Quill: a Qt eBook reader and Linux OS for Kobo e-readers

An OS, not just a reader app

Quill is described as an open-source, Qt-based eBook reader for Kobo devices, and it also comes as an entire native open-source OS, with the rootfs kept in a separate Quill-OS/rootfs repository, providing support for a number of devices. That distinction is the whole point. A Kobo normally runs a closed firmware from the manufacturer, and the reading application is part of that firmware. Quill replaces the firmware, so the reader, the browser, the file manager, the update mechanism and the security policy are all part of an open system rather than a single swapped-in application. The feature list reflects that scope: a full integrated KoBox X11 subsystem, which is the Kobo's original X11 stack, ePUB, PDF, picture and plain text display, the muPDF rendering engine for ePUBs and PDFs, Wi-Fi support with the NetSurf browser, encrypted storage with EncFS, a fast dictionary and local storage search, dark mode, ten built-in fonts, enhanced power management, a lock screen and passcode option, a factory reset option, a seamless update process and a VNC viewer app.

Signed software only, and the kernel is yours to change

One feature deserves its own reading because it is unusual for a consumer device project. The security policy is strict in that only signed software can be run on the device, and the README notes this can be adapted to your needs by recompiling the kernel, with the kernel hosted in its own Quill-OS/kernel repository. That is a coherent stance for a project whose whole appeal is that you are running open software on your own hardware, and it is also the opposite of what a typical firmware modification does. Most custom firmware for these devices exists to remove restrictions, and here the restriction is deliberately present and documented as a property. The tradeoff is that the kernel, the rootfs and the applications have to be kept in step, and the repository structure with a separate kernel and rootfs repository reflects exactly that.

Flashing it is a dd away, and you can brick the device

The installation instructions are the part with real consequences, and the README is direct about the risk. Quill OS does not provide any warranty at all and the maintainers are not responsible if you brick your device with one of the factory images or update bundles available online, with a pointer to LEGAL.md for details. The precompiled OS and binaries are downloadable, and standard Nickel add-ons are also available, though support for them has been stopped until further notice. On Windows you use Rufus to flash the image to the Kobo's SD card. On Linux and macOS you find the device node with fdisk -l, or diskutil list on macOS, and then write the image:

xzcat inkbox.xz | dd of=/dev/mmcblk0

The maintainer's own advice is blunt about the older Kobo path. Installing this old Quill requires expert Linux and compiling skills and a lot of time because it is a mess, buying a new Kobo e-reader for it is pointless, and they propose the PineNote instead.

Upgrades are file copies onto the USB partition

The update mechanism is unusually simple, and simplicity here is a risk as much as a feature. To upgrade, you copy the files for your device onto the .inkbox folder in the USB storage partition. The README notes a historical wrinkle: prior to version 1.6, Quill, then called InkBox, had to be upgraded one version at a time, and if your device's factory image ships with version 1.5 you need to follow the upgrade instructions for 1.6 specifically. If you install the Nickel add-on, you unpack the base archive in the root of the onboard storage rather than in a subfolder inside it. The seamless update process listed in the features is this: put files on the SD card, reboot. For a device that has no package manager and no recovery partition, that is a defensible design, and it is also the reason the project warns about bricking, because an interrupted update has no obvious recovery path.

The project has moved to PineNote and says so

The warning at the top of the README is the most important thing for anyone reading this today. The recently released Kobo Libra Colour uses Secure Boot, which makes it very difficult to port to Quill OS, with two MobileRead forum threads linked for detail. Beyond that, the project transitioned to Pine64's PineNote, where the maintainers say they have rewritten the OS from the ground, striking through the word plan, and that you can try it right now, usable but only for advanced users for now. The link goes to a separate PorQ-Pine repository. The long-term statement is careful. They plan to backport new features to Kobos when they reach that point, but this should not be seen as a guarantee, although it would be sad to give up on all those Kobos. The TL;DR is stated without softening: expect no support from us, although you can still ask questions with some hope, and no official updates are planned either for now.

A single-maintainer project with a long tail of hardware

The contribution section explains the situation plainly. The maintainer is a student and motivation has been the only thing that has helped maintain the project for the last few years, and developing an entire operating system alone was not an easy task, with time available but spent wisely. For code contributions, the route is to fork the repository or one of the subprojects the organisation hosts and open a pull request, with the promise to review when there is time, and there are Matrix, Discord and Linen spaces for chat. For people without the coding skills, there are donation routes through PayPal, LiberaPay and Bitcoin. The repository layout shows the scale of what is being maintained: inkbox.pro for the build configuration, libs/ for the bundled libraries, src/ for the source, plus the rootfs, kernel, bootloader and other pieces in their own repositories under the Quill-OS organisation. The license is a custom one that GitHub cannot classify, with the terms in the LICENSE file and a separate LEGAL.md.

When this is the wrong tool

There are clear cases where Quill is not the answer. If you want a good eBook reader without replacing your firmware, Quill is the wrong project: there are existing readers for Kobo and other Android e-readers, and Quill's cost is an OS replacement rather than an app. If you have a Kobo Libra Colour, the Secure Boot situation is a hard blocker for now. If you are not comfortable writing an image to a device node with dd, the install path is a bad fit, because there is no guided installer and a mistake is a bricked e-reader with no warranty behind it. And if you need a system with an active update channel, the Kobo line here is explicitly frozen with no official updates planned. The PineNote rewrite is the live branch and the one the maintainer points people towards, and it is described as usable but only for advanced users for now.

Editorial conclusion

Adopt Quill if you own a Kobo e-reader and want an open reader and an open operating system on it, and you are prepared to flash an SD card and accept that the Kobo line is frozen. Do not adopt it on a Kobo Libra Colour, because that device uses Secure Boot and the maintainer says it is very difficult to port. Before you do anything, read LEGAL.md and understand that Quill OS provides no warranty at all and that you are responsible if you brick the device, and check the PineNote rewrite, which is where the project now lives.

Frequently asked questions

Official sources

1. Issues
2. Quill-OS/quill on GitHub
3. README
4. Releases