COMMUNITY UPDATE 20
XENIAL, ANBOX, SIGNAL & MORE
[00:00:00] Marius' introduction and holiday greetings.
[00:01:38] Xenial (16.04) Updates
[00:04:15] Anbox Updates
[00:06:53] Anbox Demo
[00:08:40] Signal App Development
[00:09:50] Questions --
[00:16:11] Unity 8 on the Desktop
[00:17:58] Snap package support
[00:23:34] GPS Improvements
[00:25:15] QT browser?
[00:29:23] Spectre and Meltdown
[00:35:30] Ubuntu Touch on Desktop
[00:37:15] Anbox Integration
[00:43:45] Overall project impressions
[00:48:00] Experimental Channels
[00:49:50] Ubuntu Store to OpenStore migration
[00:51:30] Ubuntu Touch Kickstarter
[01:00:00] Deep Tree Thoughts
# Holiday Cheer
Happy holidays! A lot happend during the holiday season. Most of this can be seen on Popescu Sorin [@popescu_sorin], who has been showing a lot of this work through [videos on G+](https://plus.google.com/+PopescuSorin)
There has been a lot of progress with Xenial (16.04). We have made [tools](https://github.com/ubports/unity8-desktop-install-tools) available to install our software stack on any Ubuntu 16.04 computer. This is a step forward for the project and convergence. The desktop version and mobile version are almost identical now with the same versions of Unity 8, Mir, and some of the lower stacks.
Anbox installation instructions were promised over the holdays, but due to personal obligations, progress on this was slowed. The development team have been working on it even right up to this community update.
Installation instructions are now at [UBports Docs](https://docs.ubports.com/en/latest/userguide/dailyuse/anbox.html]
It is to be noted that Anbox currently works only on select devices (listed in the docus) running 16.04. Thus, it is meant only for testing by experienced users in its current state. Neithe Anbox nor 16.04 are ready as a daily driver. Anbox requires a kernel change, which has not been completely tested yet. This is so that the "Android Binder" will work. Eventually there will be a "two-binder" system so that the Anbox and Ubuntu Touch parts don't interfere with each other.
Marius demonstrated how to start Anbox once installed. It runs as a contain, but requires manual starting from the icon in the apps scope. It takes quite a while to start up.
Eventually it will be integrated into the main menu, but for now you must start it as you would an application.
# Signal App
The Signal app for Ubuntu Touch is in active development again thanks to Aaron Kimmig who has picked up the project. We are also thankful that [Open Whisper Systems](https://signal.org/), the creators of Signal Messenger, have said they are open for collaboration. Hopefully this means that we might get push notifications for the Signal app in the future.
# The difference between Unity 8 and Yunit
Our version of Unity 8 and Yunit are two separate forks (copies) of the Unity 8 that was under developmen by Canonical.
Yunit has become a lot less active recently and there has not been much contact with them since the very beginning. Of course we would like to collaborate, but we need to keep things moving. Ubuntu Touch requires Unity 8 to move forward, so we are developing Unity 8 on our own. Even so, our version of Unity 8 for the phones is a little bit older that the desktop version at the moment.
Xenial (16.04) for the phone might not get a stable version of Unity 8 (as shipped in the Ubuntu 16.10 and 17.04 releases). An unstable version is avaiable for testing in our repositories.
# How to get Ubports Unity 8 on Desktop
We have a [Github repository](https://github.com/ubports/unity8-desktop-install-tools) for Unity 8, but it's not stable, so use it in a virtual machin rather than your actual devices.
There is no snap version of Unity 8 yet. It is easier and feasible to use deb packages right now, so that's what we're currently focusing on.
# Snap package support
Snap packages are intended to be supported soon on the M10 Tablets and Pro 5 device.
# Will GPS be improved?
GPS is extremely difficult to work with and it is not currently a high priority. That being said, both Xenial and Halium will show great improvement in GPS.
