::out,1);
std::istream * conn_istream = new std::istream(ibuf);
std::ostream * conn_ostream = new std::ostream(obuf);
Why the extra wrapping with FILE *, even while stdio_filebuf provides a constructor where file descriptors can be put in? Because I made the experience that buffering doesn't quite work as you expect. You need to provide a buffer size, and it only seems to return data to e.g. getline() when the buffer is completely filled. This leads to weird behaviour that getline(istream,strobj) leads to strobj == "ASDF" when the buffer size was 4 and the input was actually "ASDFQWERT\n". So, that's why I'm putting FILE * in between, as it gives me the behaviour that I expect (see principle of least surprise).
Saturday, September 17. 2005
In this blog, I already wrote about developing a new content management system. Well, after doing some research, I concluded that, in order to get a reasonable performance, I need to develop a high-performance servlet container for Ruby (think of Tomcat for Ruby).
I evaluated a few available ready-to-use HTTP server stacks, but I liked none of them (I was always forced to use threads), so I decided to develop my own one, in C++. Currently, I'm at an early level, all I've got is a nice and easy-to-use wrapper around Unix sockets, providing only the socket functionality that I need. But a nice feature is that it enables me to do the network communication via C++ iostreams. Yes, that's right, no more low-level read()/write()/send()/recv()/whatever.
Tuesday, September 6. 2005
Tja, seit gestern ist es in den Nachrichten, nämlich dass es wesentlich weniger Tschernobyl-Tote gegeben hat als anfangs befürchtet. Wie diese Zahlen zustande kommen: ich weiß es nicht. Was ich jedoch dazu schildern kann, ist, was sich nach Tschernobyl im Kollegenkreis meines Vaters abgespielt hat:
Eine (damaliger) Arbeitskollege und guter Freund meines Vaters war damals, relativ kurz nach der Tschernobyl-Explosion, bei einem Manöver des österreichischen Bundesheers dabei, und kam dabei voll in den radioaktiven Niederschlag. Relativ kurze Zeit danach (genauere Zeiten weiss ich leider nicht mehr, mir wurde das nur vor einiger Zeit von meinen Eltern geschildert) dieser ansonsten vollkommen gesunde Mann schwerstens an Krebs (Hodenkrebs war es, wenn ich mich recht erinnere), und Chemotherapie sowie Bestrahlung zeigten keine Wirkung. Schließlich ist er am Krebs gestorben.
Seine Hinterbliebenen, im speziellen seine Frau, haben dann versucht, vor Gericht Schmerzensgeld zu erstreiten, es ging da nämlich auch um eine eventuelle Mitschuld der Republik Österreich durch unterlassene Information über die bevorstehende Gefahr. Das scheiterte letztendlich, und zwar meiner Information nach, weil das Gericht keinen kausalen Zusammenhang zwischen Tschernobyl und der Krebserkrankung sah. Ich nehme an, dass dieser Todesfall wohl nicht zu den Tschernobyl-Todesfällen gezählt wurde, immerhin wurde er schon nicht von den österreichischen Gerichten als solcher anerkannt.
Der gesamte Bericht ist übrigens hier einsehbar. Als besonders zynisch ist mir dabei folgender Satz aufgefallen:
It should be noted that the average doses received by residents of the territories contaminated by Chernobyl fallout are generally lower than those received by people who live in well known areas of high natural background radiation in India, Iran, Brazil and China.
Hmm, wo hab ich sowas ähnliches nur gelesen? Das muss wohl in Wallraff's "Ganz unten." gewesen sein, als der Subunternehmer den illegalen Arbeitern erzählt hat, die Strahlung im AKW wäre harmlos, "so wie 2 Wochen Strandurlaub".
Monday, September 5. 2005
This is definitely an acceptable speed (within the ISP's network):
21:29:04 (475.55 KB/s) - `linux-2.4.19.tar.gz' saved [32219641]
It has taken a very long time, but finally, I have cable TV and internet in my flat! The technicians came today at 7:15 am, the worked for almost an hour, and then everything was done, and I could just plug in the RJ-45 cable into my notebook, switch on DHCP, and be online. Same with TV: I only had to start the automatic channel search, and after a minute, I had all 36 TV stations available.
Later, I have to do some performance tests, how fast everything really is, whether i really got my 4 mbit down/1 mbit up. And I can finally move my servers from my old place to the new flat.
Sunday, August 28. 2005
In June, I published a few food photos of stuff that I ate at that time.
This time, I will present two pictures of wonderful dishes. The first one is Ramen, some kind of Japanese "noodle soup" (although it's an own kind of dish in Japan cuisine), very spicy, and easy to prepare because there's cheap Instant Ramen available:
The other dish is also Asian, also noodly, but not spicy at all. It's called Udon, also available as instant food. The one I bought was obviously directly from Korea, with the importing company putting own preparation descriptions (in German) on the packages, as the original package only has a Korean and an English version of it on it.
