practical filesystem design

In November 1998 there was a book released about file system design taking the Be File System as the central example.

“This is the new guide to the design and implementation of file systems in general, and the Be File System (BFS) in particular. This book covers all topics related to file systems, going into considerable depth where traditional operating systems books often stop. Advanced topics are covered in detail such as journaling, attributes, indexing and query processing. Built from scratch as a modern 64 bit, journaled file system, BFS is the primary file system for the Be Operating System (BeOS), which was designed for high performance multimedia applications.

You do not have to be a kernel architect or file system engineer to use Practical File System Design. Neither do you have to be a BeOS developer or user. Only basic knowledge of C is required. If you have ever wondered about how file systems work, how to implement one, or want to learn more about the Be File System, this book is all you will need.”

If you’re interested in the matter I definitely recommend reading it – it’s available for free in PDF format and will help to understand what those file system patterns are all about – even in terms of things we still haven’t gotten from our ‘modern filesystems’ today.

Source 1:

History Lessons: Trial of the Major War Criminals before the International Military Tribunal in Nuremberg

History was one of my favourite classes at school – I liked it so much that I even wrote one of my final examinations at the A levels in history. I like to know how stuff happened and I like to know what people got from it.

Being a german there’s a lot of history in the last 100 years guiding the interest. You can imagine that the darkest parts of those 100 years are the first and the second world war. Thankfully my generation never had to suffer through such a terrifiying time.

So for the equally interested reader of this article I have good news. In times of the internet we get access to documents that were previously hard or expensive (or both) to get. Like the original documents of the so called Nuremberg Trial – the Trial of the Major War Criminals before the International Military Tribunal in Nuremberg.

You can get them in english, over 16.000 pages of PDFs, packed into 42 PDF files. Or in the official translation in German on

That will keep me reading for a while – but there’s even more. With the progression of scan projects more and more original sources are becoming available for everyone.

Source 1:
Source 2:
Source 3:
Source 4:

ELV MAX! Cube and the Solar-log 500 – state of the reverse engineering and h.a.c.s.

It’s been some weeks since I wrote a status update on the ELV MAX! cube protocol reverse engineering and integration into my own home automation project called h.a.c.s..

So first of all I want to give a short overview over what has been achieved so far:

  • I wrote a C# library, highly influenced by a PHP implementation from the domotica forum, which allows you to continuesly get status information from the ELV MAX! cube with current (1.3.6) firmware. It is tested so far with a fairly big set-up for the ELV MAX! cube (see below)
  • I was able to integrate that library into my own home automation project called h.a.c.s. – There the ELV MAX! cube is just another device, alongside a EzControl XS1 and a SolarLog 500. The cube is monitored using my library and diff-sets as well as status information are stored automatically with the h.a.c.s. built-in mechanisms. In fact you can access for example the window shutter contact information just like you would with any other door contact in the EzControl XS1.
  • You can use events coming from the ELV MAX! cube to create new events – how about switching off/on devices when opening/closing windows?
  • Every bit of information from all integrated sensor monitoring and actor handling devices come together in h.a.c.s.

I started the reverse engineering with just one shutter contact and one thermostat. After all my test were successful I went for the big package and ordered some more sensors. This is how the setup is currently configured:

ELV MAX! set-up

I’ve learned a lot of interesting things about the ELV MAX! cube hardware and software. One is that you need to be ready for surprised. The documentation of the cube tells you the following:

Did you spot the funny fact? 50 devices – we’re well below that limit. 10 rooms – holy big mansion batman! We’re well over that. How is that possible? Well take it as a fact – you can create more than 10 rooms. And that is very handy. I’ve created 13 rooms and there are probably more to come because those shutter contacts are quite cheap and can be used for various other home automation sensory games. The tool to set-up and pair those sensors just came up with a notice that said “Oh well, you want to create more than 10 rooms? If you’re sure that you want that we allow you to, but hey, don’t blame us!”. Cool move ELV! – As of now I haven’t found any downside of having more than 10 rooms.

All my efforts started with firmware version 1.3.5. This firmware seemed to have some severe memory leaks – because just by retrieving the current configuration information every 10 seconds the device would stop communicating after more then 48 hours. Only a reboot could revive it – sometimes amnesia set in which led to a house roundtrip for me.

With some changes in the library (like keeping the connection open as long as possible) and a new firmware version 1.3.6. the cube was way more cooperative and hasn’t crashed for about 1 month now (with 10 seconds update times).

