I recently connected with Matthew Miller, distinguished engineer and Fedora project leader, to discuss the project and the future of Linux. His insights are invaluable not only to Linux users but to anyone maintaining a Linux distribution or those considering creating their own flavor of the open-source operating system. Here is what Miller has to say.
Jack Wallen: What is the one thing missing from the Linux community’s efforts in marketing the operating system for the masses?
Matthew Miller: I think, fundamentally, the problem is that there’s not a mass-market for operating systems at all. Some people, of course, find technology at this level fascinating — probably a lot of the folks interested in reading what you and I have to say about it. But, relative to even other geeky pursuits which have become mainstream (hello, grown-ups who build cool LEGO things! hello, fellow D&D nerds!), caring about your operating system at all is pretty esoteric.
SEE: Become a Linux expert just in time for the 2022 boom (TechRepublic Academy)
There is certainly a market for operating systems at a corporate level, in the enterprise and for millions of different technology use cases businesses need to solve. Something has to power the cloud, and there needs to be a platform for all the software that a modern electric car needs to run. Those markets have actually already decided that the answer is Linux, and those are definitely markets with a lot of money at stake.
But for the masses, the desktop operating system (and, ever-increasingly, the OS for mobile devices)? The OS is just an implementation detail as part of a whole experience, and talking about that level generally makes people’s eyes glaze over. Of course, when it doesn’t, I know I’ve found a kindred spirit, but, again, a rare kindred spirit. So — thanks for bearing with me; I’m getting around to a direct answer—what are we missing in marketing Linux for the masses? I think it’s a lost cause to try to “sell” our quirky technology interest to people who don’t see it already. We need to take a different approach.
I read your article on 5 things Linux needs to seriously compete in the desktop market that you probably never considered with great interest, and I think there are a lot of interesting ideas. (Any celebrities ready to endorse Fedora Linux? Call me!) But, I actually very strongly disagree with one of your core points. I think our message, at its root, has to be around open source.
We’ve got some great features Windows or OS X do not — and they have some technical strengths as well. (Yes, even Windows. Let’s be honest!) Any campaign based around those things ultimately ends in the eyes-glazed zone. But, I think the fundamentally important difference isn’t in the technology anyway. It’s something else entirely. Anyone can be a Windows fan or an OS X partisan, or, if you really want, an OS/2 Warp enthusiast. And you can even go work for the companies that make them, well, not OS/2, but the others! You can buy stock. But, they won’t really ever be yours. Ultimately, they exist as, well, products to fill market opportunities.
But with Linux, when you install an open-source distro, you’re not just part of a fan community. You’re part of a colossal, global effort that makes software more available to everyone, makes that software better and better, and makes the whole world better through sharing. You don’t need to be a coder or have some special skill or do something to give back. Just by using it you’re sharing in this amazing undertaking, part of a move away from scarcity to an economy based on abundance. When you install Fedora Linux, you really get something that in a very real sense belongs to you: the licenses for all of the software we ship are designed to include you, not keep you out.
And, of course, that software can do all of the things you need: communicate, design, create, play, learn, work — everything a computer can do. Wherever it doesn’t meet someone’s needs, that’s an opportunity for more growth — and while a company might decide those needs aren’t worth the investment, open source isn’t limited in that way. It’s not about the market opportunity, it’s about making this thing better together.
SEE: Best practices for Linux admins (TechRepublic Premium)
I saw a tweet recently warning that if an app is free, it means the product is you — your personal information, some kind of invasive tracking, at least somehow monetizing your attention. But since we’re not going after a market, since we’re making this for a reason other than finding something to sell, that kind of thinking doesn’t apply. It’s a different, better way for the world to be.
This is a long answer, starting with the argument that there isn’t a mass-market for operating systems, and it comes around to that actually being a good thing, and why Linux is ultimately the best answer even then. I think that’s the message we need to take to people, and I think one which resonates far more broadly than any technology story.
Jack Wallen: What can Fedora 36 do for that effort to win over new users to Linux?
