Ubuntu Touch Q&A 63
ARM64 merge into Edge channel


Looking for the Audio-only version of the Q&A?

You're just one big orange button away -->

News and Update

Easier newer devices porting

This week were present Marius, Florian and special guest Jörg Wurzer from Volla Phone.

At the beginning Marius explained that the ARM64 version of libhybris has now been merged into Edge; this will make porting to newer devices easier.

It also helps with the Raspberry Pi build. That matters a lot because it will help create a true development platform for UT, for the first time!.

Florian: Our Raspberry Pi community is still very small but I encourage to join them because as Marius said: this is very good development port.

New MIR progress

Marius also explained that new Mir can run on Wayland using the Wayland protocol. One server is running as root and another is running in the user session; this allows the user session to be suspended, which saves battery. That separation also improves privacy and security.

New bluetooth support update coming?

Florian announced better support for Bluetooth headsets in the works with an upgrade to bluez and especially a patch for Pulseaudio.

Raspberry Pi 3 running UT

Ubuntu Touch was shown running on a Raspberry Pi 3 with a 7" touch screen. Raspberry Pis are really maturing as development platforms and will make access to Ubuntu Touch application development possible for a much wider base.

Volla Phone Kickstarter

Jörg Wurzer introduced himself and explained that he is the founder of Hallo Welt Systeme UG, the company behind Volla.

He set out his aim to create a phone for freedom, with simplicity and security. He noted that modern smartphones operating systems consume a lot of people’s time and attention and he wants to provide an alternative that allows people to focus on the things that matter instead of being interrupted by notifications or losing time searching for apps or calling up hidden features.

It is important that we should decide for ourselves with whom we share our data.

Jörg Wurzer also commented that at the moment there aren’t any alternatives which consumers can go to a supermarket and buy, letting them protect their privacy and focus on life, right out of the box.

Therefore, Volla has started a Kickstarter to pay for making that possible. A small international team of twelve people has already done the groundwork in creating a new user experience which Volla would like to provide.

The starting point is a system which is based on Android Open Source Project but, to ensure this freedom by simplicity and security model, some alternative operating systems are necessary.

For this purpose, Volla thinks that Ubuntu Touch is one of the best candidates and Volla appreciates the support of the community and the foundation in discussions before the project started. Although there has already been some investment in prototypes for Sailfish, what is more important for this Q&A is Ubuntu Touch.

Sponsors were thanked.


The News section of the Forum is the best place to pose questions for the Q&A. YouTube live chat, Telegram and Matrix are other places to post a question.

What about Focal?

Gizmochicken has noticed that the Bionic branch of Unity 8 is a hundred commits behind Xenial. What is happening about Focal, which is the next LTS build after Bionic?

Well, Bionic uses systemd to handle applications, in place of upstart. Limitations of time also mean that we have not been able to put much work into the desktop implementation.

As for Focal, we are holding back a bit because we will probably be handling a newer version of Qt for that, as well as other modules. We don’t want to waste effort adapting to module versions which may not end up being used in Focal. If there are not going to be major changes, we can start slowly after Christmas with those elements which will not be impacted much by any upgrades.

If Canonical abandons armhf we will have even more work to do ourselves. In the newer version of MESA we won’t have Mir support anymore, so that will rely on a Wayland implementation. That has to be done on Xenial first, before we can move to Focal. Basically, we will not make a switch to Focal until that platform is mature and established.

A switch of LTS will mean that some older devices can no longer be supported. That is about kernel issues and the lack of security patches. So a bit of delay helps users who are not yet able to get a newer device.

Banking apps?

Momaksoks said that the ‘one swipe away’ function of UT combined with the freedom and security makes Ubuntu Touch a dream! In Sweden though he needs a special banking app, which is not available on UT. Is Anbox any nearer so that the Android app can run?

Lots of users have the same issues, especially around banking and transport. Marius explained that in Norway they use a SIM toolkit, not something embedded on the phone. UT can do the same natively, as it is just SIM dependent. Marius has been playing with it at a demo level and expects it to become more refined.

Anbox is a sandboxed environment, so one of the challenges is creating the hardware layers which enable the actions that normally would run natively in Android.

Jorg Würzer confirmed that in focus groups involving phone users from 17-80 almost all had just one or two Android apps that they really needed. That is why Volla started their development from Android One. That exposes a political problem, since those essential apps are not being offered for all platforms. So some lobbying is needed. That applies particularly to the delivery of public services.

Marius gave a major example of this from Norway, where the plan is to build driving licences into just Android and iOS apps, creating a monopoly for those platforms.

There is a growing awareness in the European Union of privacy and security issues and a recognition that at the moment all of our data ends up in the USA. So there is a logic to having a phone system which would allow in principle for Europe-centered privacy controls.

UBports cloud server?

Someone asked whether Ubports had ever thought about setting up a Cloud server as a side project? What if that was offered as a paid service?

Ubports has never intended to do such a thing. Setting up a ‘cloud’ on a local server is an easy thing to do but scaling that up to a large scale, with promotions, pricing and contracts would be a highly complicated undertaking. There is a huge responsibility attached and regulatory controls on data handling.

Central servers are in any case insecure by design because authorities can require access to the data they hold – as has happened recently with Tutanota.

Solutions need to be found by using distributed systems!

Stable Teleports release?

There was a question about Teleports and whether there is a date for a ‘stable’ release. Rather than a date, Florian mentioned the features which need to be present in a ‘finished’ version. Sticker sending needs to be there, along with basic group administration – creation, adding and removing users etc.

Is libhybris needed for Volla?

Gizmochicken asked whether libhybris will be needed for running UT on the Volla phone? The answer from Jorg Würzer was yes, since it is not possible to recompile the Android drivers to run natively with UT. Hopefully over time with the help of Volla’s partner Gigaset, it will be possible to remove the Android elements and so not need libhybris anymore.

Is libhybris needed for Pinephone?

The second question is whether libhybris will be required for the Pinephone? No, it will not need libhybris. The cpu is non-standard as far as phones are concerned.

It uses an Alwinner chip, which is heavily used in China for tablets. It also uses Mali for graphical drivers. Therefore we have access to some of the upstream which has been tailored for that chip. It supports upstream Linux kernel. We use Lima amd Mesa for graphics stuff.

Libhybris will still be present but it will be there just to detect whether the device is Android based or not.

It will give us vital experience in using Ubuntu Touch with mainline kernel!

Ubuntu Touch Q&A 64
OTA 12 development progress