So what does my library do? It is designed to run in it’s own thread. When it’s started it opens a connection to the cube and retrieves the current status and configuration information. Those informations are stored in an object called “House”. This house consists of multiple rooms – and those rooms are filled with window shutter contacts and thermostats. All information related to those different intances are stored along with them. The integration into h.a.c.s. allows the library to generate sensor and actor events (like when a temperature changes, a window opens/closes) which are passed back to h.a.c.s. and handled in the big event loop there.

With all that ELV MAX! cube data I wanted to plug a quite nice tool that I am using in the iPhone and the iPad. It’s called “Moni4home” and it allows you to control the EzControl XS1 directly. Because it’s only accessing the EzControl XS1 I used h.a.c.s. to “inject” additional sensor data into the standard EzControl XS1 data. So basically data flow is like this: iPad app accesses h.a.c.s. which acts as a proxy. h.a.c.s. retrieves the EzControl XS1 sensor and actor data and injects additional virtual sensors like those from the ELV MAX! cube. h.a.c.s. then sends that beefed up data to the iPad app. Voilá!

After the successful integration of the ELV MAX! cube I’ve started to work on the next bit of networking home automation equipment in my house – a solar panel data logger called “Solar-Log 500”. This device monitors two solar power inverters and stores the sensory data.

Solar-Log 500 built-in statistics page

“Funny” story first: this device has the same problem like the ELV MAX! cube. When you start to poll it every 10 seconds (or less) it just stops operating after about 20 hours. Bear in mind: In case of the Solar-Log I just http-get a page that looks like this in the browser:

And by doing so every 10 seconds the device stops working. I am using the current firmware – so one workaround for that issue is to planable reboot the Solar-Log at a time when there is no sun and therefore nothing to log or monitor.

Beside that it’s a fairly easy process: Get that information, log it. Done.

that’s how the console output of h.a.c.s. looks like with all sensors and devices active (Mozilla+Wilma are the two aquaria :-))

So there you have it – h.a.c.s. interfacing with three different devices and roughly 100 sensors and actors over 434mhz/868mhz, wireless and wired network. There’s still more to come!

A lot of people seem to dive into home automation these days. Apparently Andreas is also at the point of starting his own home automation project. Good to know that he also is using the EzControl XS1 and in the future maybe even the ELV MAX! cube. Party on Andreas!

Source 1: ELV MAX! cube progress
Source 2: Reverse Engineering the ELV MAX! cube protocol
Source 3: ELV MAX! cube – home automation for the heating
Source 4:
Source 5: h.a.c.s. sourcecode
Source 6:
Source 7:

second Tokyo Trip 2012 – Rakuten Technology Conference 2012

This October I had the pleasure to fly to Tokyo for the second time in 2012.

The development unit of Rakuten Japan was hosting the 7th Rakuten Technology Conference in Rakuten Tower 1 in Tokyo.

The schedule was packed with up to 6 tracks in parallel. From research to grass-roots-development a lot of interesting topics.

Source 1:
Source 2: Recorded Lectures

open source audio codecs getting better

Some weeks ago I heard about a new audio codec which is being developed as open source – very similar to vorbis – the previous open source approach to audio codecs.

This time it seems that they’ve got some standardization into the play so it might be more successful than vorbis was.

“Opus is a totally open, royalty-free, highly versatile audio codec. Opus is unmatched for interactive speech and music transmission over the Internet, but also intended for storage and streaming applications. It is standardized by the Internet Engineering Task Force (IETF) as RFC 6716 which incorporated technology from Skype’s SILK codec and Xiph.Org’s CELT codec.”

Source 1:
Source 2:
Source 3:

Learn to code

Knowing how to deal with those personal computers is getting more important by the day. Not everybody needs to know how to write code – but since writing code and making those machines do what you want them to do isn’t as hard as it used to be it’s worth the try!

On the mission to learn to code this page is probably very interesting for anyone wanting to learn:

Source 1:!/exercises/0

generate C# classes from JSON data

It’s a common use case: you’ve got some JSON formatted data and you want to interface with it using your favourite programming language C#. You can write the appropriate classes yourself, or you could use the fabulous json2csharp helper page.

Source 1:
Source 2:
Source 3:

a font for number people

OpenType is a font format which I personally might have underestimated in the past. Well you know – fonts and stuff. This all seemed not too interesting up until now. Now that changed dramatically when a font came to my attention which can be used for various purposes and as a font does not resemble the normal numbers and characters scheme. But what can it be used then if not to type numbers and characters?

