LINux on MOBile

As you may have heard or rather read, the Spanish company Liberux recently launched a crowdfunder for their new mainline Linux "Nexx" phone on Indiegogo - starting at 8 GB RAM/128 GB eMMC/LTE for 799 EUR and going up to 32 GB RAM/512 GB/5G for 1300 (during the crowd-funder).

The specs include impressive things, such as two USB-C ports and a headphone jack - quite unusual these days. It's also somewhat modular (cellular modem, RAM, and storage are on modules) and they aim to open-source the hardware and plan to manufacture the devices in Spain - meaning, the Liberux Nexx will be (successful funding assumed) be one of the only (?) smartphones designed and build in Europe.

I asked Liberux for an interview, and they were happy to answer my questions. While they would have prefered to do this as a video interview, I just could not swing that (we're preparing to move, and I have to stand-in for a colleague on holiday at the dayjob right now) - so gladly, they agreed to do it in a back and forth via email. With that said, here you go:

The interview

Let's start with something broad: Why did you decide to make a Linux Phone? It is a tiny niche, the software ecosystem is still somewhat nascent, many challenges have not been solved and the market keeps adding new ones (e.g., Voice over LTE (VoLTE) or Rich Communication Services (RCS)). In addition to that, making specific hardware for FOSS (and/or privacy-) nerds is really difficult, as these people can be very nitpicky? So: Why?

We know it’s not an easy path. But it’s a necessary one. The dominant mobile operating systems are black boxes, serving interests that often don’t align with those of the user. We’re concerned about surveillance, the lack of real control over our devices, and the opacity of the software. We wanted to build something different: a phone designed with respect for the user's freedom, running an auditable OS with no backdoors. It's a bet on a future where you don’t have to give up your privacy to have a useful device. It’s not for everyone, but it is for those who value their digital autonomy.

What level of prior experience do members of your team have with hardware projects/phone or consumer electronics manufacturing and software development?

Our team has a mix of backgrounds, all with solid experience. Pedro, for example, developed the operating system (TuxumOS)¹ for the first Linux tablet PC running on x86, and has collaborated with companies like Toshiba, Lenovo, and Fujitsu on Linux support and power management for their laptops. Carlos has designed mechanical keyboards and consumer electronic products. Our hardware design partner, Pleeda, has already developed and marketed its own portable game console. Ernesto Mansilla from Ecaman, our manufacturing partner, has extensive experience in the production and assembly of electronic boards. The team at Collabora, particularly their programming wizards, has been a huge help in defining the DTBs needed to boot the board. So no, we’re definitely not starting from scratch.

Some people have tried to cast doubt on your origins or even your integrity. What’s your response to those rumors?

Well, at first we found it quite amusing — some of the things being said were wildly outlandish, involving bankers and politicians. It was kind of funny to see ourselves portrayed as being on the “other side” of the conspiracy for once.

The truth is, free software is our life. Pedro started programming computers (an IBM 5150 his father brought home) when he was just 7 years old. He installed his first Slackware in 1993 using a stack of 3½-inch floppy disks, and he's been a Debian user since the Potato release — with the occasional flirtation with BSD (nobody’s perfect).

In the end, we honestly don’t understand who would want to spread false information about a project whose sole goal is to provide the community with a freer, more secure phone.

You write that you've looked into the PinePhone (Pro) and Librem 5. What were the biggest mistakes or failures in your analysis of the PinePhone (Pro) and Librem 5 that you don't want to repeat? What did PINE64/Purism do really well in your opinion?

We deeply respect both projects for being pioneers. The PinePhone, for instance, was very affordable, but it was more of a development platform than a daily-use phone. The Librem 5 took so long to come out that its hardware was outdated by the time it was usable.

That said, they got many things right. PINE64 was brave enough to release several iterations that became key development platforms. Purism achieved very high build quality, even if the final design was bulky. Both have taught us what to avoid — and also what’s worth preserving.

Where do you see yourself regarding software development? More aligned with PINE64 (who just do hardware) or with Purism, who basically made GNOME a thing on Mobile again?

We feel clearly more aligned with Purism. We want to offer equally polished and solid hardware, as well as a strong software development effort to go along with it.

