Linux Audio Development (LAD)

Mixxx

Linux Audio Developers Spotlight interview with Owen Williams (aka DJ Huxley) from Mixxx

This interview was conducted by Amadeus Paulussen in 2026.

Dear Owen,

I'm delighted to have the chance to chat with someone from the Linux Audio DJ universe. 😎 I always wanted to learn to DJ but never actually managed to do so. Nonetheless, making it easier for DJs to use Linux as their performance center piece is genuinely close to my heart.

For music producers and musicians on Linux, audio interface support and audio software availability (plugins above all) are often the "reasons" people stick with macOS or Windows. What reasons do DJs most commonly mention for not using Linux?

Hardware support is always the biggest barrier to entry for DJing on Linux. There are hundreds, if not thousands of DJ controllers, and they all come bundled with software that runs only on Mac and Windows. Users are usually going to stick with the officially-supported software rather than experiment with third-party software. Also, DJs are an artistic bunch and not always the most technically savvy. Linux is still seen as an advanced operating system for computer experts, so DJs opt for the more familiar consumer options.

I personally use Mixxx mainly for two things: editing tags in my music library and playing playlists via Auto DJ. 😜🫣 And, I occasionally practice DJing with a Pioneer DJ DDJ-200. 😎 And, to me, the software feels mature, the UI, UX, and feature set coherent and overall surprisingly polished. Would you say that most DJs currently using proprietary software could switch to Mixxx without significant issues, or are there areas where Mixxx lags behind commercial alternatives?

Mixxx works great for the heart and soul of DJing: selecting tracks and mixing them together in real time. It also supports advanced features like 4 simultaneous decks, loops, hotcues, keylock, and many many other features listed on our webpage. There are some advanced features missing from Mixxx, like certain advanced effects (which are often proprietary and have no free alternatives), streaming music directly from download services (which is not legally possible with Open Source software), or auto-removal of drums or vocals from tracks. But usually the barrier to adoption is perception of stability and high switching costs. Mixxx is seen by many as a "toy", whereas paid software is "professional". In an industry where perception is a huge part of success, showing up to the club with a free program is frowned upon versus showing up with a USB stick ready to plug into Pioneer gear. And practically, it's a huge amount of effort to move from one program to another. The user interface is different, the audio library may not translate with all the correct metadata, and even the way the music sounds may not be the same. Mixxx does not have a big killer feature other than its price and openness, so it is hard to make the argument that people should switch from pro DJ tools to Mixxx.

Mixxx supports Linux, macOS, and Windows. How has cross-platform support evolved over time, and which platforms currently present the biggest opportunities or challenges?

Mixxx has been cross-platform as long as I've used it (since 2009) but the first version was Linux-only. Our non-Linux versions are by far the most popular, with Windows downloads much larger than the other options. We have always tried to maintain parity between operating systems, with no "preferred" OS. At times, a lack of macOS developers has led to some performance issues on that platform, but recently those issues were solved.

Has Mixxx always been available for Linux, macOS, and Windows, or did development start on a single platform?

The first version of Mixxx (circa 2001) was Linux-only, with Mac added in 2003. Windows support was added in 2009. Since then we have always released for all three platforms.

Many plugins are developed using frameworks like JUCE. Can you tell us a bit about how you develop Mixxx?

We use Qt, a low-level cross-platform framework for developing C++ applications. It handles graphics, file access, etc. For sound support, we use PortAudio. There are some other libraries for specific needs, but those are the big pieces.

Is there a Mixxx core team, and if so, how are you organized?

Yes we have a core team, and we operate on a consensus model to decide what gets approved and merged. Contributors who stick around and participate are invited to join the core team when we feel they've demonstrated reliability and an ability to work together through disagreements. Anyone may post pull requests for review, but a core member review is required and only core members can merge code.

Are you primarily Linux fans, or are there macOS and Windows devotees among you too?

Because cross platform support is critical to Mixxx, we have developers on all three platforms working regularly with us. Probably most of our core team is on Linux but we don't actually track who uses what.

How important is the Mixxx community to you, and are users and contributors usually different people, or are there also users who contribute code at the same time?