Wednesday, August 24. 2005
Wie manche schon mitgekriegt haben, ist nach dem Inn jetzt auch die Donau dran mit Überflutung, nachdem der Inn ca. 2/3 des Wassers der Donau nach Passau liefert. Dass das passiert, war schon heute in der früh abzusehen, da fehlten nur noch ca. 40 bis 50 cm, bis die Donau "übergeht".
Was ich aber eigentlich erwähnen wollte: ich wurde schon angerufen, ob es mit den ganzen Absperrungen etc. in Linz-Urfahr (der Teil von Linz nördlich der Donau, wo ich wohne, für alle nicht-Linzer) überhaupt möglich ist, meine Wohnung zu erreichen. Zur Beruhigung: ja, es ist möglich. Wenn nicht, hätte ich sowieso ein anderes Problem, dann könnte ich nämlich heute abend nicht nachhause. Aber soweit wird es nicht kommen, denn sowohl der Hauptplatz als auch die Nibelungenbrücke liegen mindestens (grob geschätzt) 5 Meter über dem Niveau der Donau, immerhin ist die Donau für die Binnenschifffahrt freigegeben, also müssen da auch Schiffe unten durch, und da viel Abstand sein (naiv formuliert), im Gegensatz zu z.B. anderswo.
Update: beim Suchen eines aktuellen Pegels der Donau bin ich auf diese Galerie vom Hochwasser 2002 gestoßen. Hier kann man sehr schön sehen, was die Extrema in Sachen Wasserstand waren, und wieviel Spielraum hier noch vorhanden war.
Update 2: beim weiteren Stöbern bin ich auf DoRIS, das "Donau River Information System" gestoßen. Von Linz gibt es anscheinend keine Messwerte, aber da Linz zwischen Wilhering und Mauthausen liegt, sind diese Werte am interessantesten.
Not me, fortunately, but the construction workers that currently refurbish the house next to my flat. The view in the morning was really amazing: one construction worker tried to fire up one of those "flame throwers" you use to melt plastics or foam against flat surfaces. For firing it, a big gas bottle was attached, and he first turned the flame thrower to his face, turned on the gas, and then used a lighter to ignite the flame. Well, the flame was a little bit bigger than expected, so he got frightened, dropped it, and the flame went out. He tried it a second time, but it failed.
That was the time when he started to open and close several valves apparently at random, trying it over and over again, even opening something that looked like a "test valve" where he was able to ignite a test flame two times, with the second time having big problems blowing it out again. After a few more tries, he called for another construction worker, who was also clueless, and so they played with the valves a little bit, and even managed to ignite the flame thrower again for a few seconds. I thought that a huge gas explosion was imminent, and so I decided to better leave for the bus than to watch this fiasco any longer.
Friday, August 19. 2005
I just submitted a lecture to 22C3. Contrary to my contribution to the CCC Camp 2003 and the 20C3, it is only computer-related in the way that computer programs are used as tools.
I named the lecture "Military intelligence for terrorists(tm)", and is supposed to show how easy it is for terrorists to gather and correlate relevant intelligence information using only a few free tools. The topic itself it inspired by this blog entry that I posted some while ago, where I successfully identified the location of a nuclear power plant in the US, including satellite photos and stuff.
I hope that the lecture will be accepted, because I have a few pretty interesting things to show, where you will really ask yourself why the US government hasn't censored that yet (although that's futile...).
Monday, August 15. 2005
Although the last week was quity stressy, I found some time to extend my proposal on developing a new CMS. You can find it here. Comments are welcome.
What has changed since the first proposal? Well, I wrote more about the proposed programmability, including eRuby example code of possible use cases, including the XML integration (people would also like to store structured data, after all...). I also wrote something about logically separating the different parts into so-called "facilities" each of which has a different responsibility, including an (incomplete) example XML-RPC interface how it might look like.
Tuesday, August 9. 2005
...or "why wasting several thousand euros a day on external developers and consultants is ok, but spending money on soft, triple- or quadruple-layered toilet paper does not seem to be ok."
I really hate using toilet paper that is at most double-layered, it makes me feel uncomfortable, and it definitely reduces my productivity because of that. Dear managers, next time you pay EUR 2000 per day for an external project manager, think about buying some high-quality toilet paper to make your dump-taking developer crowd happy. It shouldn't cost a fortune.
Monday, August 8. 2005
So singt rappt zumindest Bushido. das will ich einmal sehen...
Sunday, August 7. 2005
After I saw the Stellent Content Server at a customer, I thought about developing my own programmable CMS, because I liked some of the ideas. After thinking about all that for some time, I'm currently quite clear about how the overall principles and the design should look. I wrote down an overview about the general concepts, which you can find here.
I'm asking everyone for feedback. Please be aware that the explicit goal is not to create yet another easy-to-use CMS, but instead to provide a CMS that can stand up to high demands on programmability, flexibility and scalability, even with huge amounts of data.
Thursday, August 4. 2005
I'm sure some tele.ring employees bit their ass:
(for those people who don't speak German, the second heading means "Money rain for tele.ring employees", while the first one means "No money rain for tele.ring employees")
|