by Frank Karlitschek
posted on Thursday, March 14th, 2013
posted in Community
Today we released ownCloud 5, a very important milestone for the ownCloud community and perhaps the most important release so far. But before going into the details I want to take a step back and look at what the original idea of ownCloud was at the beginning.
The idea of ownCloud was and is to enable everybody to host, control and sync and share their personal data without giving control away to the big data silos like Dropbox, Google Drive, Skydrive and iCloud. I think today we have all the features in place to say that we reached this goal. Everybody from a home user to a big enterprise can host their own personal cloud installation. I'm also super happy about the integration into KDE and GNOME because this is important to provide a really seamless experience for users.
It's a coincidence that CERN invited me to give a talk about ownCloud and data silos that I will give here in a few hours at the exact same day ownCloud 5 is released. CERN is also the place where Tim Berners-Lee invented the World Wide Web 22 years ago. It's interesting that the Web was built as a completely decentralized system where no distinction between someone who is publishing data and someone who is consuming data exists. There is no concept of a centralized entity that everybody connects to. Everybody can be sender and receiver at the same time. Just as Berthold Brecht proposed in 1932.
Interestingly, the web looks a bit different today where a huge amount of the traffic goes through websites like Facebook, Google, Dropbox and Amazon. Where is the idea of a decentralized and federated web?
Today we are deciding how the world will look like in the future. We, the IT community, set the course of the train that is called "open society" now and we can decide into which station the train will roll into in 5-10 years. Is it the one where all the people still control their own data and information and can decide who has access to the personal files, photos, contacts, location data, chat messages and other personal information or will we live in a future where all the personal data of all the people in the world are stored on the servers of just a few big organizations and commercial interests, terms of services and secret services decide who has access to the digital life of everybody?
If you care about these questions then join the ownCloud community or other free software projects and work on decentralized and federated alternatives.
ownCloud 5 is the result of the work of our awesome developer community. More and more people join and are getting more involved. To me this is a sign that we are doing something right and that ownCloud is not just a crazy idea that no one needs but something that is very important to a lot of people.
We did 2 major developer meeting during the development of ownCloud 5. One in Berlin and Ann Arbor in parallel last fall to do most of the ground work. And one just a few week ago in Stuttgart to really streamline and polish ownCloud 5. We will do the next developer meeting in a few month and everybody is welcome.
ownCloud 5 is also proof that a company and an open source community can work together on a product in a very open and effective way. It is needed of course that both parties have a shared interest in the success of a product and that the development happens in the open. But I must say that I'm proud that we managed to set this up in a way that works very well.
We added a ton of cool new features in ownCloud 5. The features are interesting enough so that they deserve a blog post on their own. So tomorrow I will blog about the new ownCloud 5 features. But as important as the new features are three other things:
A lot of work to improve the quality of ownCloud went into the version 5. We launched a quality imitative during our developer meeting last fall. We introduced peer reviews for all commits that go into the core. We launched a new documentation system that has great new docs for users, administrators but also developers. We have improved application templates and sample code to help newcomers, we launched a new Jenkins-based continuous integration testing imitative with a lot of tests. We switched to a new and better bugtracker and provide daily builds of the server and the clients. This all helped us to increase the quality of ownCloud 5 significantly.
A lot of work went into ownCloud 5 to improve the overall performance. One of the key components is our filesystem cache and abstraction layer. This was completely rewritten to improve the performance significantly. Some tests show up to 500% faster performance compared with ownCloud 4.5. if you work with a lot of files or you have a server with a lot of users. We also looked into the overall database structure and optimized it for big installations. The sync protocol was also improved to reduce the roundtrips between the clients and the server to sync faster.
Thanks a lot to everybody who made this release possible. You guys rock!