Buckets of (cold) Water

Almost 11 years ago I got a call from some Matthias guy in Oslo, asking me if I'd still be interested to start an internship as the Support Technician at Troll Tech. A bit more than one week ago, after Stephen Elop's presentation of the new Nokia strategy, I thought for a very brief moment - and for the very first time in those 11 years - that the ride with Qt was finally over.

When Nokia "bet the farm on Trolltech" we started with a Qt that was great on the desktop, but notoriously unsuccessful on devices - too slow, too big, and with UI components that were designed for keyboard and mouse but not for the touch screen. With the release of Qt 4.7 it looked like we had, after three years in Nokia, finally managed to turn Qt into a rock-solid platform for embedded Linux, Symbian and MeeGo. Something we never managed as Trolltech, in spite of Qtopia! We got fantastic feedback to Qt and Qt Quick from our colleagues in Nokia and you guys in the developer community really believed with us in a Qt/QML based developer offering for hundreds of millions of Nokia devices.

Cold Water in the Face

Last Friday's change of strategy was a surprise and shock for the whole Qt organization. I personally can understand very much that the Qt community is upset and disappointed, and that nobody would be surprised to see the end of Qt in Nokia. And if you don't believe or trust Rich Green's latest communication regarding the future of Qt in Nokia, then I don't think any speculations or promises from my side will help.

So instead of speculations and promises I'll share a few facts so that you can draw your own conclusions.

"We are the people who build the company" (from the Trolltech values)

We are currently ca 260 people in the Qt Development Frameworks organization in Nokia. In Oslo, where Qt was started, we have a team of 60 engineers, product managers and technical writers. We have organized ourselves in teams, each of which is responsible for a technical domain in Qt. The Earth team develops the foundations of Qt such as the object model, data structures and low-level OS services; the Fire team works on the rendering stack that turns data structures into pixels on the screen, using as much of the GPU as possible; the Water team makes shiny and flowing interface components and works on UI concepts, such as accessibility or input methods; and the Air team makes the cloud accessible to applications, through internet technologies such as WebKit and JavaScript. We don't have a Love team - yet! - but we have the A-Team that is responsible for documentation, examples and demos.

In Brisbane we have ca 60 software developers and testers, and that's the home of the QML engine in QtDeclarative, of many Qt Mobility APIs and of Qt/3D. In Berlin, 40 Qt developers work on the tools side of the Qt SDK - Qt Creator, debugger and toolchain integration, QML designer, Qt Assistant etc. In late 2010 we started a new site in Finland which is now our biggest team. Ca 100 engineers were hand-picked from other organizations in Nokia and are now working both on APIs and features for Qt as well as releasing and testing.

On top of that we have our Qt sales, marketing and services teams in Oslo, California, Germany and Asia, and they continue to advocate Qt for projects and use-cases beyond Nokia. This is fantastic for at least two reasons - more developers are using and contributing to Qt, and we can introduce new Qt technologies into real projects without having to wait for the next Nokia device.

During the next 2 months we will have 8-10 new guys join us in Oslo, and we continue to hire. We intend to grow the organization from currently 60 to ca 80 by the end of this year, and we are doing a lot of recruitment these days.


So what are we working on? As usual, we are developing in a number of projects that are very relevant for Qt's future success, irrespective of what platform our colleagues in Nokia decide to build the future smartphone business with. For instance, the Qt Quick Components project and Project Lighthouse continue the effort of making Qt more suitable for form factors and runtime environments other than the desktop, and the QML Scene Graph is going to radically improve the performance of the rendering of QML UIs. And there are a lot of structural projects as well, like the Modularization of Qt and the ongoing effort of bringing Qt further under Open Governance.

Upwards and Onwards

In other words, we work on exciting stuff, we grow the organization, Nokia continues to invest into Qt. It is of course naive to pretend that Qt was not on the burning platform, and that we are not affected by the strategic change. However, I personally like to think of Qt during the last years as the bucket of water that we tried to quench the fires with - and in the end that bucket turned out to simply be too small for such a big task. Now a part of Nokia will gradually move to a different platform for a part of the company's strategy, and as long as that platform doesn't burst into flames we might not need Qt to be a big bucket of water again.

That is not a bad thing - because we don't really like to throw buckets of water at a burning platform if instead we can make breathtaking fountains and rainbows.

Blog Topics: