Thursday, October 27, 2011

Introducing Inqlude, the Qt library archive

Today I would like to introduce you to Inqlude, the Qt library archive. The goal of this project is to provide a comprehensive listing of all existing libraries for developers of Qt applications. So if you are creating applications using the Qt toolkit, and are looking for libraries, components or modules to use, Inqlude is meant to be the place where you find all information and pointers to get started.


The Inqlude project started at the KDE sprint at Randa this summer, where we discussed the idea of a "CPAN for Qt". There is a thriving ecosystem of libraries around Qt, KDE obviously being a big part of it. But there is no easy way to get the complete picture of it, and simply get the libraries which fit your needs best, independent of if they are part of Qt itself, of KDE, or of any other of the numerous places, where people publish their code. Other languages and frameworks have systems for this, like Perl has with CPAN, or Ruby has with Rubygems. So these served as inspiration and I did a prototype on my way back from the sprint on a Swiss train.

During the Qt contributors' summit, the desktop summit, and recently this week at the Qt contributors' day at the Qt DevDays in Munich, we had more discussions about it, collected feedback on the prototype, and refined the concept. During the last SUSE hack week I sat down to bring the code and the web site up to an alpha state, where it could serve a useful starting point for the project, for application developers to find libraries, and for library authors to contribute data about their projects.

The main component now is the web site. It's listing the data about Qt based libraries. These are only pointers, there is no hosting of code or packages involved. Inqlude is not meant to duplicate any of the functionality of code hosters, or distribution packaging systems, but integrate well with what's already there.



The meta data is stored in a git repository. So it's easy to contribute, following the well-known procedures of submitting patches. More information can be found on the "how to contribute" page on inqlude.org.

In addition to the web page, there is an inqlude command line tool. This can be used to get libraries and install them locally. Right now it's a prototype. It works with openSUSE 11.4 using the openSUSE build service as source. In the future we'll add more backends for other distributions and add more information about packages, so that you can easily install the libraries you need. Again this is not meant to duplicate any existing tools, so we'll make use of the native package management stacks and build tools which are already there.

The command line tool is inspired by the Ruby gem system, which makes it amazingly easy to distribute and get software. The inqlude command line tool prototype uses this mechanism itself. See the instructions how to get and use it on the inqlude.org page for more details. With C++ it's of course more difficult to distribute software, because there is a build step. But here comes the open build service to the rescue, which takes care of this step.


So what's next? This blog marks the official beginning of the alpha phase. The general structure of the web site is in place. So the next steps are to complete the collection of library data, and make the web site ready for end users. When this is done, we'll enter the beta phase. The goal of this phase will be to make the command line client ready for end users. Then we'll release Inqlude as 1.0, ready for production use.

If you want to help, you are more than welcome. Just go to the inqlude@kde.org mailing list. There we'll coordinate development, collect data, discuss, and help, if you have questions. See you there.

Friday, October 14, 2011

Fifteen years of KDE

Fifteen years ago Matthias Ettrich started the KDE community. On 14th October 1996 he wrote his famous email to the de.comp.os.linux.misc group on Usenet. He called for other programmers to join him to create a free desktop environment for Linux targeted at end users. Many, many people joined. Thousands of developers wrote millions lines of code. We did 90 stable releases of our core set of applications alone, not counting all additional stuff and the thousands of 3rd party applications.


So now, fifteen years later, we are done. We have a wonderful free desktop environment for Linux which is used by millions of users every day. Well, are we done? Not quite. While we have reached the original goal of creating an appealing desktop which makes Linux accessible to everyone for everyday tasks, our scope has broadened. There are whole new classes of devices in need of free and friendly user interfaces, it's not only the desktop anymore. The cloud presents new opportunities for connecting computers more than ever, and presents new challenges for freedom of software and data.

We are on it. The last two weeks saw frentic activity in the KDE community. Last week we released version 4.7.2 of our flagship product, the classical desktop. On Sunday we released Plasma Active One, our speedboat going into the waters of tablets and the device spectrum. On Tuesday we released ownCloud 2, our helicopter going into the cloud on its mission to retain your freedom and control about your data.


The KDE desktop is part of our live. We and millions of others use it every day. It's great for getting work done. We are on par with other desktops, but we are still pushing the limits and innovating, for all the users out there. This won't change anytime soon.


