Lomiri in Debian - Convergence for all Device Form Factors

Hello and welcome back um to this afternoon talk. It's about Lumier Conve factors by Mike Gabriel. Give a warm applause. >> Yeah, thanks everyone for coming. Um so I've been here on debump and debc camp for quite a few days and um I've been sitting in front of several notebooks and working on graphical UI stuff and and people came around hey what is Lumiere actually and so and this is what the talk is about what is Lumiere what did we do for Lumiere with Lumiere and Debian over the past two years at least maybe more actually and also it will be as you can see in this slide it looks a bit Ubuntues Um uh we will als I will also talk about Lumiri upstream development. So and why the Ubuntu background you will see in a minute. So um of course it's not only about Lumiri upstream it's also about Lumir and Debian and uh we tried to make this visible if you install a Debian 13 system and um so Lumier to spoil that is in Debian 13 and we still need a lot of help testers back reporters um and um yeah so I'd actually be happy if some of you try that on their notebooks or their their tablets and and get some feedback on that. So what is special about Lumier? Um that is convergence. So what is convergence? Um so convergence means basically you have so it's absolutely in the realm of desktop environments of user environments interaction between user and device. So it's a desktop environment but let's cross out the desktop. It's an operating environment because it's not for desktop only. It is for every device form factor. So as you can see in the slide, it is for tablets, for phones, for uh notebooks, for notebooks with external displays. So it can scale all the way up from a little device in your hands uh to your uh uh desktop working environment in your office. Um and the thing is that it's always the same software. So the same the software Lumiere as a software has the ability to adapt to your uh display layout your um also the performance um um it can be used as a phone as a tablet and as a desktop PC in your office with um office office software and you can actually switch live from phone mode. I'm not going to demonstrate here this because I don't like live demos, but it's possible. You can attach a phone via USBC to a docking station and the phone switches over to becoming a PC. So, with all applications still open, they just resize, rescale the windows and uh you can continue with what you started while you were still on the way to your office at your office. So, and this is this is convergence and this is what makes Lumiere quite special. Um so the history so some of you might have known Unity 8 and Unity 7 in Ubuntu and this is where the background comes from. So it's Lumiere is a fork of Unity 8. Fork means we took the code we renamed everything and uh published it again under uh same licenses but under different names. uh we have different maintainers and um the ma group of maintainers is in fact the Ubuntu touch team. So it is the operating shell in Ubuntu touch but this is not going to be a talk about Ubuntu touch it's about Lumiere. So I'll come to the difference between Ubuntu touch and Lumiere in a second as I said it's just one operating environment you can have in your enterprise context on all the devices in your company. So you could have, you know, you have your staff and the staff can have a phone, can have tablets, it can have desktops and just running the same operating environment using the same software to actually manage system manage this pool of devices. Um, Lumiere is fully free and open source. That's hopefully clear and but it is definitely compared to other Linux-based phone operating systems. Um and the the the development is mainly communitydriven but we are as a community financially backed. So this is really good to have. And so maybe for some of you the word mir means a thing. So Mia used to be an idea at canonical to become a new display server protocol like x11 like wayand like mir and it originally started at that but mir now has to be renamed now it is still mirror but it's a display compositor for wayand so it's good if you have still the mere display server word in your head wipe it out replace it by mir way compositor and also this is a continuation of a talk I already gave at depcon on 17 in Montreal I think um which was about Ayatana indicators and Ayatana indicators are in Lumiri the little icons in the top mostly right corner u but they can be used on marty desktop they are used uh on budgy desktop in kubuntu they are used so it's it's a component in lumiri but also uh applicable or usable cross desktop environment so Um so what is Lumiere and what is Ubuntu touch? When I joined the project which is the basically the Ubuntu touch project uh run by the UB ports foundation. Um this was not really clear because it was all one GitLab area with the code repositories and it was not really and and no there was just the Ubuntu touch release workflow. So it was not really evident okay if we ever wanted to you know package Lumiere on its own for distribution where do we take the code from from latest githads which doesn't really work well because you don't have version numbers for packages and so on. So the first thing I try to land in the project is okay can we please differentiate. So Ubuntu touch is a derivative it's it's a phone operating system on the first end but it's also a derivative of Ubuntu and the developers of Ubuntu touch have a deep interest in Lumiere because it's the thing we try to sort of implement and and we have achieved that is um that