Let's continue with pricing. I've seen discussions on the Fediverse, and you have added going two cheaper (yet still powerful) options with 8 or 16 GB RAM and 128 GB storage and LTE. Can you explain why the Liberux Nexx is more expensive than the average Android smartphone people may be comparing it with?

First, the components: 32 GB of RAM, 5G, and 512 GB of storage are not typical for devices in this space. On top of that, we manufacture in Europe and in small batches, which makes everything more expensive—from assembly to logistics. Every phone will be hand-assembled in our offices. Then there's the intangible: we don’t rely on big manufacturers or opaque supply chains. The price reflects our commitment to quality, transparency, and independence.

Let's get into hardware. Why did you decide to pick the RockChip RK3588s as the core for your product, and not, e.g., a chip from Qualcomm (e.g, QCM6490)? What makes the RK3588s a good choice for phone? Why pick it despite requiring proprietary firmware for GPU and DDR RAM?

The RK3588s is a well-known and respected chip in the Linux community, used in projects like the MNT Reform Next. Yes, it needs proprietary firmware for the GPU and DDR RAM — but so do alternatives like Qualcomm, which typically include even more closed components. One of the key advantages of the RK3588s is that it doesn’t have an integrated 5G modem. This lets us isolate the modem physically and control it via a real kill switch—something much harder (if not impossible) with integrated Qualcomm SoCs. We also aim to work toward freeing the DDR firmware, something that's almost unthinkable on Qualcomm due to much stricter restrictions.

On design choices and details: Why does the Nexx have two USB-C ports?

Simply put: you often need to charge your device and connect something else at the same time. But there's a more ambitious reason too—we want to make it easy to use the Nexx as a desktop computer, and two USB-C ports open up a lot of possibilities, especially with docks or keyboards. And let’s be honest… it just looks awesome!

Does one (or both) of the USB-C ports support DisplayPort Alt Mode, so that I can plug into a display without carrying accessories?

Yes, at least one of the ports will provide direct DisplayPort output, allowing you to use the Nexx as a desktop without additional docks.

What are the transfer speeds that can be expected from the USB ports (USB 2.0, 3.x)? What are your plans regarding USB Power Delivery support?

Both ports will be USB 3.1. We’re also working on implementing support for USB Power Delivery, with proper energy delivery for fast charging and accessory compatibility.

Regarding WiFi/BT and LTE/5G: Purism made the unusual choice of making these socketed and exchangeable; I presume this is not the case with your design, the hardware kill switches ensure that components can be truly turned of. Sadly, there don't seem to be any blob-free choices. What are the main criteria to pick hardware here?

Our priority was balancing performance and Linux compatibility. The 5G modem will be replaceable—it’s mounted on a small FPC board, making it easy to swap without the bulk of traditional sockets. Wi-Fi and Bluetooth are soldered directly to the board, but both are connected to physical kill switches that cut power entirely, ensuring they’re really off. We know there are no fully free options yet, but we’ve chosen components with good Linux support and low power usage, minimizing blobs without sacrificing usability.

Battery life has been an issue with the existing native Linux phones. How do you plan to optimize for this?

It’s one of our top priorities. We’re developing a dedicated daemon to manage power more efficiently—it will shut down and wake up CPU cores depending on load, adjust frequencies dynamically, and optimize idle performance. We’re also working on smarter suspend and resume behavior, including options like RTC wake, wake-on-WAN, power button, or even tapto-wake. Brightness management will also adapt more intelligently to real usage conditions. All this aims to extend battery life without compromising the experience.

Some important phone features are set by hardware choices. Unfortunately, Voice over LTE has become mandatory in some markets (and the number of markets, where this is the case, will only grow) - is it a major deciding factor when choosing the 5G part?

Yes, we considered that. We want the Nexx to feature modern hardware compatible with the latest technologies. VoLTE is increasingly essential, and we see it as key to ensuring a good user experience and long-term compatibility.

Aside from necessary certifications, you are also going for a OSHW certification - does this mean Liberux Nexx will be Open Source Hardware?

Yes, that’s our goal. We want the Nexx to be open source hardware, which is why we’ll release the schematics with the first units shipped. Other assets (like full mechanical designs and manufacturing files) might come a bit later. Our initial idea was to release everything once we hit about 10,000 units sold. What matters most to us is building something open and transparent—even if we have to take it step by step.