Matthew Miller: I read your very nice pre-release review of our upcoming version — hopefully coming May 3, assuming we can get a few final blocker bugs ironed out in time. I’m glad you noticed the focus on simplicity because I think that is indeed key for winning new users. When the OS gets in the way, it drops from the conversation I want to have about big ideas to … well, the boring technical details that people never want to deal with. Every time we have to explain those, we’re back to the glazed-over-eyes problem—which, to be clear, is perfectly reasonable. There are plenty of other things in life to worry about. It’s crucial to provide a streamlined experience so users can worry about all of the things they actually want to do. We’ve worked really hard on this.
Fedora’s vision isn’t “our operating system running everywhere.” It’s for a world where open, inclusive communities work together on this grand shared project that benefits everyone. We provide an easy, powerful onramp to, that the OS and a vibrant, friendly, helpful community that can help with any questions or problems. And if you are interested in learning more about how it’s all made, how you can help make it even better, and how you can get more involved, we’ve got easy paths for that, too.
Jack Wallen: What’s the biggest difference in Linux today vs. Linux of 10 years ago?
Matthew Miller: I think first we have to start with just the amazing ubiquity of it. Ten years ago, it was cute to find a TV that ran Linux. Now, not only is it definitely powering your TV, you’ve probably got Linux running on your lightbulbs! It’s everywhere. And while Linux had pushed proprietary Unix from the server room, ten years ago Windows-based servers were pushing back. The cloud changed that—now, the cloud is Linux, almost completely. (Anything that isn’t is a legacy app that it was too much trouble to port!) From tiny devices to the most powerful mainframes and supercomputers: Linux, Linux, Linux.
Because it’s all open source, that ends up benefiting everyone, including desktop use cases. Containers originated as server technology — cloud technology — but the same concepts are crucial for easier, safer desktop application deployment. We have a flavor of Fedora Linux called Silverblue which explores this idea in particular, taking ideas that came from CoreOS (and the short-lived Red Hat Atomic Project) for cloud and containers and exploring them on a desktop OS. I think we’ll see a lot more of that in the future.
SEE: Linux turns 30: Celebrating the open source operating system (free PDF) (TechRepublic)
We’re also seeing big changes in the way apps are written for the cloud to come to the desktop, not just in containers but in the programming language stacks that they’re made with. I’m going to try to keep this at the level that won’t go back to those glazed-over-eyes — readers can correct my oversimplifications in the comments.
Basically, every modern language provides a lot of building blocks that usually come from other smaller open-source projects. These are libraries, and they do things like format text, handle images, connect to databases and deal with talking across the internet. Projects like Fedora or Debian used to work to try to package up every such library in our own format, made to work nicely with everything else.
Now, every new language — Rust, for example — comes with its own tools to manage these, and they don’t work nicely together with our old way. The sheer scale is overwhelming — for Rust alone, as I checked just now there are 81,541 such libraries. We can’t keep up with repackaging all of that into our own format, let alone that plus all of the other languages. We need to approach this differently in order to still provide a good solution to software developers.
I think a lot of that will need machine learning and automation … we’ll need to keep adjusting so we can provide the value that Linux distributions give users in trust, security and coherent integration at an exponential scale.
Jack Wallen: If Linux has an Achilles’ heel, what is it?
Matthew Miller: Linux and the whole free and open-source software movement grew up with the rise of the internet as an open communication platform. We absolutely need that to continue in order to realize our vision, and I don’t think we can take it for granted.
That’s more general than an Achilles’ heel, though, so right now let me highlight one thing that I think is troubling: Chrome becoming the dominant browser to the point where it’s often the only way to make sites work. Chromium (the associated upstream project) is open source, but isn’t really run as a community project, and, pointedly, very very few people run Chromium itself. I’d love to see that change, but I’d also like to see Firefox regain a meaningful presence.
Jack Wallen: What does Linux need more of: New users or more big companies behind it?
Matthew Miller: Oh, I think we’ve got plenty of big companies, and honestly, I think we’re getting the new users, too. What I’d really like to see more of are more non-technical contributors. I mean, yes, we can always benefit from more packagers and coders and engineers, but I think what we really need desperately are writers, designers, artists, videographers, communicators, organizers and planners. I don’t think big companies are likely to provide those things, at least, not for the parts of the Linux world which aren’t their products.