The Mixxx community is everything! There are a lot of levels of users+contributors. Some people just use Mixxx. Others are power users who help other users with installation and configuration problems. Some people help write mappings for controllers with XML and JavaScript, even if they can't be a full code contributor. Others work on graphic design or writing blog posts. Even among the code contributors, some are heroes who work on everything, and others work on small pieces or specific bug fixes. So there's a huge range of contributions that people make to Mixxx.

As I mentioned, I only have surface-level knowledge of the DJ world, but I'd guess that marketing (similar to audio plugins cosmos) plays an increasingly important role in making software visible. How does that play out for you?

We don't do much marketing. We are in a privileged position as the only full-featured DJ software that is Open Source, and the only full-featured DJ software that runs on Linux. As such, anyone looking for music performance software is going to find us quickly with a simple web search. As the proprietary DJ ecosystem becomes increasingly enshittified, we might want to do more outreach to disgruntled DJs that there is an alternative platform that does not charge fees for features and does not practice planned obsolescence.

What importance would you give to user interface design—not only, but also, in terms of visual identity and ergonomics? And, how do you approach this topic?

The interface is critical for a DJ, and we need to balance clarity of information with the density required to expose all the features users may want. We have dedicated skin designers who create and improve our skins as the code is developed. Our biggest active project is actually a complete user interface rewrite, migrating from our custom XML schema to the open QML standard. This will open up a huge range of possibilities for the Mixxx interface, allowing for a more dynamic and responsive UI.

I keep hearing from developers that they wish more things would be Open Source, not only software, but at the same time that they consider it a difficult model from an economic standpoint. How does that look for you?

As an all-volunteer project, we do not have economic considerations. We make enough money from donations to cover our basic infrastructure expenses and the occasional merch sale. This frees us to focus our energy on the features and fixes we think best benefit our users.

These days, there's incredibly good Open Source software in almost every corner of the audio landscape. What is your personal standpoint on Open Source vs. Closed Source software?

My professional history is in the video and film industry, and I've used a lot of closed source software in the course of that work. There are a lot of cases where the closed products are much more capable than their Open Source counterparts, so I never begrudge anyone that feels they can't use the Open Source alternative. At the same time, we've now seen a feature-length movie made in Blender win an Oscar award (Flow(2024)). Open Source alternatives play such an important role in the software world and we should all be working hard to keep them maintained, improved, and relevant. Even if many people don't use them, it's important to carry the torch and show that there is another way.

Open Source also presents a unique opportunity for DJs to customize their setup just how they like it. My own version of Mixxx has lots of little tweaks in the user interface and controller mapping to suit my style of play.

How have transitions like Wayland and PipeWire affected Mixxx development? What improvements have you seen, and what challenges remain?

We still have graphics problems under Wayland, and not a lot of in-house expertise to solve them. But we are fixing issues as we find them. As for sound, Mixxx works best with exclusive direct hardware access, so ALSA remains the best experience. For a while we have supported PipeWire via PulseAudio compatibility layer, and we have a new project for native "pro" low-latency PipeWire support. The biggest challenges for us have been to keep up with Qt's evolution, and the transition from Qt5 to Qt6 took a very long time. The transition to QML is the current big project.

Does the infamous Linux fragmentation ever give the Mixxx team headaches, or is that, in your opinion, a myth worth debunking?

We used to get bug reports from people with strange Gentoo systems that we were unable to reproduce, but that hasn't happened in a long time. The most common issue with fragmentation is when a distribution uses an old version of a library we depend on, leading to bugs we can't fix without the OS updating their packages. We also see issues in hardware variations, like Intel vs ARM, or issues with OpenGL support on graphic cards.

You offer Mixxx for Linux through a variety of channels—from source code to Flatpak to distribution-specific packages. What are your thoughts on the topic of software distribution on Linux, and do you think Flatpak has the potential to establish itself as a standard in the medium term?

I think Flatpak is the best way to get Mixxx into people's hands that's consistent and reliable. While native distribution packages have their advantages, often package maintainers have to create custom patches or they link against too-old versions of the libraries we rely on. This can generate bug reports for us that take time to sift through.