Accessories. Maybe this is just my age (and growing grumpiness) showing, but - and especially from my experience with existing Linux Phones - why the Wireless Dock (W-Dock) for the "use phone as a PC" use case? I would have rather gone with a nice USB dock that has a silent fan to keep the phone cool to not suffer from throttling.

We totally get that! In fact, we’re not ruling out that more traditional option. But Pedro had already developed an interesting solution to access the device remotely via RDP, which opened the door for us to explore a wireless dock. The main advantage is that you can keep using the phone for things like calls without unplugging it. Still, we know a USB dock with passive or silent cooling has real value, and it’s definitely on the table as a future accessory.

The Liberux Mechanical Keyboard sounds interesting. Any idea on pricing? Will it be available separately?

Yes, we plan to offer it separately—even in a black version—and it will likely be under €200 (excluding taxes). It’s inspired by the legendary IBM Model F, but in a more compact format. We even want to implement the buckling spring mechanism those keyboards had almost 50 years ago, since we think it still outperforms any modern keyboard. The goal is to deliver that unmistakable mechanical experience in a more portable and modern design.

Let's tackle software. You write that "LiberuxOS (based on Debian 13 Linux) is an ethical and mostly opensource operating system." Which parts will be non-open-source? Just firmware, or also drivers / custom user space software (e.g., apps)?

Our goal is to release everything we can. There will be some unavoidable blobs, like firmware for the GPU or modem, but all apps we develop will be open source. We won’t include closed software—either ours or from third parties. Any new developments, such as interfaces or tools, will be published from day one.

Are you intending to ship a (close to) mainline kernel, or a Board Support Package (BSP)/vendor kernel and make it work with a libhybris/Halium approach?

We’ll go with bare-metal Linux—no Halium, no libhybris. We want to stay as close to mainline as possible and actively contribute upstream.

Do you intend generally develop software bits in the open where possible or just where required by license?

We’ll develop everything openly. Our policy is to publish code, collaborate with the community, and be transparent. Free software, for us, is a matter of principle—not just legal compliance.

_Why go with a distribution of your own and not partner with existing projects (e.g., postmarketOS, Mobian, Ubuntu Touch, Sailfish OS)? Given that you are basing on Debian, a collaboration with Mobian (or, if vendor kernel Droidian) may lead to obvious synergies. Are you exploring this?

Yes, we’ve considered it—especially Mobian, since we share a base. But building our own distro lets us better tailor the system to the hardware and to the kind of use we want to promote. That said, it’s not mutually exclusive: we want to make porting other distros as easy as possible and collaborate with anyone interested.

Why GNOME Shell Mobile and not Phosh? In my experience, Phosh is better at dealing the occasional app/dialog that's not quite ready for mobile.

We considered Phosh. But GNOME Shell Mobile gives us more room long-term in terms of customization, animations, and graphical performance. That said, we’re doing significant work on top of GNOME Shell to make it more mobile-friendly. Phosh is more mature in some areas, but we’re betting on something we can evolve more freely.

Which parts of GNOME Shell Mobile do you intend to improve on for the Liberux Nexx?

We’re working on better window management, visual improvements, and a more refined touch experience with intuitive gestures. We want using the Nexx to feel smooth and pleasant from the very first boot.

Since I've attended [Akademy]( last year: Why not Plasma Mobile?

We think Plasma Mobile is a great option and respect it a lot, but for our goals, it didn’t align as well. Given our limited resources, we had to focus on a single interface, and GNOME Shell Mobile fits our vision better, even if it also needs polishing.

Are you in talks with community members/known Linux Mobile developers?

Yes, we’re already in touch with various developers and projects.

[I have asked a follow-up and known-to-me names have been mentioned, but to preserve privacy, these names are not included here.]

Are there ways the community/interested people can help you make the Liberux Nexx happen, despite ordering on Indiegogo (especially for those, that would like to get the phone, but can't afford it right now)?

Any help is welcome: from spreading the word to contributing code. We also accept donations so people who can’t afford the phone can still support the project.

We want this to be a community effort—not just our own.

More questions? Let me know!

I'd like to thank Liberux again for answering my questions. If you have more questions, let me know as a reply to this post on the fediverse or via email - or ask Liberux directly. Thank you for reading!