john.layt.net

[Edit: Updated the sponsors list]

I'm home from the Platform 11 Sprint in Randa, Switzerland, where I had a fantastic 7 days with the KDE Core team discussing the future of KDE's Development Platform. The temperature outside may have been 0 degrees and snow on the first morning, but inside it was an intellectual hothouse as some incredibly smart people (and me) debated, designed and hacked all hours of the day and night. The summary emails are now slowly making their way onto the mailing lists for wider community discussion and a dot article will follow to communicate the big picture to the wider world, but what I want to share here are my personal thoughts on the conclusions we reached on where the Platform that our Applications and Workspace are built on is heading in the next few years.

Firstly though I want to thank Mario Fux and all his volunteer crew (especially his parents) for their fantastic effort in hosting 60 hungry hackers in such a beautiful corner of the world (those mountains sure are distracting when you're supposed to be staring at a screen) and look forward to the opportunity to visit for another sprint. Thanks also to the sponsors, especially Bar Informatik AG for the internet access, Swisscom for the Swiss Army Knives, and openSuse for their help with everyone's travel costs and t-shirts.

In the KDE community we produce many fantastic Applications and a great Workspace which have lead to the production of some equally fantastic common libraries and frameworks to share common functionality. The problem is after 15 years of striving to produce a Kool Desktop Environment most of our libraries are rather unsurprisingly structured to support a monolithic desktop environment. The reality is we now live in an incredibly diverse and even more rapidly changing tech world where no one platform or form-factor will dominate like the old wintel desktop days when KDE was coming of age. While that may mean KDE will never achieve World Domination via the desktop, it also means there has never been a better opportunity than now for KDE to establish a significant presence in these new markets. I want to use all those great KDE Applications on every Kool device I own, even those where I can't run a KDE Workspace (and even better on those where I can). To do this we need to adapt our libraries to meet the challenge of surviving in such a diverse world so we can equally support these exciting new form-factors as well as our classic and much loved desktop.

In recent years we have slowly been moving the Development Platform towards such a modular approach, the 4.0 release made some major strides towards this, and now with a modularised Qt5 being announced under Open Governance we have the opportunity to complete the task. By evolving our Development Platform to a more modular and flexible set of Frameworks that integrate better with Qt we can continue to provide the same stable high quality foundations our existing Applications and Workspaces rely on while providing a flexible and light-weight foundation for exciting new developments to innovate on.

We're very mindful however of the pain a disruptive transformation can cause and that's something we don't want to do, which is why one major overriding goal for Platform 11 was how to achieve what we need to with the Frameworks while having the minimal impact on our existing Applications, Workspaces, and Users. I think the plan we developed manages to do this, and if we succeed then most people will not even notice something significant happened beneath their feet.

One benefit we're hoping from all this is opening up KDE developed libraries to be used by the wider Qt developer community, making KDE the go-to place for libraries and frameworks extending Qt. More users of our frameworks means more eyes-on-code and possibly more contributors to that code, and can be a gateway for more involvement in the KDE community. I really believe Open Governance is an exciting opportunity for KDE to improve and grow and one which we need to work hard at to make succeed to guarantee our future survival.

For the KDE Core team, finishing Platform 11 is only halfway through an incredibly busy couple of weeks work with the Qt Contributors Summit coming up next week where we will be discussing with the wider Qt community how to get involved in Qt Open Governance. The KDE agenda will be a mixture of helping establish the new community and figuring out how it will operate, and working out the technical aspects of where KDE can contribute code and features to Qt5 . Platform 11 gave us many great ideas on how we can better work with Qt and simplify our Frameworks by making some key contributions to Qt and we will be working hard to make it happen. For me it will also be the chance to pursue the topic of Printing, which didn't feature at Platform 11, to finally get my patches into Qt and to work to improve Qt to the level of our much missed KDEPrint.

Now, I really need to go and catch up on my sleep :-)