News and Update
Florian and Alfred presented.
The frequency of the Q&A broadcasts has reduced sharply but that doesn’t at all mean that activity has ceased.
Ubuntu summit and Ubuntu Touch
Ubuntu Touch OTA-24 and OTA-25
There were lots of good things on the agenda so it was heartbreaking that we had all those delays.
OTA-25 will arrive very soon and will consist almost solely of security fixes for Xenial. Although Xenial is end of life, there are some upgrade packages which can still be applied and we will aim to import all of those to the latest versions for that distribution. It will therefore be quite a big download.
After this, there will be only Focal development. Our aim is to get Focal out around the end of December or perhaps into January. There is no absolute promise, so go easy on us. At this stage, we only have certainty about a small number of devices being able to run it but we are confident that the list will grow. Don’t push us on that at the moment. We don’t yet have a ‘final’ list and we expect the number to grow over time, just as it did for Xenial. The porter of a device will have to do some modifications in preparation for Focal and even if a device is suitable in principle, if the porter has gone awol that device will be lost.
Ubuntu Touch 20.04 news
Alfred is working to ensure that the code fixes work for tablets as well as they do for phones.
This takes us on to ‘laboratory experiments’. Incidentally, if community members are doing something in this category, please let us know and we will invite you onto the show.
Alfred has been playing around with kinetic scrolling for Qt-able applications. This is very important when a touchpad is used. Hopefully those tweaks will appear in both Xenial and Focal.
The second mini project is blurred indicator panels. On Jingpad or similar, indicators are made to appear as a transparent overlay. Of course as soon as you make transparent you end up with grey on grey, which isn’t great for readability. It is what it is though. Alfred will try to implement the feature as a user defined setting so that those who need an emphasis on accessibility are not inconvenienced.
We have translucency on the drawer, so in UI terms it looks patchy if there is solid black on the indicator panel. Consistency looks much more professional and modern. Let us have feedback on your opinion. Does it need to be user defined or should we just do it?
We haven’t had any recent discussion of camera notches and round edges but we have been working on those in the background. Maciek in particular has been playing around with that. The Fairphone 4 and the Volla phone are cases of this. We need to get it right, especially if the time disappears of the right-hand corner!
The News section of our Forum is the best place to post questions for the Q&A. YouTube live chat,Telegram and Matrix are other places to post a question.
If you didn't know, the Forum questions get priority.
Catalyst asked whether it is best now just to develop apps for Focal or is there still some point in building apps for Xenial? If the app is something like QML which can run on any version, your contribution will not be accepted unless it is primarily designed for Focal. Development for existing apps can be undertaken for both platforms if you want and you can adapt new apps to run on Xenial if you want to. The move to Focal will be substantial though so any work on Xenial will benefit a declining minority of users.
Waydroid and Focal
Is there an indication of when Waydroid will arrive for Focal? No idea at all. There is a Waydroid telegram group and you will get a detailed status report there. We will install some enabling packages for Waydroid in the root file system, in preparation. We are already confident that we can overcome the problem of having to install Waydroid from scratch after every system update.
Nextcloud online accounts
Can Nextcloud be added to online accounts in Settings for Focal? Right now, there is no integration of any online accounts in Focal. It is something we will have to work on.
Xenial extended support
LNG asked how the integration of extended support for support for Xenial (courtesy of Canonical) is going? As explained above, the new updates will be added in OTA-25. It shouldn’t have any implications for app developers. It will just mean that the platform they are developing for is more secure than before.
Devices promoted list
Electrodate asks whether more devices will be added to the ‘promoted’ list? Volla have three devices now and we have a very good relationship with them. We hope very much that they will continue to produce new devices which run very well on UT.
We talk actively with Fairphone about FP4 At this stage those discussions are obviously about Focal. Upgrading from a Xenial build would be a mess and a waste of resources. Certainly our plan is that you will eventually be able to buy a FP4 from their store with UT already installed.
As we have said before, Xenial is an absolute deal breaker for any potential commercial partners and with Focal we hope to bring new partners on board.
Fairphone 4: finally the computer in your pocket?
ConvergenceUser said that since 2015 they have been a fan of the idea that you can carry around a small computer (a phone) in your pocket and then convert it into a full computer by plugging it into a large screen. Despite caution, they are getting quite excited about Fairphone 4. Will that be the device to finally deliver on the promise? Are there shortcomings which are not obvious? Well FP4 has the right HDMI hardware which is a great start. You should not start out with the idea that it will be ‘just like a normal desktop’. You will have to refine which apps you work with and you will have to learn to adapt to using them in a different way. Think of it as learning a set of new skills.
A tablet alternative would be great but there is no real replacement as yet for the dying Jingpad.
FP4 functions at the moment only on Devel channel. That is a bit frustrating because it is pretty close to being daily driver material. The Devel channel breaks of course from time to time, so from that point of view is off limits for daily driving. It is too early to talk about a Stable channel but pretty soon we can start thinking about an RC. Florian has one and will try it for a few weeks as daily driver and see how it goes.
The hardware offers 5G and eSIM. We don’t have support for either of those now. We will certainly use this device as a test-bed for developing those features. 5G is not much of a technological challenge. It is more of a UI issue, letting users know that they are using it and enabling them to make a choice. Florian is waiting for an Austrian provider to give him an eSIM so that he can play around with it and see what needs to be done.
On top of those issues, FP4 has multiple cameras and we want to get switchable camera selection into Camera app. Again, FP4 is the perfect test-bed.
