About Reginald

I grew up in a typical middle class household in Graz, Austria. My mother was a doctor and my father taught Chemistry at University. I am the middle-child with an older sister and a younger brother.

My dad always was very interested in technology and got me into electronics and computers early on. Already in my school years, when most families didn’t even own a computer yet, I learned to program (initially on a Sharp MZ 700 and later on Atari ST) and started earning a few Schillings here and there by selling small utility programs I wrote and published in shareware magazines.

During University, I started experimenting with Linux, mostly because I wanted to build an interactive website with CGI scripts. For this, I needed to set up a web server, which was best done on Linux. 

This was in 1997. My experience prior to that was programming on Atari ST and later on Windows 3.11 and Windows 95. I was always drawn to GUI programming, mostly in Pascal which was easier to learn and understand than C++ during my highschool times.

After spending a lot of time getting Linux set up and running, coming from Atari GEM and Windows, the lack of a proper Desktop Environment initially irritated me, but also attracted my interest to do something about it. Through an article in the computer magazine c’t, I learned about a new project in very early stages (pre-1.0 alpha). It was called the K Desktop Environment with the goal to build a desktop environment for Linux and Unix. So I went ahead and installed a very early version of KDE (which meant downloading loads of sources and libraries and compiling everything yourself). This alone was quite an experiment and experience. But I was determined to get it all up and running, and soon I started playing around with KDE. 

This got me into C++ programming and GUI programming with Qt, which KDE is based on. I soon started to contribute small patches to KDE and got involved into the community. Back then, working remotely, distributed over the whole world, was not common at all. Besides learning a lot about software development, the spirit of the KDE project, the way of communicating and collaborating and being part of an open source project very much influenced me and how I later built a software company.

After initial small patches here and there, I became one of the driving forces behind KOffice, a rather bold attempt to create an alternative to Microsoft Office. In 1999, KOffice was even cited in testimony in the United States v. Microsoft antitrust trial by then-Microsoft executive Paul Maritz as evidence of competition in the operating system and office suite arena.

At this point in time, the KDE project, and open source projects in general, were mostly driven by students and freetime programmers without any commercial motivation. To some extent, it all was rather idealistic and the common goal was to build a free alternative to the Microsoft world, which was considered the bad evil among the Linux people.

Slowly companies, mostly Linux distributions who sold CDs with pre-packaged Linux setups to make it easier to install, started to work with open source projects by paying developers to contribute to projects which in turn made Linux more attractive and usable and allowed them to sell Linux CDs to more people. KDE was one of the projects which Linux distributions started to support.

In those times I started to work as a freelancer for a German entrepreneur living in the US, who envisioned building a Linux distribution which would be easy to install and usable even by computer novices. One needs to understand that, at these times, usually you needed to do a lot of fiddling around with scripts to get Linux up and running. There was no plug’n’play and one click installation of applications. The idea to configure everything via a GUI and offer applications to be installed with a single click from a central repository, was very innovative but also very challenging to follow through with at this time. It would take about one decade for this idea to become reality in the incarnations of what we know today as App Stores.

While the gig with this company only lasted a few months, as a student “hacker”, it was great to earn a decent amount of money while working on very interesting projects in the Linux space. It also allowed me to travel to New York City, where the startup was located, and experience this city and the different culture for a few weeks.

Through the involvement in the KDE project, the makers of Qt, the Norwegian software company called Trolltech, contacted me when they were looking for some programmers to help with a project they got. The project involved porting the Opera Web Browser from Windows to Qt to make it available across Windows, Linux, Unix and MacOS. The Opera Browser was a rather successful commercial (yes, this was possible in those days!) web browser of the early WWW days built by a bunch of Norwegians.

Through that project, staying two weeks in the dark and cold winter in Oslo, I got to know the folks at Trolltech very well. We spent most of the days and nights in the office working. While we mostly worked on the Opera project, we also spent a lot of time discussing the design and architecture of Qt 3 and visions of cross platform GUI development.

When I returned to Austria after two weeks in Oslo, just a few days later I found an email from Eirik, one of Trolltech’s founders, in my inbox. He asked me to join Trolltech and move to Oslo.

Without any formal education in software development, I was offered a great job with a more than good salary working on what I was burning for. So it didn’t require a lot of debating with myself to decide to drop out of university, quit my freelance contract with the US based Linux startup and pack my bags to move from Austria to Oslo, Norway.

In the spring of 1999, I finally moved to Oslo and joined Trolltech. By then, I don’t recall it exactly, we were about 10 people in total. This was prior to any funding rounds.

Working at Trolltech was great. The vision of Trolltech’s product Qt was great. Building a framework which allows software developers to build GUI applications which run on any platform – with a single source code. But what was even better, was the business model.

Predating software as a service times, and not very common at this time, Qt was made available as a commercial offering as well as open source offering. The idea was, use it for free if you develop free software. Pay, if you build commercial software and make money with the software.

While the spirit was easy to understand, coming up with license terms which clearly distinguished between open source and commercial use, and satisfied the Free Software Foundation, was a tough task. It took a lot of discussions, fights and several revisions of the license terms over the coming years.

The huge benefit of this business model was that Qt became very popular due to its free nature, but commercial users generated enough revenue to continue developing Qt and building Trolltech. Many programmers got to know the open source Qt versions at universities or in pet projects and later introduced Qt in their jobs, where their employers then paid for Qt licenses.

Shortly after joining Trolltech, the company started to seek funding to accelerate its growth. I wasn’t really involved in these discussions, but one very memorable moment was when the whole team met the then-CEO of Borland, who wanted (and later did) invest into Trolltech. This was the first time I encountered the very different American business culture.

Everything was big, big numbers, big money, big growth. During the speech to the Trolltech team, when looking into rather astonished faces, Borland’s CEO pulled out a big bundle of 100 Dollar notes and waved them around, just to demonstrate that aggressive growth will benefit everyone in the company.

With this, a period of more aggressive growth and getting investors on board has been initiated. Fortunately, the developers were well shielded from the business side, so working at Trolltech was still a programmer’s paradise. 

In the beginning of this growth phase, when many new people were added to the team, Harri Porten, whom I’d later co-found froglogic with, also joined the company.

In this phase, while still mainly doing software architecture and development, I also learned more aspects of a software business. I was able to travel to several trade shows and also visited customers. I got a small team of developers assigned which I was responsible for, and I (had to?) experience(d) what team coaching means.

While working as a software developer at Trolltech still was great, the strong growth and burning money while not all plans worked as expected, also had its negative sides. Certain projects, which objectively seemed a bit far fetched, got the full support from management as the Next Big Thing. On the other hand, the core products, which generated the majority of the company’s revenue, got too little attention since they didn’t look like high growth opportunities.

With all that, more and more management was hired and at some point likely the number of managers outnumbered the number of non-managers in the company. With that, the work culture started to change. It wasn’t the developers anymore, who worked countless hours, days and nights, who had the power to decide about the product and its future. Suddenly managers, who often didn’t really understand the products, became the decision makers.

In that time, I realized that this likely wasn’t the environment anymore which I wanted to continue to work in. But more about that in a future article.