first of all in in people's heads there's the Lumiere upstream and there's Ubuntu touch although it's sort of in the same GitLab area but for distributors uh what is really possible We we we we um set up two different release workflows. One for the phone operating system and one for the 130 I don't know whatever components of Lumiere. So and then individual release numbers for the different components. So and that was actually the foundation of being able to package uh Lumiere for Debian uh and also for other distributions. I'll come to that at the very end of the presentation. So um for Lumiri and Debian um personally I had a few goals. Um, so one was that I wanted really to show that Lumiere or Unity 8 at that time can be brought to life outside of the Ubuntu ecosystem because Lumiere was dependent on stuff in Ubuntu that I couldn't change in Debian or in Fedora or wherever um that that is in Ubuntu components core components for example account service is one of them and Unity 8 was really written for that. So it was you know um tightly inter interwoven. So I wanted to show it works. So and with being able to have Lumier or former Unity 8 on non Ubuntu you also create visibility for the whole operating environment. So we wanted to create visibility. Um then I really was in need of because I have customers in my company uh who would like to see a Linux only tablet. So not like Android with some Helium stuff and then so just Linux machines as tablets. Um I wanted to also enable other distributors to package Lumiri. So that was also one goal which we also achieved. Um and then um so originally Lumier was designed as a con or Unity 8 was designed as a convergent operating environment and the plan at Canonical was and people may correct me if I'm citing that wrong. they wanted to bring Unity 8 to the desktop. So as a um as a successor of Unity 7 which they didn't do they switched to Gnome at the end and and dropped the Unity 8 project in 200 207 when the Upports Foundation took over or the Ubots community at that time. Um but then the main goal was to get Unity 8 fork then as Lumier to work properly on phones. So the desktop use case was a bit neglected, ignored or yeah so so features were added, code was changed that worked well on the phones but not so much on the desktop use case. So in where applications are not in full screen where applications are windows. Um so then for Debian in Ubuntu touch there is this mia display compositor which in Ubuntu touch is still the mere display server. So it's an old version and this is still required by some components. While it was clear we can't ship that in a distribution that is state-of-the-art. So we had to actually move forward to the latest MI version that's MI one versus MI 2. And so we wanted to ship that. So which we also achieved and then there is for groupware applications like contacts and calendars there is a synchronization framework now in use in Debian that's called boutiosync and that's actually um taken over from safe and we also cooperate with the maintainers there the upstream maintainers so a bit in Debian we we try to sort of stage the next generation of Ubuntu touch what will be on the phone me to boutio sync framework is now already available in Debian 30 or in the to be released Debian 30. So and oh why the hassle yeah because I love Debian Edu I'm also maintainer of Debian Edu head of Debian Edo at the moment and I would I wanted to provide my school customers and school friends and people working at schools with an uh alternative solution for iPads so tablets. So lumary. So and then as another thing I've been working on recently is the Windows to Linux workstation migration in the northern company northern state of of Germany which is called Hstein. So the whole whole federal state is migrating away from from Windows and M from Microsoft products. So and but they don't have a solution for phones yet. So I also wanted to have you know something that that could be interesting for enterprise environments. So this is the really final final goal which the four and five haven't been achieved yet but the others quite well in fact. So um we achieved while while working on all of this we achieved quite a bit. Um so simple things partially but for developers really important. So one thing was that we so we had like merge requests being merged crisscross across this history tree and it was like if you would visualize the the parallel trees that were merged again. It was really spaghetti. So we tried to sort of boil it down on a similar merge strategy in GitLab which I really recommend if you if you run projects in GitLab use that. So all commits that go together as one unit are in one merge request and the next merge request you have to rebase before you can merge it. Um so we wanted to detach the development more from the Ubuntu touch development and also from the release cycle and um so and and and we also so one thing was Ubuntu touch still but we are about to release a newer version still uses Ubuntu 2004 and we have an agreement with Canonical to have the ES uh ESM packages so the extended uh support um versions of security updates, but we wanted to we wanted to make sure that although we use this old Ubuntu version underneath still and it would change in October, um we wanted to make sure that our upstream components still already built against latest latest third party components like a latest GNU system. So we we ran basically our CI against uh the Ubuntu target version and Debian testing. [snorts] Um so and we also as part of this project we also contributed to um to to Lumiri