Swapnil Bhartiya: Who is the target audience of Fedora?
Matthew Miller: Historically, we’ve looked at a single default user for Fedora – a desktop or workstation user that could also be a potential contributor. “Contributor,” however, is a broad term; there are so many ways to contribute to Fedora, from code to content, that almost anyone interested in Fedora can contribute in some way, shape or form. Because this is so wide, it’s been hard to make actionable decisions based on this definition. And, on the other side, the desktop focus as a default doesn’t really capture everything people do and have always done with Fedora. We have people running Fedora for wide-scale web hosting, for home and small business servers, for research computing, and even as crucial infrastructure for financial services — all in real-world production. Those community members have been somewhat masked behind the desktop-OS view of Fedora, so with Fedora 21, we are redefining how we look at our user base.

As we go forward, we are looking at three distinct user groups with three distinct Fedora classes – Fedora Server, Fedora Workstation and Fedora Cloud. Fedora Workstation appeals to our traditional “default” user base in many ways, but will have a greater focus on software developers and content creators — and as the name implies, more towards people who don’t want to do all of their computing from their phone. Fedora Server will appeal to a broader, production-class audience, including some nice new tools for simple service deployment and management. And Fedora Cloud is where we will track the shift in computing to a “pets versus cattle” mindset, and where we’re experimenting with a lot of new ideas like Docker containers and Project Atomic (http://www.projectatomic.io/). The idea is that we’re not just collecting innovation for innovation’s sake, but assembling emerging technologies that can actually improve people’s specific use cases, whether it’s desktop software development or hosting massive applications in the cloud.

That focus doesn’t mean that we’re taking away the broader mission of the Fedora Project, though, and as part of this, we’re working tools and processes which will make it easy for Fedora to address other specific target audiences in the future as well.

Swapnil Bhartiya: What will be your focus as the FPL in context of Fedora.next?
Matthew Miller: Fedora.next is the umbrella term for our planning process, both for Fedora 21 and beyond – so in reality, Fedora.next is itself the definition of my focus. We want to build upon the 10 successful years of Fedora releases by ensuring that we aren’t just watching our own development process, but the progress of computing in general. I’ve been working on Fedora.next with other people in Fedora leadership and the Fedora community for almost a year now, and the FPL role gives me a new perspective where I can help coordinate the planning we need across the project. The FPL isn’t a “benevolent dictator” sort of position; my job is to listen to the community and integrate their feedback into a single, cohesive direction for Fedora moving forward.

Swapnil Bhartiya: What are the biggest challenges for Fedora?
Matthew Miller: The Fedora Project is very successful at what we do: we deliver an innovative, fully open operating system every six months – there’s no denying our success in this regard. But we want to be mindful of becoming blinded by routine, where we just allow ourselves to be walk forward in exactly the same direction, release after release, with no regard to the landscape around us. This is where Fedora.next comes in – this process will help us examine where we are, where we came from and, perhaps most importantly, where the community and our audience at large what us to go in the future.

Swapnil Bhartiya: We often see a lot of forks or duplication of work in the GNU/Linux world. How does it effect the projects? Can there be any efforts to encourage people to work together on projects instead of creating their own forks?
Matthew Miller: Forks are an integral part of the open source world, as they allow for experimentation and expansion of projects to address problems and use cases that may not have existed at the project’s conception. That being said, communication and collaboration are key, to ensure that all communities understand the end goals and that bad blood doesn’t develop. As an example, Fedora actively encourages Fedora Remixes, which are essentially Fedora forks that are designed to solve a specific use case that doesn’t fit properly under the umbrella of the project itself. We ask that credit be given to the original community and that the group doesn’t wall themselves off; if ideas/concepts are good, there should be a natural bridge back into the original project (with credit where credit is due, obviously).

Swapnil Bhartiya: New CentOS is also under Red Hat’s wings, so what kind of collaboration is there between Fedora and CentOS team?
Matthew Miller: There’s no secret backroom collaboration – we’re both community projects, and all collaboration between Fedora and CentOS is in the public. Watch our mailing lists and you can see it going on. We share a lot of similar technologies as the upstream (Fedora) and downstream (CentOS) of Red Hat Enterprise Linux, but there are a lot of differences between the projects as well. It’s interesting to watch as CentOS transitions into being a full-fledged community project rather than functioning as a singular development team, and a lot of that is being done with experience gained in Fedora.

One specific area of collaboration that I’ve been watching just this week is between Fedora’s Server Working Group and a new CentOS “Simplified Linux Server” Special Interest Group. Turns out both groups are interested in the same problem space, and they’re meeting together to come up with joint solutions rather than inventing two different approaches for no real reason. That kind of collaboration and communication benefits both projects, and demonstrations a positive cycle in our distribution ecosystem.