Interview: Carsten Munk

Carsten Munk will give a talk about "Beyond traditional mobile Linux" at FOSDEM 2012.

Could you briefly introduce yourself?

I'm 27 years old, currently living in Warsaw, Poland but of Danish nationality. I've been involved with various open source projects for 12 years, first within IRC daemons (UnrealIRCd) and since then working within mobile devices (which became my career) after getting hooked on Maemo with a Nokia N800. I was involved intensely in the MeeGo project, doing hardware adaptation for the Nokia N900 and have always had a heart for open source, transparency and community. I now am the project architect of the Mer project, an open source mobile core.

What will your talk be about, exactly?

I want to talk about the paradigm shift in how we use computers and its influence on the open source community - which we already see the beginnings of in the way we use our mobile devices today. People are switching away from PCs to tablets, your phone is more like a computer in your pocket and the monopoly of exclusively using the PC and the traditional WIMP interfaces has been broken.

A paradigm shift has the positive effect of creating new interesting projects and challenges, but also has a tendency to obsolete projects that are less or not at all relevant past the paradigm shift. As an example, a lot of the work we've put into the PC-oriented user interfaces may be a lot less relevant in mobile settings. New methods and ways of thinking are needed.

Compared to the previous setting with Linux having to 'win' on the desktop, Linux and open source already has a significant stronghold in the mobile and embedded domains. But we need to move beyond the machine room where we usually are and start actively developing the world. The best way to predict the future is to invent it.

What do you hope to accomplish by giving this talk? What do you expect?

My hope is to inspire people to think beyond the PC, beyond the cellphone - to begin collaborating and working together on solutions. To a world like in this video. I don't believe such a world can function without open standards and open source.

I will expect some people to be lightly insulted by what I'll be suggesting in my talk - we have many people within the open source community who actively resist change, conservative, the kind that complains if they can't run Motif-based applications on their cellphone. But new paradigms require new architectures, new settings, new ways of doing things.

While old and stable can be good, this paradigm shift is coming all around us, whether we want it or not. We need to take advantage of our ability to work together, in the open, to create a better future, fitting in this.

Cory Doctorow talks about the coming war on general computation - where he talks about the tendency to lock down devices as specific appliances. And what's the best way to ensure our software freedom and hardware freedom? To create the future and make open devices that people would actually buy and be able to use. And to that we, in the open source community, need to work hard, to fit in this paradigm shift.

What are currently the biggest challenges for mobile Linux operating systems and how should we tackle them? For instance, most mobile operating systems have very similar user interfaces. Are these good enough or do we need a radically new approach?

The biggest challenge for mobile Linux operating systems right now is that they're too handset-oriented. They're too inflexible, unportable to different form factors. We're moving to a model where even a bathroom mirror could be a computer.

The biggest problem for Open Mobile Linux is that we don't have a common application story. We can't do our own application API and expect everyone else to use it. Some open source initiatives exist to settle this, such as PhoneGap, HTML5 and perhaps the Tizen APIs, based on HTML5 and WAC.

My proposal is that we work together on cores - the Linux system stuff we all have to do anyway - and work together on common application stories. We're moving to a world where it doesn't matter if we're RPM or DEB. We need simple and working tools to make a product of any kind of device, on any architecture or chipset. And then put a UI on top and a hardware adaptation. My own initiative is the Mer project.

MeeGo's two founders Intel and Nokia have moved on to other operating systems. You were the first non-Nokia/Intel/Linux Foundation person to be nominated for an official position in the MeeGo project. Looking back to this, what went wrong with the MeeGo project in your opinion?

The MeeGo project failed for multiple reasons. One was that in practice, all decisions were made behind closed doors. Another was a misallocation of resources between MeeGo.com and the system that runs the Nokia N9 (which isn't MeeGo). So it didn't get the attention it deserved. The architecture and the code itself are fine. Too many politics can kill a project. Other issues were the difficulty to convert the daily routines of people - it's always easier to walk over to someone than to send a message to a mailing list about a direction choice. If you want to start a open source project, spread the people across multiple cities so they're required to work in the open and communicate in that manner.

I've learnt that corporations often fail in launching open source projects. They do it the wrong way. They should send engineers and pioneers ahead, let them run the project in the open and start it up, help them with investment for the things they need. So that the people who work on it has working in the open in their daily routine - else you can't do a successful open source project.