Plasma Active is one of the results of our innovation activities. It brings KDE to tablets, and it comes with a strong vision to create an elegant, desirable user experience for a spectrum of devices. It builds on the foundation of the KDE software, but it opens new doors, explores new areas. I'm really looking forward to what we can achieve there.


ownCloud goes beyond what we have done before. It runs in the cloud. But it's built on the values and the community of KDE. We deeply care about software freedom. Enabling users to retain control about their data and their privacy is a big part of that. ownCloud does that where other cloud solutions fall short. It brings fresh blood and energy, and opens up a space, which we weren't able to address before. Here as well I'm looking forward to what we can do.

Fifteen years of KDE is a long time. I joined the community in 1999, and I can say that KDE changed my life. I probably wouldn't have the job I have today without KDE, I wouldn't live where I live today, and I would miss a lot of great experiences and friends I wouldn't have met. In the interview I gave for the "people behind KDE" series, there was a question, what my favorite feature of KDE was. My answer was: "the community". Many other KDE people answered the same.

That's the core. KDE is an awesome community. I'm proud to be part of it, and I'm looking forward to be part of it for the next fifteen years of KDE.

Thursday, October 13, 2011

The demise of the Windows platform

I bought a Windows game last week. What I got was a scenic tour through the demise of the Windows platform. I knew that Windows as gaming platform was troublesome, but it never was as clear that it's actually moving towards irrelevance. If you ever have seriously played games on Windows you know this cocktail of driver updates, googling error messages, entering illegiible cryptic codes from stickers hidden in game boxes, waiting for online activation, going through update popups of various origins, and what not. It took me something like two hours before I was even able to start the game. I love games, and I have played quite some games on Windows, but I might be done with this now.

Of course Windows as a platform won't go away anytime soon. There are hundreds of millions of people running it. But the interesting part is that there are less and less reasons to do so. One of the arguments why people don't use Linux on their desktops always was "I need Windows to run my games". I heard this a thousand times. But as this argument becomes irrelevant, the only argument left is "I have always run Windows".

The free Linux desktop is mature. It's not only on par with proprietary desktops on other operating systems, it actually is innovating and moving beyond what other systems do. It covers all the needs of the vast majority of use cases. It has a variety of office suites, it runs several fine web browsers (another area where Windows has lost relevance up to the point of being made fun of), it has excellent tools and applications in almost any area you can think of, it's a primary choice for software developers, it even moves beyond classical desktops to netbooks, tablets, and more.

In addition the free desktop has some inherent attributes where Windows just can't compete, first of all the software freedom, but also the development model, and the distribution ecosystem. When you install a Linux desktop you have a fully functional system, you can browse the web, you can send email, you can edit your spreadsheet, even printing works out of the box these days. When you have Windows installed you are at the beginning of an odyssey to add all the bits and pieces you need to have make your system functional and secure.

Some might argue, that the desktop itself is becoming irrelevant. For some areas that might be true. People will use their phones or tablets or game consoles for things they have done on a desktop before. But there are so many people using their computers for work and other serious things, where you do want to have a solid desktop, probably not exclusively, but it won't go away for a very long time.

You might also argue, that people are using web applications instead of desktop applications. This also might be true in some areas. But the interesting part is that web applications have developed into another kind of desktop applications. The platform is different, it's Javascript and the web client environment, and the apps are heavily connected to web services. But you absolutely need a solid platform on your local system to be able to run the web apps. This platform is more than ever based on Linux (think Android), and there is no reason why this shift shouldn't continue to happen very quickly.

KDE is in an interesting position here. We are one of the key players on the free desktop, we have a mature classical desktop, we are expanding to other form factors and into the web, we have a strong community, which can make things happen, nobody would have expected to be possible (or who thought that Matthias announcement from 15 years ago would result in something like we have today).

We even have a story on Windows, where we have central applications like Kontact or Okular, which are more than a replacement for the apps, which are only running on Windows. This is more important than ever as a migration strategy. It's very easy to arrive at a point, where all the applications you run on Windows, are actually running natively on Linux as well. Running LibreOffice, Firefox, and Kontact on Windows? Fine, move over to Linux, it will run as well and better, and you will enjoy the freedoms of free software.

We have a chance to change the world, let's do it.