SEE: Hiring kit: Linux administrator (TechRepublic Premium)
We need people who think the whole grand project I’ve been talking about is important, and who have the skills and interests to help make it real. Of course, getting more users is one way to do that, but we also need to make sure our projects are structured so that users feel welcome, have a sense of belonging and are inspired to get involved.
Jack Wallen: What can other distributions learn from Fedora 36?
Matthew Miller: I recently did a talk looking back at the almost two decades of our project. I started going through our history, release by release, and as I did, several big themes came up. We’ve made some mistakes and had a lot of low points along with our successes, and I hope that we can learn from them — and that other distros and open source projects, in general, can too.
First, work on your process for big changes, and make sure that community is part of the decisions that get you there. No matter what your project governance structure is, exercise transparent, open decision-making all of the time—for small things as well as big ones. That will make it work better when big things eventually do come up. Trust your community — that’s what makes any process work.
Second, community teams need someone to hold momentum. Someone has to always be there to welcome new folks, to keep meetings going, to keep request queues from going unanswered. It’s really, really easy to let everything fall to one amazing person working in some area — documentation, release engineering, whatever — and eventually that person will get burned out or win the lottery or just find some other interest, and you’ll find that whole important area collapse. Make sure those folks have support, make sure no one ever feels like they’re doing their part alone and that if they take a vacation, or decide to go raise llamas and never touch a computer again, they’ll know that someone else is there to keep things going.
SEE: Checklist: Essential support sites for Linux admins (TechRepublic Premium)
And finally, let the community lead, even when it seems scary. Let people experiment, and as a project, remove roadblocks, and that can mean especially the areas you initially thought needed to be restricted. Here, I’m looking particularly at businesses with open-source interests. When your community wants to do something which you’re worried threatens your model, it’s time to change your model. My talk has several concrete examples: Red Hat had initially decided that 64-bit support would be an “enterprise” feature not done in Fedora, but a community member created that anyway. Now, we know that bringing up new architectures in Fedora first is actually way better (hello, ARM, and now RISC-V!). Red Hat invested a huge amount in their own proprietary system-update framework … but a community-developed approach was better. (That was called “yum” — the precursor to DNF as used in the distro today.) These kinds of things take leaps of faith, but those leaps pay off. Again: Trust your community.
Jack Wallen: How is Fedora 36 different from other distributions?
Matthew Miller: I think we fit in a sweet spot! We’re fast-moving, but we do careful quality assurance. We work on bringing new features to users quickly and make it easy to go to a new release from a previous one, but we also allow you to plan for when you make that update. We try to follow the leading edge without making our users feel like it’s a bleeding-edge instead.
As we do this, we’re deeply committed to the open-source vision I’ve been talking about. Our goal isn’t just to make a better OS, to make something that belongs to just us. We want to make things better for everyone. We work closely with various upstream projects, the people who make the code that we integrate, and you’ll frequently see the technology we pioneered show up in other distributions a little bit later.
And, we’re a community-directed project — we have a stable sponsor in Red Hat, who pay me and a few others to work full time, but I’m not a project dictator, and we strive for community consensus around all decisions. Obviously, Red Hat benefits, but the bigger benefit is to everyone who is involved (and, of course, to our users). We’re also an amazing, fun community to be part of. Of course, other projects have this too, but I personally think we’ve got something particularly special.
Jack Wallen: What is the five-year plan for Fedora 36?
Matthew Miller: We’re actually working on this right now! We feel like we’re riding an amazing wave of success, and we need to make sure that we have a solid plan for the next phase so we don’t falter. Developing this plan will be (hopefully no surprise by this point, after all I’ve been saying!) a community process, out in the open — you can follow our conversation on the Fedora Discussion forum.
I don’t know what the technology will look like. I don’t think anyone can predict technology trends five years out. But one of the big flags I’m putting on our horizon is that we should double the size of the active Fedora collaborative community. We know our user base is growing, and we need to grow the project to match. To do that, we’re going to focus our investments in mentoring, in accessibility and inclusiveness and generally in community health. That will power success in whatever technology looks like in 2025 or 2027 or beyond.
Subscribe to TechRepublic’s How To Make Tech Work on YouTube for all the latest tech advice for business pros from Jack Wallen.