Fedora, Red Hat, and investing in the future

It was just about 4 years ago that I hopped on a plane to go to Raleigh, North Carolina to go meet up with some folks and work on Marketing Things for an open source project that I had recently started contributing to, called Fedora.   The Fedora Marketing Team was having a FAD (Fedora Activity Day) – and I was sponsored to come out, get things done, watch some hockey, and eat some barbeque.  With the exception of my significant other, this was pretty baffling to most of my family and non-internet friends; not about what one might expect, which is, “You’re doing things for free?” – but mostly, “They’re *paying* for you to fly out there?” 

Flash-forward to the present – and while I certainly didn’t know everything that I now know today, standing in the Fedora Project Leader shoes, four years later – my answer is still remarkably similar: Red Hat invests in Fedora because it is the upstream for Red Hat Enterprise Linux. 

Red Hat’s investment in Fedora is significant; more than a dozen people support Fedora’s community infrastructure, both “people” and “technology”, in their full-time roles as Red Hat employees. Hundreds of engineers who work on open source projects upstream of Fedora integrate their work into the releases we do every 6 months.  Budget is provided for collaborative events, such as Fedora Activity Days, and FUDCons & Flocks, as well as for equipment, bandwidth, swag, event sponsorships, media, and other various services. 

Of course, being the upstream for RHEL means that Fedora is much more than simply an *integration* point. The Fedora Project community is made up of contributors from countless viewpoints and interests, both in terms of contributions and use cases. If you’ve read “The Lean Startup,” you’re familiar with the notions of “build the right thing,” and “faster feedback loops”; Fedora provides this exact model which has enabled the success of RHEL. Our rapid, 6-month cycle enables Fedora to quickly integrate the latest and greatest technology advancements – and to backtrack, tune, or adjust how those features work based on feedback in time for the next release.  This process has in turn enabled Red Hat to produce a release of RHEL every three to four years that is not only consumable by their enterprise customers, but is also expected to meet their current technological needs.

The Fedora Project recently celebrated its 10th anniversary – and its 20th release – of developing the operating system we know and love as Fedora. Over those 10 years, the technology landscape has changed dramatically, not just in terms of what and how things are produced, but also in terms of how they are consumed. It’s not particularly a chicken-and-egg situation, but more simply an evolution where technology and use have grown together. 

  1. Breadth, complexity, and velocity: We’ve seen the emergence of compute virtualization, cloud, big data, virtualization round 2 (The Network Edition), and containerization technologies, one right after the other – primarily propelled forward by technologies developed in open source communities.
  2. Agility and resilience, in both business and infrastructure: The ability to consume ever-increasing volumes of information – either about your business, or your infrastructure – and rapidly make decisions based upon that data, and *act*, is what separates successful organizations from dysfunctional ones. Increasingly, people are not building culture, or infrastructure, with permanence in mind; the need to be agile also drives the need for resilience – the ability to bounce back from failure.   More specific to infrastructure technologies, the ability to abstract, simplify, and automate enables the ability to scale in size and more rapidly develop New Stuff – which has manifested itself in a emerging sea of packaging, configuration, orchestration, and other glue-ish tools for infrastructure, many of which were born from the need to more efficiently deal with the operating system.  Organizations strive to build the right thing, the Fedora Project included, and choice abounds when it comes to technologies to enable that building.

The Fedora.next initiative is paving the way for Fedora 21 and beyond; to the most casual of onlookers, the biggest change from previous releases is the shift to building purpose-specific versions of Fedora – namely, Workstation, Server, and Cloud-image products – rather than the “one Fedora to rule them all” release that we have produced in the past.  This is, essentially, putting us far closer to “building the right thing” than we’ve ever been; it helps us to make the technologies we develop more consumable for our users and contributors, and enables a tighter feedback loop on what we are producing in a world where the pace of technology is moving at warp speed. And Fedora’s success in shifting focus to a more diverse audience via a change in product set directly enables Red Hat and other companies to have more successful projects themselves.

And speaking with my red fedora on – Red Hat, of course, does hope to benefit from these new purpose-specific products and the emerging work around them. Just as a single-purpose Fedora has helped select technology for today’s RHEL, Red Hat hopes this diversity will do the same for future RHEL. The communities that are springing up around Linux and open source development have become very diverse, and so have Red Hat’s customers, and product lineup. The more appeal we generate with Fedora for those communities and use cases, the more value Fedora adds to the cycle of participation and integration. Since Red Hat’s engineers end up working on many parts of that cycle through free and open source upstreams and integrating in Fedora, it’s no surprise they’re interested in helping Fedora get these new products well thought out via the working groups. Bettering Fedora’s appeal also directly impacts Red Hat’s ability to build its ecosystem and thereby bring even more participation to, and investment in, Fedora.

All that said – our own need in the Fedora community to build resilience and agility, in both our infrastructure and culture/community – are key to successfully launching three products. The process isn’t going to be as easy as flipping a lightswitch (sorry, folks!), but rather more of an evolution. Many new things are already underway in terms of new technology – such as our work on coprs, collaboration with Docker, or the (IMO) exciting work going on in the Cloud SIG around atomic upgrades  – as well as rethinking some of our existing processes around how we build and test our products. As we navigate through this process, our fearless program manager, Jaroslav, will be helping to coordinate and plan how all of these pieces fit together – and I encourage you to keep an eye on those planning details and dependencies so that we can deliver a Fedora that is prepared for the next 10 years of technological innovation.

I’m a fan of the concepts behind the new purpose-driven products, and I encourage you to bring constructive inputs to the mix. Of course, I’m also delighted for people to bring contributions around the products —  just as we’ve done for our past 20 releases. It’s an exciting time for Fedora, and a great time to be involved and to influence the next 20 releases to come. (Or more!)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s