# What about using Qt Browser?
Qt application support is available in Unity 8. However, we don't have Mir support in the Qt web engineg for the phone yet. We are looking for people help us migrate from Oxide to Qt web engine eventually. Oxide os based on Chromium currently, so it will take time for us to make a switch.
Our goal is to provide more options for applications (like the browser), but it is not a high priority at the moment.
# Fixes for Spectre and Meltdown
Meltdown only affects Intel chips, so no Ubuntu Touch devices are affected by this.
We can patch Chromium, but until LineageOS patches their kernels, we don't have a real solution. It is a very difficult and time-consuming task that we will not be able to focus on.
The reality of this vulnerability being exploited in an Ubuntu Touch device is quite low. An attacker would need to install packages on the device itself, and this is not very likely. Servers are more vulnerable and easier targets.
The Ubuntu desktop patch is [scheduled for release](https://insights.ubuntu.com/2018/01/04/ubuntu-updates-for-the-meltdown-spectre-vulnerabilities/) around January 9th.
# Ubuntu "Touch" Desktop Version by UBports
We would like to create a version of the OS for the desktop but we don't currently have the experience to do this. We need someone who is experienced with ISO building for desktop.
# Anbox integration and WiFi Control
Anbox will not be able to control WiFi. It will connect to the Internet, but it won't know any details about the connection. This is on purpose to maintain security and keep the Android parts contained. Eventually we may need Bluetooth control, but as we develop we will keep security in mind. Ultimately the user should have control over what Anbox can or cannot access.
# Is Halium used in 16.04 images?
None of the existing devices will run Halium. It will be used with new devices. Dalton is doing [a lot of work](https://github.com/ubports/ubuntu-touch/issues/404) to boot Ubuntu Touch on vanilla Haium. In short, it's not here yet.
# X11 Apps
We would like a lot more X11 apps to be avaible on Xenial. Xmir will make this a lot easier. The Libertine container was difficult to worth with, but we want it to work.
# Are you happy with the project overall?
We are quite happy and we are getting better all the time. We have made a lot of progress in self-hosted documentation, organization, and with the foundation. We are grateful that Canonical kept their servers alive for so long, but now we are completly running on our own servers.
It's unbelievable how fast things are going, but we are excited. Even some former Canonical employess have been helping out in their spare time.
# Is UBports presenting as FOSDEM?
Unfortunately we didn't get accepted into any of the talks at [FOSDEM 2018](https://fosdem.org/2018/). Some of our developers might attend, but otherwise we will not have an official presence.
# Experimental Channels
"Experimental" channels have been moved to Community Channels, where people can make their own channels and deliver releases via our infrastructure. Soon we will try to get this up.
# Ubuntu Store to OpenStore migration
Some apps have been imported from the defunct Ubuntu Store into the OpenStore. There has not been a lot of effort lately because we don't want to flood the OpenStore with old, unmaintained apps or unnecessary webapps.
# How about Kickstarting a Phone like Purism did?
Right now we have no plans for that. We want to focus on making our software good before making a product out of it. We have a small team, and our development time is better focused on the software, as we are not hardware developers. If we can grow the team in a healthy way, then we may consider a crowd-funded campaign, but creating a phone is very challenging, and not all problems can be solved just by throwing money at them. We want to maintain high quality and not try to hack together something like many Kickstarted campaigns do.
In order to make a new phone, you have to design it yourself. It's not like laptops where you can just select and ODM model and ship within a year.
We hope that we can get Ubuntu Touch onto the Librem phone if they succeed. $2M is a slim budget. There is a reason why the Ubuntu Edge project had such a gigantic goal. It's not easy.
# What about uMatriks?
Marius plans to get back to the "mario griptest"--an updated and improved version of uMatriks for communicating on the [matrix](https://matrix.org/) network. But, this work require more time outside of UBports development.
# If you were a tree, what tree would you be?
Jan: A [balanced binary tree](https://en.wikipedia.org/wiki/Self-balancing_binary_search_tree)
Marius: It's Christmas. Pine tree. I want to have a lot of prickles... err needles.