Of course most devices don’t have HDMI but Alfred is trying to extend the ethercast option to more devices, starting with Volla phones. A team to do this would be great and since Alfred himself has limited time, maybe he could direct the work on that?
Alfred was asked about impressions of his Jingpad setup at the Ubuntu Summit but in fact he left that behind. Perhaps next time. Maybe his presentation will revolve around the Fairphone 4 next time? There is a core group of people who are still very enthusiastic about tablets but the market is being squeezed and there are fewer of them.
FOSDEM is at the beginning of February next year and we are planning to be there again. This year there will be a dedicated mobile devices dev room, so we will have a formal presence.
Native Linux apps on Ubuntu Touch
Vlad asked in live chat whether some ‘native Linux apps’ will make it into UT? Well in one sense all of our apps are native Linux apps but what the question is hinting at is ‘desktop apps’. The headline apps of that kind are LibreOffice and Gimp and the obvious issue there is the small screen. The core team will always have the mobile app experience as its first priority. If you want to extract features from ‘big’ apps and scale them to run well on a phone you can do that. You can also adapt those apps so that they are hosted on the phone but run smoothly in convergence mode on a big screen. Just don’t expect the core team to be much involved in that.
OTA testings process
Messaisto guessed that the testing phase didn’t go very well this time round? Is there anything that can be done to make OTA testing easier, to draw more people into the process? One big thing we learned is that all major push requests should come with a prepared test regime, designed to see whether they work as planned. The test procedure should not be an afterthought. If you try to work out a test regime later on you will have forgotten a lot of the details. Having the test procedure ready to go should be part of the acceptance process. We need to tighten up on this.
S60W79 asks if cell broadcasting is supported? It isn’t supported in developer version, in terms of UI and interface etc. At a low level though it has always been there, just not implemented. LionelD has been looking into it for some time. There is a cell broadcasting test in Germany on 8 December and we will do some tests there to see if the prototype solution works. Don’t expect anything slick and integrated yet but we may get to the point soon where you will be able to receive some sort of notification and then go digging in logfiles to see what the message was. If anyone has advance notice of tests in their country, maybe we can use that opportunity too.
Core apps OS integration
A question seemed to ask whether the list of ‘core apps’ could be changed when we move to Focal? The obvious candidate would be to make Dekko part of the system. Well the thing is that an email app is really a huge thing. Like a browser, it is almost as big as the operating system itself. The idea of ‘core app’ implies some commitment by the core team to maintain it. There are already examples of much smaller core apps which currently don’t have a maintainer. It would be dishonest to give the impression that we are able to devote a great amount of energy to Dekko and be held accountable for that when the reality is that we just don’t have sufficient resources.
A maintainer is more of a mechanic and organizer than a designer, so someone with lower level skills may be able to take on that work and we always welcome people to fill those gaps. We need maintainers for around half of our core apps. We have enough to deal with in the 20.04 change. Yes this issue matters but we should look at it when things are less busy.
Someone asked whether UT can run on devices which are based on Android 11. Yes is the simple answer because FP4 does. Generally with porting, don’t pick below a 13.8 kernel and go for 18.1 rather an 16.0. It is a bit late now to be playing round even with Android 9. We need to be keeping more up to date. We are switching to System D and if you don’t have compatibility you will be left behind. Patching kernels is possible in theory but huge patches are honestly not worth considering. That is why 3.18 is the cut-off point. Anything lower is pointless. We can now say that with certainty.
If you are well advanced with a device which will only work with Xenial it may be worth finishing the job as it will be usable for a while. If you are in the early stages, abandon the port and find something newer to work on.
A user asked whether it is possible to port a device with a Mediatek processor. Volla phones are so in principle, yes it is possible. It used to be a much bigger problem with closed binders but these days our fixes generally ‘just work’.
Alfred has been doing some Halium work on the daemon which is used to implement some of the functions in Mediatek (or non-Qualcomm) chips, such as wifi switching and ethercast mode, peer-to-peer communications etc.
Battery usage information has quite a lot of problems and there is a desire to fix those. App development procedures will change a bit under Focal and we will need to provide new documentation for that. Lots of tasks are building up.
In other news, the OpenStore is now capable of hosting apps for both Xenial and Focal. It is already possible to upload two versions. We will ensure that devices only see the set which is appropriate for them. Inevitably there will be quite a lot of apps which were available for Xenial which will not carry over into Focal.
If you update your system to Focal your old apps will still appear as icons and the software and date will still be present but if you try to start any of those apps they will crash. We will need some sort of process to take care of that. In effect it will be a clean break and a new phone. You will have to do backups where possible. It will be anything but seamless.
The right manufacturer for Linux phone
Amarnath asked what a manufacturer would do if they wanted to support the adoption of Linux on their phone? A very simple answer. Provide open source drivers! Florian commented on a rumor that the big reason behind closed source is that it hides the evidence of theft from competitors…
Hardware manufacturers should stick to what they are good at which is building hardware. They shouldn’t try to extend themselves into software development where they truthfully have limited skills. Drivers yes but operating system elements no.
Google imposes an absolute timetable so manufacturers rush to the exact deadline, sometimes with quite trashy results.
Quick note Q&A sessions
Finally, we would like to throw open an invitation to someone to lead on getting out Q&A sessions, so if you would like to be a presenter, let us know. A once a month slot might be better than the fortnightly session we used to have. Bring your ideas too.
See you next time :-)