Well. What about typing graphs?

Everything in the above image is generated by a font… like in your Word-processor (if it uses that font)

“Designed by Travis Kochel, FF Chartwell is a fantastic typeface for creating simple graphs. Driven by the frustration of creating graphs within design applications (primarily Adobe Creative Suite) and inspired by typefaces such as FF Beowolf and ­­FF PicLig, Travis saw an opportunity to take advantage of OpenType technology to simplify the process.

Using OpenType features, simple strings of numbers are automatically transformed into charts. The visualized data remains editable, allowing for hassle-free updates and styling.”

Source 1:

baking with the PI

Do you know what happens during the push of the power button and typing your log-in information inside of your computer? No? You should. At least from a software side. Not that it is necessary to use a computer. But in order to understand what this wonderful machine does and why.

For those teaching and learning purposes the Raspberry Pi is a perfect device. It’s cheap and now there is a course you can take online which shows you – starting from the very beginning – how to get the device up and running and how to make it do what you like. And that’s without installing an operating system. You are about to write your very own.

“This website is here to guide you through the process of developing very basic operating systems on the Raspberry Pi! This website is aimed at people aged 16 and upwards, although younger readers may still find some of it accessible, particularly with assistance. More lessons may be added to this course in time.”


Entwickler in Hamburg – die Developer Conference 2012 in Hamburg

Der Freitag der vergangenen Woche begann sehr sehr früh. Es ging nach Nürnberg um den Flug nach Hamburg zu erwischen. Erstaunlich wie günstig die heutzutage sind: der Flug nach Hamburg (50 Minuten in der Luft) sollte nur 10 Euro teurer als die Zugfahrt zurück (4 Stunden auf Schienen) sein…

Jedenfalls war es ein schön kurzer Flug und schwupps stand ich vor der Otto Versand Zentrale in Hamburg… Es war Zeit für die…

… Developer Conference Hamburg 2012.

Es war meine erste DevCon-HH und dementsprechend kann ich keine Vergleiche zum letzten Jahr ziehen. Die Räumlichkeiten – direkt bei Otto – waren jedenfalls sehr ordentlich aufgebaut, alles sehr bequem. Kurze Wege zwischen Kaffee und Vortragsstuhl. Die 2 der Vortragssäle waren leider nur über den Hauptsaal zu erreichen. Was ein-zweimal dazu führte dass Vortrage in den kleineren Sälen bereits beendet waren und die Menschenmengen durch den Hauptsaal Richtung Kaffee strömten während die Zuhörer im Hauptsaal noch versuchten zuzuhören. Hier mal im Bild erklärt: Rechts der große Hauptsaal und Links ein kleinerer Vortragssaal. Ich stand beim fotografieren direkt im Türrahmen.

Es ging für mich mit zwei sehr guten und interessanten Vorträgen los. Die Keynote des ersten Tages gibt es mittlerweile auch, wie es sich gehört, auf Slideshare:

Insgesamt war die Qualität der Vorträge sehr hoch. Ich fand die Mischung zwischen harten und soften Themen rund um die Software-Entwicklung mehr als gelungen und sicherlich werde ich versuchen nächstes Jahr wieder zu kommen.


Source 1:

Congratulations Mr. Heil!

I am very pleased to congratulate a good man and ex-colleague for his accomplished mission: He handed in his PhD thesis with the Subject (german): “Anwendungsentwicklun für intelligente Umgebungen im Web Engineering“:

“This book describes a holistic approach to develop complex software systems based on the WebComposition process model. It shows how to integrate soft- and hardware components in a cost efficient and effective way using  Web technologies and the Semantic Web. The WebComposition Concurrency System, a formal language to predict system dependencies and conflicts, allows efficient planing and monitoring of the development and operation process of the overall system.”

I had the pleasure to work with Andreas on several occasions. One that I remember with the strongest feelings is a 2 1/2 day around-the-clock hack-a-thon at Microsoft Research. We got it working back then!

For the last 8 years I am constantly trying to get him interested and convinced to work on things directly or remotely connected to some of the stuff I do – but up until now luck wasn’t on my side. Maybe someday :-)

My sincere compliments on achieving his goal on this. Congratulations!

Source 1:
Source 2:
Source 3: