This Q&A session covers the latest developments in the Ubuntu Touch platform, including the release of OTAA 1.3 with snap package support, ongoing Morph browser porting to Qt6 with Vulkan integration, device porting improvements for Fairphone 5 and other devices, and the community's collaborative development process at the Luminary Code Fest. The discussion highlights the integration of desktop gaming capabilities through Wine Box and GL4ES, the evolution of the Open Store for snap applications, and the ongoing challenges in achieving a stable, user-friendly mobile experience.
Approfondir
Prérequis
- Pas de données disponibles.
Prochaines étapes
- Pas de données disponibles.
Approfondir
Ubuntu Touch Q&A 190Indexé :
No description provided by the creator.
Hello, we are live. I see. And welcome everyone to the Ubuntu Touch Q&A episode 190 streamed live on May the 16th, 2026.
My name is Alfred and joining me tonight because we are uh at a very special place right now uh gathering together.
You would not believe it if you were not here by yourself. Uh very impressive day and very productive day I see. And uh maybe we should just >> turn around.
>> Turn around the camera what we have here.
Hello everyone.
Hello everyone.
So, this is a very nice atmosphere that we've been having over the whole day.
Uh, some people brought their laptops. They started hack hacking right away on whatever they came up with. And first and foremost I think we should start at what this is because this is uh an event organized by OS science open source science and open source science uh have been hosting this lumary code fest here uh here in the Netherlands in Tber and yeah we've been taking all the rooms available at our disposal and filled them with people who were interested in hacking on Ubuntouch apps on Lumier like native looking apps and yeah was a very was a very productive day I would say I don't know how how was your feeling about >> it's it's it's really wonderful it's my first maybe not first time but it's rare to see so many people who interested in model touch and what we are doing and rising apps.
>> Oh yeah definitely.
>> So uh yeah let's start with uh have we should we just this guy is preparing here. Hello.
>> Hello everyone. Before we forget somebody here.
Uh this is Mike Mike Gabriel.
>> Sorry you get my joining request or you mostly know me as not >> perfect.
It's time to It's about [ __ ] time. We have Okay, that was We have It's about time.
Excuse me.
It's about time. We have a face to the name. This is the guy responsible for a lot of things happening in the porting area and integrating a touch into various different devices. So thanks to him we have a lot of things working the way they are. So and they work well.
Okay we have somebody joining.
Okay.
>> So, >> yeah.
>> Okay. So, we use one microphone on Alfred's notebook, >> which is the phone.
>> Which is a phone.
>> But it doesn't matter anymore because it's all portrait now. We are three people splitting it into three and it's portrait now for everybody and nobody notices.
Fairphone 5 wins.
Yeah, Alfred is actually streaming this from pair five connected to max dock.
>> Yes.
>> Right. If if someone doesn't really talk, >> this is the there there's a magnetically attached Fairphone 5 uh attached to the next dock and USB C connected to the next stock uh to provide the video capabilities and the keyboard and the touchpad.
And this is my daily driver setup. This is what I work with. This is what I compile my Ubuntu touch packages on in Lexi, for example. and install them and test them and work on the platform and it works. So I'm happy about it.
All right. So we should start with a little bit of news uh regarding our uh request for or call for people coming in as Ubart's members uh officially people need to if you're already a member please consider nominating yourself to become uh somebody in the membership committee or in the board of directors uh as people working in specific areas.
And the last one is the trustees. So or the first one is the trustees. So you become a trustee. You have certain meetings to to join. You have your chance to voice your opinions. Uh you have chance to organize a few things. So this is roughly what you can expect from being part of the whole um democratic system of UB ports as a foundation.
And yeah, so that's that.
Next one is OTAA42404-1.3 has been released. So, we have a great new update released for people out there who are interested in running Ubuntu touch on their devices.
The newest hotness has been released in the 2404 department. This is 1.3.
This is a bug fix maintenance release.
uh mostly stuff trickling down from main development that we decided to pick into the 1.x the X branch and this is what we release and it also includes the ability for pseudoless installation of snap packages for example or doing privileged system administration or system tasks via policy kit.
And the great thing about that is because I'm going to spoil the means right right now right here uh spill the beans uh and that is the open store has recently been uh contributed to uh the ability to have snaps and it's been merged overnight.
So this is in now this is in the code of the open store. So you can expect one of the upcoming releases of the open store uh bringing in the ability to install snaps by searching for snap packages uh right on your phone right on the device. Doesn't matter if it's a command line snap, if it's a uh desktop snap, if it's some like scum VM. I've seen I've tested it works uh some SDL applications and OpenGES applications way apps uh Audacity works and I also want to improve on Audacity with regards to the microphone support.
So that's not completely finished but I hope to get something working really quickly.
>> So will will all the snap apps blend into the click apps and mix with them or how will it look >> in the permission system? Yes. So in the permission system actually we merged that uh in system settings where the necessary change was to to just check for desktop files in the path where snaps usually put them and that one was enough to just and also patch to pulse audio uh so that the trust prompt that pops up shows up above Audacity like animates on top of it and then you just allow or deny the access to the microphone and that stores it in the list that you can access in the system settings.
>> Yeah. But how will things appear in open store io? So so will you be able to install snaps directly via open store.io and open store offers a list somehow of available snap apps or I >> I don't know what what Brian has planned completely. What I can tell is that the data is scraped off of the snap store into the open store which means it can be displayed in any way we desire or Brian desires. So if you have something special coming up or in in mind here.
>> Yeah. So, so the next question basically I have if if there's actually a method to filter out like to so so so because in in in the snap store there are a lot of apps that are not really well usable I'd say on a phone because they're not convergent or not phone optimized. So if there is some filtering capacity or some some mechanism that that's sort of coming up in in the next release or in one of the next releases >> so that that there's a pre-selection because there's just a really a whole lot of snaps. Maybe not everything is really helpful and open to touch. I don't know. Just an idea.
>> Uh, doesn't open already have a feature for featured apps?
>> Yeah, featured apps. But featured apps are only showing clicks at by default.
So, they're not intermixed. They're not intertwined. Uh, no mixing of clicks and snaps there, but they're mixed into the installed apps and in the search results.
uh so that you can have clicks which mostly show up at the top and probably snaps coming coming up later. Uh that works pretty well.
>> It just would be nice to have some kind of curated lists of snaps that really work well.
>> Oh yeah. I mean, I would definitely put the Android deu uh the Android platform tools into that list because that's how I de debug the Pixel 3a Holium 12 port from the Fairphone 5 that tell from something that I packaged for the snap store which I released as new dash inet tools or inet utils. Uh and you just have talet in that package. You just open talet and that's what I needed for debugging the inits and yeah now we have everything that we need for debugging in snap store if it's not in the click store.
>> Mhm.
>> So we are finally selfhosted.
>> We safe hosted. Yeah.
Compiling kernels on the device is also possible. Slow but possible.
>> Yeah. I I remember you showed it on JPEG before. I I I I tweaked the process a little bit. I tried it with fax and it's way more accurate to execute the executables, the the Linux binaries, GCC client binaries from Google uh for the Android kernel. those start but slow and it probably took like already like 20 minutes 30 minutes to get to the point where it's saying okay I I saved the last config in the kernel object >> that's slow >> that's slow >> so it's slower than GitLab CI >> right so so I wonder I wonder if we before we dive into the really low levels of tech here um so so you you said we are at the OSS ci.com and that is the Lumiere confess and I wonder if we should like honor what's happening has been happening today and maybe also give a little preview of what's happening tomorrow.
>> Yeah.
>> Um so so basically I think we were um at the beginning we were around 50 people is that correct looking around in the audience? Yeah around. So, and um there was there was a short introduction given by Eric Molz who's um well, I guess he's the person who's running ossci.com. It's an educational facility Intel work uh with the focus of teaching people how to do and how to work and how to actually make a living out of open source development and also deployment and whatnot. And um so and and then um as a as a um second talk or presentation uh Sunday Clober actually told us about he gave a really good tutorial in how to write uh Lumiere apps and actually want to touch apps. So, so he explains the whole workflow using clickable and whatnot to to get a default or a a sample example app uh up and running so that you can actually start tinkering with it and then develop your own stuff.
And then there was a lot of space provided and also computers to land and uh uh uh phones to sort of uh have for the day to actually tinker with app development and um and and and and work on one's own app. So that was really really awesome and I actually saw some people who did quite an amazing bit of work considering that it was just a day.
So it's absolutely doable. I learned from that and um so and as a side note so so um um a guy here from NH and uh who's who's also working on uh Lumier core at the moment um we we both looked at okay how is that actually don't I've been hacking a lot on on the Lumiri core but not ever touch nav really. So um and it was really easy to set it up and then of course we diverged a little bit because uh we thought ah QD6 is around the corner. So how does it actually work to sort of what is what needs to be converted in this in this template to actually make a QD5 a QD6 Lumia app. And so one outcome of that is that there is now a merge request for the open totouch templates uh that clickable users. So um once that gets merged that's the first step actually to start clickable development against uh Lumier and QD6 and the other thing is that um Adrian now has the foundation for testing his his Q6 version of QDP pimp. So everything that's related to address book and events and calendars and whatnot which is currently the topic he's working on.
So, so that was actually quite quite a good kickoff and uh an inspection for me personally into something that I don't do during weekdays. So, that was really nice. And my problem was then what app do I want to write? I don't know. I don't need any apps, but that's a different problem. So, yeah. So, so and tomorrow um there will be um some uh there will be some some some talk about risk five. I learned uh I don't have a schedule here and I've forgotten name of the speaker. I'm sorry for that. And also I will be I will do some kind of a state of union about Lumier Core development where we are at the moment and where we not are. Um and I must admit I have forgotten the rest of the schedule but it will probably be really great. Also is we have to mention that there is we that everyone got a swag bag and there were two big books inside and these books um have done by well a lot of people here at lsi.com um Richard Vagis to name one and and also Agisha and I forgot Nigisha's last name so but he's been the graphics designer so in these these books it's a level two and level uh level one level two book uh about lumi development and it really explains everything from scratch and we we worked through the first third of the first book Ariel and I and it's really it's really well explained and every problem you encounter is actually documented there.
It was quite amazing. So um yeah so um big thanks to these people here also there was food and coffee and everything we need for such an event. So it's really well organized and it's I I really advertise if it happens again here or some event happens come come here. It's really good place to to work on open source.
>> I must admit the coffee was particularly good.
>> The coffee was good. Yes.
>> Okay. Nikita had some coffee.
>> Great.
>> Welld deserved coffee for everybody.
Okay, good. Uh do we want to go over to some news coming from GitLab or about >> Yeah, let's start. Huh?
>> Let's see what we have here. Uh I see some change to Lamiri RMDR only if such directory exists in the migrations.
Uh that sounds like preventing it to exit early. Uh Rajan Ran made that change. I'm going to check this.
>> So is that for 2.x right?
>> That is for main. Yeah, that's 2.x X and 2604 1.x uh calling it which can happen in a script to run a new device which have nothing to migrate add the condition.
Okay.
So what else do we have? Uh synchronizing Debian packaging I see uh here for Lumiere UI extras. Um, we have a correctness fix for scaling and calculating the grid units in Lumir UI toolkit apps. Uh, tested on Wayland on on the next stock as the second screen.
uh tested in all the apps in that use way uh not way mute client and also for cute mirror to shell. So that shell scaling also is not affected by it negatively has been verified works and yeah so the next one is what else do we have here the picker fix the position view at index if called on the second day of the month wow okay that one sounds very specific >> yeah it's it's basically we so so we have one developer on the team Henrik and Henrik loves unit tests so it's what he learned at university and what he used to do at work for a long time already and now we hired him and I said, "Oh, that's great. We need someone who knows how to unit test." So, and he's now going through Lumia UI to get QD6 stuff and tries to fix all the unit tests. And that's probably one of those fixes, I guess, but it's amazing. So, um we we were with QD6 with our Numiri components uh QML module, but we didn't have any kind of test coverage yet. So, and this is currently work in progress and um and I guess it we will call it stable when the lumiri components when that when that code repository actually has full test coverage there.
I see a fix up for the kernel of the mopform quintas uh which has a USB USB gadget function uh disables gather skeptic gather uh whatever that specifically means. I would guess it disabled something in the USB gadget kernel module so that it's fixed up. It's by Akali.
Uh you can find it on GitLab uh under the reference device ports uh HGM 13 specifically. You can find the Quintos there listed freshly off the uh process for you.
And I would say let's go to the next page if I can find Oh, right. We have something to complain about with something that we merged.
Uh oh, it's gone.
>> Oh, I'm scared.
>> Yeah, you see this the the scroll bar pops up sometimes when you resize it.
But not but then when it's gone. Okay, now I I have the mouse on top of it. I can scroll down.
Wow, that was that was a lot of effort.
So yeah, so we have uh sensor FW data type headers to follow wire format change. This is a change to way sensors by Rajan. It was >> backported >> because um sensor MW upstream itself is they changed the V format in for the bus and they updated the KC plugin but they butroid sensors were using direct debas interface >> right similarly to what repowerd does >> yes >> right actually the code is from repd copy pasted >> yeah Uh >> but yeah uh so also sensor packaging update to the version has been reapply uh version 0.15.1.
Uh space war has seen space war is one of those community ports that are so seeing a lot of changes right now uh happening.
>> That's nothing font one. Yeah, nothing phone one. And we see a lot of people like just buying them now, which is exciting to see. We have a new community port that takes the cake uh for itself.
And yeah, so this is a like uh maintenance change suppress hardware adaptation arcs missing error uh for the device port. When I click on it, >> what about it?
fixes in USB mode D for the Nothing Phone one. Likely not making big improvements, but doesn't hurt, I would say. Okay. Yeah, that's a fair change to have.
Okay. Yeah.
So, going back to the list of mergers.
>> What about this introduced white wine change?
>> Yeah, I wanted to mention that next. Uh, I need to wait until this has reloaded.
Okay, the Fairphone 5. This is one of the thing things that I want to I want to set the expectation straight with what you can get from the Fairphone 5 as a daily driver experience for your laptop. Uh, when you want to run something extensive on it, it's going to thermal throttle.
So, if you want to run a game, which I've been able to do, I've actually uh something that I want to release as soon as possible if I find the bandwidth and somebody to uh upload it for me like from the open store team, the click to the click store to the open store uh wine box. I've been using uh white box and gl4s been able to run open arena which is an Intel x8664 binary compiled for Intel machines. You execute it from the terminal just dot slash. It chooses the right interpreter as for the binary and then it just runs it for you. It doesn't need any special box 64 command. It just runs it for you. And for those who really want to use the hardware for for this game, they run it with gl4s equals 1. All uppercase GL4ES equals 1.
And that's necessary to load something that is compatible with the lim hybrid drivers where you can use the GPU to some degree to render the game for you.
And it was fluid until it was thermal throttling. So I'm very glad see to see that working properly. So that gives me hope that some if somebody want to package a desktop OpenGL 2.1 app they can support it using GL4AS.
So that's an accomplishment and thanks to Petit Sap uh out there on the interwebs uh maintainer of GL4As and Box 64.
So because without that expertise we wouldn't have been able to accomplish that. So white wine clickable add stuff for white wine >> this is oh for the click merge.
>> Yeah it's it's actually two two things.
So um Dimitry Dimitry Menco um um has been adding uh white wine support. Well, white wine plug-in download support to mo browser. So, what is that exactly?
So, the white wine plug-in is proprietary. So, um um it's not open source, it's not free software, but it will extend more of browser in a way that you can watch the DRM protected or encrypted or what you call it content.
So, like I know all these well-known video platforms that uh you pay for instead of watching public television.
And um so and and and in want to touch we always had the problem that these sites were well they were unknown to us basically. So with that extension in morph it's there's a plugins menu and it knows just one plugin and it knows where to find it where to download it. So you can download it you can enable it you can disable it you can uninstall it. uh the white wine plug-in for well it's actually for chromium but morph browser is chromium engine based so and that is merged in morph browser and we'll need like I don't know maybe two more iterations to get some UI glitches out of the code or some better improvements into the code and also maybe improve some some improve the error handling when something goes wrong so like >> but it's a good start I see because the change is download manager uh shipping I think the download manager and just an additional tool to unpack the thing.
>> Yeah. Yeah.
>> Uh and hey that's hey sounds like using the right tools for the right job.
>> Yeah. Yeah definitely definitely. So so at the beginning it was like all happening in background and you couldn't control what was downloaded and what the progress and the status was but this is now all implemented and um and has been tested on AMD 64 and um and ARM 64. I'm not sure about actually but um yeah it's there and um so it will come with the next more of browser component release.
I mean it's in it's it's available in 2.x already. So if you if you are running the 2.x or uh if you um yeah have the 2.x dailies 2442.x dailies installed then you probably have that already. So please try it out and also provide feedback >> using GitLab please directly that would be most most convenient.
>> I would suppose this would need for us in the 2.x department to ship unsquash of in the root of >> uh I think we have added that to the dependenc dependency. Okay. Okay.
>> Yeah. Yeah. So it's getting pulled in via the dependencies in the package.
>> But it also requires a uh we actually have the morph dilemma here. Yeah, we have a morph dilemma. Have we have you talked about it?
>> We have not talked about this. This might be interesting to have an overview of the >> stat of the morph dilemma. So shall we shall I share it maybe?
>> Yeah, maybe. Yeah, go ahead.
>> I have to hide. Okay. Okay. So we had for a long time. So, so we didn't have QD6 in wood touch at all and all was good back then. So when then we started let's actually port the browser to QD6 because the chromium engine in QD5 web engine is rather old something 88 or I don't know. So most websites look really shitty when you when you use morph QD5 which probably all of us know. So we started on developing like porting the whole stack below morph uh lumir UI toolkit as a you know the big biggest kit on the block uh but also content hub download manager and whatnot uh started porting that to QD6 and started porting more of to QD6 and the lowest uh version of QD6 we would want to support was QD6.8 which is also in stable at the moment. Um not so in Ubuntu noble. So we had to vendor which means take packages from salsa debian or which is the gitlab side of Debian sync them into U ports and build the packages against the noble system and provide them for installation via our UB ports uh Debian dep package repos. So, and all was good then. And then we thought, ah, let's move onward.
QD6.8 is already pretty old. Let's get everything to QD6.9 and finally to QD6.10.
So, we started that and then there was some I don't know miscommunication between me and Rajan. So, he started merging. So, so I prepared everything and he started merging the first Q6 package and um then he immediately realized because I ped him on Telegram and I saw that and then he dropped the build but my test my tooling was broken because one of the merge requests was gone and then I thought well it can't be so bad and Rachel agreed and yeah so let's merge QD6.9 into Noble into the 2.x 2442.x. So that's what we did and then we realized that um the the rendering and this is Alfred you have to explain it because I don't know the >> the rendering department has changed in cute web engine because uh it's more expecting it switched over from using its own implementation of openges APIs uh with Vulcan which it now uses for everything graphical uh which means even uh video presentation needs to be adapted against a Vulcan environment and we're slowly growing. We just started with supporting Vulcan. We're not finished yet with that. Not everything has been merged and tested together. Uh but it's going to mer be merged soon. It's going to land soon in the main branches or in the main in in everything building off of the main branches. Uh and yeah.
>> Yeah. Yeah. So, but but then the problem then is that actually morph browser will crash when you start it if you don't have that right.
>> Yeah. If we >> if we switch the order in which the libraries are getting loaded then we fix the issue but it's not there at the moment.
>> Yeah. And and so so and our second problem is so that's where the morph browser becomes a mere dilemma. um that so so there is a solution that allows us now well in some test environments to use morph with QD6.9 with lip vulcan and use that via the villain protocol on top of mia compositor although just works with mia 2 compositor but all of the text is with mia one still so and so we have been like navigating back and forth okay so bring vulcan support into me one. Oh, this is a whole load of work, you know, and we want to kick that out in half a year anyway or a year. So then the second solution was um don't ship a browser. No, that wasn't just just kidding.
>> No, no, that's not a solution at all.
>> But the >> I know people are at home screaming what?
>> So, so but the next solution which is really way forward is okay, Alfred has been working on trust phones already. So they they are using weighted extension and he has this prototype phone which has a mia one for system compositor and mia 2 for the shell and they are nested on top of each other. We have mia nested anyway but mia one on m one but alfred has a prototype with mia 2 on mia one and and trust trumps are work trusts are working in that and mia 2 has very comprehensive way support. So Vulcan and Morph and Wayland and me too. Ah that works. Okay. But >> that's a lot changing in the stack at the same time though.
>> Yeah. But it's it's it's a lot. And so what we are actually so the solution we discussed yesterday over actually that we revert back to the QT6 web engine uh 6.9 to 6.8 and then we are good again.
And then finally 2.x will support morph browser with Q6.
>> Yeah. So but sort of we really so >> we have everything we have to be sorry for that.
>> Yeah it's it's a slight delay in releasing the thing but I think we will we'll we will make it to our promise that we release at the six months of the release of 1.0 then we release 2.0 I think we're still on track on that one.
>> Yeah. So yeah we lagging a bit behind the I guess.
>> Was it was it September September and March? No, when did we didn't we inventor?
>> Didn't we release 1.0 in winter?
>> I forgot.
>> Maybe. Okay. So, anyhow, uh yeah, I see this one.
>> Wait, wait, wait, wait. So, so, so fine.
No, 24042.0 will come with MI and with a little bit of QD6, not the full SDK and more Q6.
Yeah, Morph Q6 that but not MI 2. So me too, we really need to postpone because that that would just be like too much to chew on.
>> Okay, I see a question that I want to take real quick of the live chat because it involves Steam and uh I tried the Steam Snap on the Fairphone 5. Uh sadly, it's uh it is able to smoothly run a few games if they were old like Counterstrike 16 and Open Arena as I told you earlier. Uh, and the therm the the performance really slows down once the thermal throttling kicks in. So, uh, it's not for a long gaming session. If you really just want to kill time for five minutes, maybe you're hit the mark properly.
So, yeah, but you can give it a try.
It's going to be soft oriented most of the time though. Uh the good thing is that if we get uh JLS, J4s and other APIs like welcome working it means that once we have more powerful hardware supported we could run more things.
>> We can run what?
>> More things without >> more things. Yeah.
>> Yeah. Because we can rely more on the GPU doing the work for us without copying a lot of uh stuff around between the CPU and the GPU in in technical terms.
Uh so yeah so that's something and also uh I was just going through old commits that we did for other in other areas and I I stumbled upon this cute Ubuntu camera change where I implemented the camera uh viewfinder mapping into user memory into user space memory where you can just map it uh as a character pointer basically or a void pointer. as you want to do it and for the QR code reader and doing that was using technology and tweaks and use usual mechanisms that you can use in GL4s. So, and I shared the information with petite sap uh so I'm looking forward to talking with that guy again about this topic because we can land some cool improvements there.
Okay. So maybe maybe we should take a look at the time as Adam already mentioned here. So uh let's let's maybe look at the pet and see what we what we still have on the agenda. Right.
>> So it's loading sadly. That's because it's running out of memory. Seems memory preserving.
>> At least it does not crash.
>> It doesn't crash. No. Let's go with the good old handle.
uh applications for the open store. I I guess we don't have experiments. We don't have more to say about GitLab. Uh applications from the open store. So, we have five new apps here coming in for this fine new episode and a bunch of updated apps. So, I'd say we go into detail with the new apps and the updated apps. we just go weeks through and just mention they are updated. So, LX Keys by Deser Venat.
Yeah. Uh another keypass app basically uh I'm not sure if what the difference is between the that one and what has been released. Yeah, I I read some comment um in in the app because I was also wondering because we have it's the second second or third keys we have now and I think maybe the second and this one has right support it seems and the other one >> it does >> doesn't have so and Alex Alex keepass will only exist as long as the other one doesn't have right support or so that go to the comment section and uh it's explained why why there's a second key pass >> okay well but At least nice to see another app popping up that tries to solve the same problem with with managing your passwords. The next one is me time by dark uh lithium to plan for me time. Uh >> I guess >> not not Windows MVs. No, me myself.
>> Me time >> to plan for me time. Yeah, could be whatever.
>> Let's Hey, we have the magic of the internet. We can just No. Meanwhile, we can just search for it. We cannot share the screen, but we can at least look at the listing on the open store time optimization tool or something.
>> Oh, uh, I'm gonna I'm gonna read it in German actually.
So, uh, yeah, there are Oh, picture screenshots. There are screenshots for us to judge what we're actually dealing with here. Uh, start a new week. New week. How do you want to shape your me time this week? And then you can select it in a dialogue.
So, you can select your interests. It seems music making music, >> how much you've done over the course of the week.
>> Yeah, it's basically things you want to achieve and um >> if you want to learn programming, that's one way of managing it, it seems.
>> Yeah.
>> Cool.
>> So, what what's what's what's on the list as well?
>> Uh we also have uh plates by Breno Alar.
Oh, I that's for uh weightlifting. Um and calculating the weight the discs on I'm not sure what that is in English actually.
>> Yeah, for your muscle training.
>> Yeah, for the for your muscle training.
So, >> okay, cool. I want I want to look that one up real quick. Just if it has a screenshot makes it easier to judge what we are dealing with here.
>> Just five apps.
>> Oh, you have them.
Okay. Yeah. Looks very modern.
Rounded corners for some buttons.
Okay.
Yeah, this looks good.
>> So, >> looks very professional.
>> Looks like what you expect from a different platform.
So, then we have my countdown uh by Tim C or Tim Slim. Is that an L or an I?
Tim Z.
>> Countdown to your most expected events.
>> Ah, so foster countdown, >> right? That's cool.
>> Yeah.
>> And the last one is dart counter by six web application for dart tournament website.
Uh that's new app. Those are the new apps that have been released.
Uh and also some updated apps. Let's breeze through them. Sonic player has been updated. The modics uh tipper sealed so 2 coconut time management alpha has been updated.
UV box uh that one is the next one is for very impossible for me to pronounce in any way, shape or form. Do you know what that one is?
con it's a flexi is that speckle of dust on your screen >> doesn't it doesn't go away >> doesn't go away no >> doesn't go away I don't know I don't even know how to pronounce it actually >> okay let's skip this one then uh CBS 2 oh yeah sebass the text editor great stuff that I occasionally have to use uh on the go on the font 5 and it's helpful to have a capable text editor Maybe this is something that we can suggest as an IDE option for the the new folks coming in >> if they need something that has syntax highlighting.
>> Cas is good.
>> Cas has it.
>> Yeah, I can tell it's really helpful when you have some text files or logs people send you on the go.
>> Oh yeah, definitely.
Okay. Uh >> and I found the pronunciation for that uh can actually.
>> Ah okay.
So v uh victionary says it's anthology of verse collection of songs cycle of songs for light.
>> Okay.
>> Ah okay.
>> Okay. That was sebas. Uh forma has been updated. Contact bridge. I need to check my Oh no. I've I've give I've updated it already. I need to give it a shot still.
uh contact bridge syncs your contacts to uh Nexcloud or from Nexcloud actually uh to your phone and stores it in contactive which is a nice workar around for those who have been waiting for contract synchronization and haven't gotten it and now they have it thanks to contact bridge >> and we can actually tease that that will I guess we want to have that in 2.x as well. Leonel is really ambitious.
>> Oh yeah, definitely. So, so then it will be bad news for the maintainer of contact bridge because then it will be part of Buddha touch core or lumiri core uh that that the address book app will then be able to set up nextcloud remote accounts via online accounts which then allows syncing um of of contacts into into nextcloud via the boutio sync framework that that will replace the not available or not working mechanism we currently have with address books Okay, next one is I quit. This is an application for uh that helps you uh quitting your smoking habit.
Prospect mail reference CD teddit where and pickpocket have been updated and also UT Tesla and Bitcoin lightning wallet. So those are the updated apps.
If you heard one of them that you were very excited about, you have it installed on your device. Now, you can rest assured the updates will contain it. Just update your phone and you get it. Uh, and that takes us to sponsors.
So, if the scroll bar works, we're all happy.
Okay, we have sponsors.
>> I can caress it from time to time.
I'm going very hard on the thing.
>> FP5, you're doing good.
>> Yeah, that's it's capable.
Okay, so if you uh want to see what we are dealing with here, we have some sponsors here. Uh we are smooth va pine 64stein digital ocean nriifi udu community association private internet access private and gardener and associates consulting private internet access tixia gitlab cloud run code looter gi.net Brex Technologies and OS Science.
>> The location here, >> this location right here, the place that we are at right now. And we also have a few patronons.
Uh, >> wait a second. Now, ready? Come on. You need to you need to come here and speak directly into the microphone.
>> Yeah.
>> Okay.
Let me try to mispronounce the names this time.
>> Yeah, try >> conspel Escora JT K Clark Kenneth Dream Luigi 300 11 Mr. T one 01 Nina >> 33 I think 101 >> Nina Omegas >> and Zombies >> and the random files at the bottom >> Jose Evandro PZ Alves Marco M and Mh. I don't see it.
Maybe >> uh we need to grab the handle of love here to scroll. And the last one, there's another one.
>> Robert Bruce.
>> Uh Robert Bruce.
>> Perfect.
And if you want us to mispronounce your names occasionally, every two weeks at least, then join us uh and join the fun by heading over to ubiports.com/donate where you can find all the details on how to donate uh and uh bring your financial contributions.
What you can also do is head over to the ubports.com/shop URL where you can find some merch uh for Upports. For example, we have socks, we have plushies, we have um coffee mugs.
Uh and they won't make us rich, but they will make us famous as mom says so very kindly and fittingly.
So, we have some questions from the forum and I would say we start with Diego 28A.
>> Wait a sec. Can you drop me the link on Telegram for the P?
>> Okay.
>> Thank you.
>> Uh Diego asks, "With the 2404 base now stable OTAA 1.3, the community is looking forward to the 2.x series." My question is about the physical usability of the interface ergonomics of large devices like on tablets uh of like for example the bola tablet and large phones like the Pixel 3 XL having the back button and menus in the top left corner is very uncomfortable for daily use. Are the there plans in the 2.x series to implement bottom based navigation or thumb friendly controls to avoid finger gymnastics? Uh no, we don't have anything planned right now, at least not uh immediately and not in the short term or medium term. Uh and we don't see a way forward at that in any way, shape or form because the the UI is expected to be laid out in this sense where it has the header bar with the back button on the left side and the content below that. And diverging from that would mean redoing the human interface guidelines that we have created over the years and that we or that can started and we still follow for the most part without mentioning that we changed just colors over the years but everything else is is the same as connected and we are okay with those human interface uh guidelines being present and promoted. So we want to keep that >> simplify fire.
>> Let me let me just explain add something to that. So it's also changing the UI of an environment operating environment.
It's not something like let's move a button around. It's it's really a conceptual thing. And if if there are people out there who say yeah okay we we we like we have concerns. So you would like to have see this changed and and and you're a graphics designer and you can provide mockups and and really dive into uh the next generation of the human interface design of Ubuntu touch or Lumiri that is actually uh you know show up and and and and contribute that's would be really awesome but but but um changing the whole design is not done in a week or a month or you know and and it needs for especially it needs UX and uh designers UX people and designers to actually work on that. So um and and yeah contributions are really welcome but but that needs to be really done in depth and and and throughout the whole stack conceptual design and the implementation afterwards is is not is also a thing but it's not the thing >> but the UX the UX folks over on Telegram are very they are worker bees on that topic they're on that front they're very active >> and when you come up with a good looking markup >> uh they might consider it but other than that if there's no way to keep weight uh backwards compatibility with existing applications. It must be a new API or I would suggest uh then we could do something about that and make applications more aware of where to put the back button. It might be something that the app needs to implement, but like the browser, for example, needs to explicitly do it, but it's maybe a start to have not only a header bar, but a footer, a footer bar with a back button.
>> Yeah. Or a gesture or something >> or maybe something with gestures. But yeah, >> remind me a very old topic uh when people were complaining about Touch not supporting hardware back buttons on phones. hardware back.
>> Yeah. But now we have a problem sol for us simply because there are no fonts left with the hardware back buttons.
>> Right.
>> Okay. So the second question is simplified file sharing. Will the 2.x series finally integrate Bluetooth into the content hub? For a mainstream user, not being able to share photo via Bluetooth from the gallery is a major deal breaker.
So, uh, I know that there used to be in6004 and in 2004 Bluetooth files sharing applications, but but I don't know how well that one worked uh, with more recent updates. I haven't tried it in a while. Uh, I haven't had the need to transfer anything over Bluetooth. Is it actually so well we are you know um taking longer than we want actually but there's this initiative we have started about the file manager and remote storage providers.
>> Yeah. So we want to sort of revive the Lumiri storage framework and provide native support for cloud syncing into Lumiere file manager and actually also the so that you can actually use the terminal and then store files by copying with shell commands into a specific folder and then it ends up on a remote share. Could that actually also be like a similar use case?
>> Yeah.
>> Yeah. So that Bluetooth is just one remote cloud storage provider in quote unquote.
>> Well, could be we we would have to look into it.
>> Yeah. Yeah. It just comes to mind. So, okay. Um I I never heard about this. I don't share files over you Bluetooth because I'm not such a mobile phone user, which is quite funny actually, but I'm not having this use case. I'm not taking photos. That's the other problem.
So, but interesting interesting request.
So, >> oh yeah >> and maybe I should use more iPhones and Androids to be aware of these features.
>> Beyond the expert user, what is the road map to make basic tasks like file sharing, file management or sharing and file management uh setting intuitive enough so that a non-technical user never has to open the terminal. Uh this is for if you are talking about for example one success story that I can tell is installing clicks directly off the file manager without needing a terminal. This is the preferred way of installing clicks and it always works.
So we have one example where content hub where you can select the app to open the file with uh shines and it does its job properly.
I'd actually be interested in the exact problems because so so if you think that some some sharing mechanism doesn't work. So you have a certain file type, you have a certain target app and you can't connect those two while sending that file over and then please file a bug on on GitLab against the Lumi content app. So then we have to or the or the app repos I don't know. So, so we have to or just the we have this meta repo for all of a to touch but provide them as the the exact problem cases as bug reports and then we can look into it in detail.
>> Okay.
>> So, it's too generic here. Um whoever whoever contributed this question.
>> Uh okay.
Yeah, I think we should go to the next question. Uh NBD Y NI uh says or else end of last year there was this post about smooth edges. How far can we develop the daily driving regarding this post?
For example, my browser is still crashing. Bluetooth connecting to my car is still a gamble if it will work or not. Multitasking seems impossible. apps are open but switching towards it ends up to be an empty window.
Keyboard seems to improve a bit but there are still a lot of input delay and lag when typing. What can expect you guys for the near future of a Moto Touch? Love your nails Mario.
>> So Bluetooth >> some of those I want to I want to address this. Some of those things have been fixed like the dangling X11 windows have been addressed. The applications need to be adapted to go with it. Uh but we have something that should improve the experiment the the experience experience. We have some bluey come changes coming up which should improve the Bluetooth connectivity in your car as tested by Marios with multiple uh ways and cars uh with multiple devices.
Also, there was Bluetooth tech updates for some devices by Muhammad.
>> Oh, yeah. Right. For example, the Fairphone 5 has has those. So, very interesting stuff. And those should improve the the experiments a lot. So, we should have way better Bluetooth coming out of the box with 2404 2.x and 2604 uh as soon as possible. And actually with QD contacts escalite the new contact back end framework thing um there is actually a proof of concept that uh contact syncing into your car media system the for making phone calls and have there for for picking people to call from the car system that's also coming.
>> Okay. Um next question coming in from CIS admin. Uh question for Russian users. U report to touch repository mirrors.
Many Russian users face issues with no access to official reports resources like the downloads of package repositories and updates due to blocks.
Using VPN is inconvenient, unreliable and should be a mandatory crutch for an open-source project aiming at users worldwide. our official mirrors for repositories and download images which is planned in Russia like on mirror yandex rue or mirror fedora project.org /ru uh or RU um or other reliable hosts or are at least instructions for setting up alternative mirrors in system settings, content sharing for OTAA updates and open store app installations. Without this, the project becomes inaccessible to a significant portion of the audience. Well, we have no am plans on that. It's just that we are caught up in this whole mess. Uh forever whoever whoever is uh like at fault here, get your bleep together. Uh thank you.
>> Well, it's not that someone is specifically at fault. It's just that we we using Cloudflare, >> right?
>> And the Cloudflare IP address is getting blocked, >> right? So, Cloudflare is blocked and we use Cloudflare to balance out a few traffic things uh with regards to the website and the downloads and whatnot. So, yeah, >> but to answer the question, if someone wanted to implement mirror support, there are multiple places that needs to be addressed.
>> Actually, there there is there is there are open merge requests that that bring mirror as a system image server. Yes.
>> Okay.
>> So, and then but it needs people well to test that. So, so >> yeah.
>> And then we need support in installer set up the mirror for system image or to select the mirror for system image and settings >> in in the system itself >> and probably also in the UB port installed. Yeah, that's what you said that's what you meant.
>> But comp store is more difficult because that one has API right. It's not like static. Mhm. archive his deb packages.
>> Yeah.
>> Okay.
>> Okay. Yeah, it needs someone to work on.
>> We can't we can't shake this out of our uh like jackets right off the bat right right here. So, we have nothing to say more than that.
>> Same question more or less.
>> So, Vlad asks uh same question more or less. Is there a way to install new by downloading the necessary files beforehand? there is an increasing number of people living in countries such as Russia that block our servers or are unable to complete the installation.
Uh what you could do is provide a so-called Ubuntu command file that you push into a specific place in the recovery with the device being in the recovery. You place it there. You upload all the files that you want to install.
Uh but getting this command file right is tedious >> and once you do it wrong uh it get it the recovery gets reboot. You can reboot the recovery and you're back at zero.
You can rest assured that you're not in this loop of it always trying to reinstall the update.
>> I've I've actually been in touch with someone who mimic the same functionality as is in the human installer in a shell script. Oh yeah, Muhammad also did the same thing.
>> Is that Muhammad?
>> I think it was a mu in his really early stages when I even didn't know who he is or so. So So I think it was one of the first things he did.
>> Actually packaged this as a snap. I tried to package it as a snap and it starts and it works.
>> So and then this this should tell you the logic for downloading the files via VPN, caching them locally and then flashing them to phones, >> right? To be uh to be precise, we actually have scripts as part of highly engineering build tools which download the OTAA from the ser systems image server and prepare Ubuntu command file as well.
>> Right?
>> So if someone wanted to look at that, there are plenty of examples.
>> But at the same time, I wonder if we could have a way to provide fleshable images of >> right >> because um >> like fast food. Yeah, that's something that has been asked for people for a while, I guess.
>> Yeah.
>> And it would make some things easier, but we need the storage for it.
>> We need the storage for that. Yeah.
>> Yeah. That's the good thing about uh system image. It keeps a delta in the tarball and keeps the full ones in the tarball. It keeps as many as configured and it manages storage uh as is available and given by the server to the system image user data directory. So yeah >> in my opinion it's a lot of extra steps and if we had a flashable image we could reduce the uh surface for failure let's say during the installation. Yeah, >> because we already built flashable images as part of factory images for wall for example. So we have the infrastructure there. We just need some place to store the images. Actually >> we know it works. We just need to more prominently make it available uh through regular means.
>> Yes.
>> Uh which we could do through the installer still. I mean it supports flashing the partitions using fast boot.
Uh we could make it just download some prepared images from CD image >> or even select uh the image from file system.
>> Oh yeah or that. Yeah.
Having your custom builds.
So there are plenty of ways. We just need more people and more time and more hands to work on it. Uh we have very few people knowledgeable in the installer.
Uh so that is that.
Do we have any questions coming in from live uh audience?
Yes, Sander has a question.
>> I heard great news that uh steps are now available free. So I pulled the code, compiled it and installed on my phone.
>> Okay.
>> Um but but one thing I noticed when starting Snap app is that the scaling is off. or is that something that is going to be implemented?
>> So scaling is off for GDK applications and SDL applications by default. Uh SDL applications uh have more chances of it being fixed because uh the scaling is part of the application code. Cute scaling uh can be improved uh should be improved. Uh but if you use the Lumir UI toolkit in your stack, you get the correct scaling. Correct.
for GTK apps. Is it technically possible to ship a hacked GTK in snaps or it's very >> you could ship you can ship a hack in snaps too. Yeah. If you want to really package uh specific uh GTK application >> no what I mean is that like in flatp pack it was possible to share some modules. So can you replace the dependency >> for that you would have to build your own platform snap. Oh, okay.
>> Which is owned by Canonic >> and if you want to download it from the snap store uh then you hit some you had some issue because it's released by canonical. You cannot have your own. You could try releasing your own, building your own, but you have to be API in uh compatible with the libraries in the regular content snap. And once you connect those, they need to be working as tested. Uh not everything sometimes it can cause issues, but generally speaking, it should work.
So just that's the beautiful thing about snaps. They have their content snap concept where every library that you want to share across a multitude of applications uh you just put it in a content snap you hook it up by scripts and that's how the how we managed to package lib hybrid with bulan and gl4s those three things uh inside of a content snap and release that as a driver support so to say for snaps so that you can run your desktop OpenGL applications, your Wayland applications, EGL applications, you can run all of that uh in a snap environment properly on a photo touch.
>> There was a question a comment in the chat for the our previous question. Uh now we getting unified recovery. Maybe it could be a function of it flash recovery and push turbos.
>> Oh yeah, right. Now that we get the unified recovery, uh I really enjoy the work on the unified recovery because it eases things a lot and it also potentially enables us just a thought uh to have a working recovery for the Pine devices for mainline devices because we just use this mostly uh like a handful of dynamic libraries from Android uh three binaries from Android toy box the recovery maybe UN entity and uh the recovery binary specifically and running that on a mainline device should be doable.
>> It it should it should totally work actually because it provides DM rendering back end.
>> So someone needs to try it I guess.
>> Oh yeah. But it's a path forward uh for something that we've been craving for a long time which we still don't have. But now we at least have an idea of what we could do uh to solve the issue.
Great.
So let's hide that. Uh I would say we have spent a long time now on this Q&A. Uh we are at 1 hour and 10 minutes. Uh so I would say we end it and thanks again for everybody joining us. It was a pleasure >> staying with us uh until very late and yes we'll see each other next time and until then stay healthy, stay safe. Not like that. That's not healthy. Let's let's put that away. Uh, and enjoy the time. Until next time.
>> Mhm.
>> Thank you for watching.
>> Thank you, Mom, for watching.
>> Thank you for watching.
Bye-bye.
Vidéos Similaires
Learning k8s ep. 3 - The end of the VM
devcentral
102 views•2026-05-15
Iterators and Generators: Real Use Cases
jsmentor-uk
188 views•2026-05-17
TCS NQT Coding Questions Solution (One Shot) | TCS NQT Preparation 2027 | TCS Actual PYQ 2026
knacademy20
2K views•2026-05-17
The 4 Bit AI Training Trick
explaquiz
414 views•2026-05-19
Image to 3D World Workflow 👀
badxstudio
843 views•2026-05-16
Why Learn Algorithms in the AI Era
bitsandproofs
245 views•2026-05-17
NFA - Transition Diagram and Transition Table
nesoacademy
198 views•2026-05-19
DSA Topics and Algorithms Overview #coding
DSA-in-Minutes1
423 views•2026-05-15