What impact do you think it would have on Linux Audio adoption if major DJ software vendors such as AlphaTheta, Serato, or Native Instruments released native Linux versions of their software?

I think there is very little chance of manufacturers releasing Linux versions of their software. If anything, these companies are moving towards a future where DJ functions are embedded into the controllers themselves and do not use any desktop software except to organize music libraries. There is no compelling reason for them to release Linux desktop software at all.

How much of Linux DJ adoption depends on hardware vendors providing Linux support for controllers, mixers, and audio interfaces?

We have been fortunate to work with some manufacturers who send us controllers so that we can build support for their products. For the rest, we rely on the protocols being open so that we can easily create our own controller configurations for each piece of hardware. The controllers we do not support mostly use proprietary USB interfaces or require special handshake protocols to unlock themselves. When manufacturers use USB class-compliant audio drivers, MIDI, and USB HID, we can do the work ourselves.

In live DJ performance, reliability matters more than almost anything else. How does Linux compare to macOS and Windows in terms of latency, stability, and trust during live use?

I don't see a lot of difference in the stability of the operating systems themselves. Rather, Windows and macOS are much more likely to interrupt your set with a request to download and install updates for your computer. This intrusion of the operating system can be as disruptive as crashing, and Linux stays much more out of the way.

AI is increasingly entering music production and creative software. Do you see meaningful use cases for AI in Mixxx, or is the hype mostly ahead of reality?

In general we do not have a strong appetite for AI features, and will probably never integrate support for paid AI API services. Hypothetically we could use machine learning techniques for very specific features. One place machine learning processing could be useful is in Stem Separation, which is taking a song and breaking it into separate rhythm, bass, lead, and vocal tracks (as opposed to simple traditional EQ filtering). I could also imagine AI being used to suggest next tracks to play, or an "intelligent shuffle" feature for our Auto-DJ, but we have no plans to develop these.

What would you generally recommend to developers who are eyeing a Linux port of their software?

It seems to me there are two avenues for porting that are the most realistic: choosing a cross-platform foundation for your app to begin with, or using Wine to run Windows software under Linux. If neither of these approaches is workable, I suspect doing a Linux port will be much too difficult to attempt.

Do you expect more creatives to move from Windows and macOS to Linux over the next few years? What do you think is driving that interest?

I think the success of Flow (the film created with Blender) has done a lot for raising the profile of Open Source software. And as pro DJ platforms consolidate and enshittify, people are taking a second look at Mixxx as well. DJs may find that a controller+Mixxx bundled device performs just as well as a controller costing twice as much, so we may yet see more copies of Mixxx on big stages!

If you had three wishes limited to Linux Audio, what would be your selection?

1. I wish someone would add USB hotplug detection and support to the PortAudio library. Right now if the user pulls the USB cable on their controller, Mixxx crashes. This is frankly unacceptable but also an extremely hard problem to solve.

2. Mixxx suffers from poor effect plugin support, especially those fancy plugins that draw complex user interfaces. Users love tweaking effects, and I would love to see better support for even proprietary effect plugins in Linux.

3. I wish our various projects would work more closely together! Mixxx should be talking to Azuracast, and they should all be in conversation with PipeWire, etc. Right now we are in our own little worlds.

In terms of your plans for Mixxx' future, what areas are you most excited about over the next few years?

We have seen an explosion in interest in Mixxx, especially embedding Mixxx on a Raspberry Pi and connecting it directly to a controller to create a fully integrated product. We look forward to seeing Mixxx in more devices, and maybe even being bundled with hardware controllers. The other major work we are doing is to rewrite the user interface in QML. This will greatly modernize our UI and allow Mixxx to be compiled for mobile devices, not just laptops. Rewriting the UI is a major undertaking, but we're making great progress and when it's done, we will release Mixxx 3.0 :).

Would you like to touch on something that I haven't yet?

I am not only a Core Contributor to Mixxx, but also a regular user! I DJ professionally (editor's note: DJ Huxley(fur)) every month or so in nightclubs and conventions with my own customized build of Mixxx. I have never had a crash during a performance in over 15 years of playing out, proving that Open Source can share the stage with the pro options. Check out my mixes at Mixcloud.

() Back to top