upstream. Um so as I said loads of code potions were actually failing to build when the compiler version was newer than the one in Ubuntu 2004 and so on. So we we adjusted that. Um there were still some some site components that hadn't been properly renamed to the Lumiri namespace which makes it easier to bring the packages into distribution because you can be sure if the Lumiere name is in the package in the package name then you won't overlap or conflict with other packages. So we did that. Um we we were on this journey to free Lumiri from the Ubuntu only realm which we which we achieved finally in this project. Um there was various Golang projects that Gido, one one member of our team, uh really dug in deeply to actually um port them over to latest latest Go and latest libraries. Um we we worked on the on the API generation process uh for app developers. Um, we ported uh an email client over from some containerized build to something that's actually shippable in a distribution. Um, part of it was dropping a very old uh tool chain called QBS and ported to CMake and we started porting Lumier to QD6. So this is still work in progress. This would come but not for Debian 13. So and in the packaging realm so what we did is we touched around 135 packages and um so we have a team for that in Debian and on salsa and on that team you can find an issue board and so we don't use BTS because we have many issues that don't really would just clobber BTS. So there is there is a convent board that you can look at where we where we sort of track the changes we need to land in Lumiri. Um and in this on this board we had we identified around 217 issues either in the upstream code or in Debian and we we fixed those um not uh we fixed 215 of them and um we we got the latest MIA version into Debian which was also a transition had to be coordinated with Lexqd and Ubuntu. Um we gotono back and network manager support uh oppono support back into network manager. So and also um just before the Debian 13 release the KV people said ah we have to transition SyonD to QD6 and they said please don't. So we actually introduced dual builds of sign and friends so that they the build runs twice and you get the Q6 version of the client library and the QT5 version of the library. So um for app packaging so in in Ubuntu in Ubuntu touch uh apps come as so-called click packages. It's a predecessor of snap packages and you can also install snap packages on Ubuntu touch nowadays is sort of still in work in progress but it works or it's promising at least. Um but there are some apps that we definitely wanted to ship in Debian as native dev packages because we you you could install snaps but there are not so many Ubuntu touch uh apps already packaged as snaps. So um so we wanted to make sure that we have quite an ecosystem of applications available. So uh I won't read them all. You have the list on screen. Um, and the idea is to bring more apps in after the Debian 13 release and make them available via Trixie backports. So, and and also if you if you want to chime in and say, "Hey, Lumir is my thing and I'm a developer or maintainer and I would like to sort of grab some app and make it make it work on Debian, feel free to do so. Please get in touch." So, we haven't achieved everything. Um so one thing we are not so happy about but we are working on it and we hope to be able to bring it to uh Debian Trixie backports so that also Debian 13 users can benefit from that. The browser that comes with Ubuntu touch that is also very you know can be used uh on touchy interfaces um is still on QT5 which means it uses a backend for rendering websites that's basically Chromium 87 and many websites these days won't work at all or complain about that. So we hope to this is the current effort that's going on in the in the background. We are getting the underlying libraries to Q6. We are porting morph browser to QD6. So we have a recent web Q web engine and then we are at version chromium 100 and whatever. So um not the latest one but a considerably recent one. So then the network indicator um we identified some let me call it chaos because an older version was forked at the time Lumiere was created that didn't have the patches from uh a canonical developer for cable non support. So you can attach a cable and you get an IP address but network manager is doing it in the background and it's not reflected in the UI. So we are working on that. So then we were a bit late for the push service that is related to Lumiri and uh the MMS um well one component that's important for MMS sending and receiving if in case you have a device with a GSM modem that can support that um there will be at some point there will be a backup solution for the user space. So this wasn't planned but it will come in in the cycle of Debian 14. And if anyone would love to develop a QD or QD quickmlbas based uh native matrix client uh everyone is welcome because we don't have that uh we have a matrix client but it's not packageable for Debian. Um and then Lumiere has never been um so there there was never the need to actually do the XCG desktop portal uh um stuff on the touch phones because they had a different approach for that. Um so but we realize if you combine Lumier with other GDK applications or so then it would be good to have that supported. So so one one downside of that for example is that you can't do screen sharing in video calls with your web browser. So and this is basically a call for help. So maybe people watch this uh after the conference and and feel ah I tried this and I maybe I can help. So uh we have various issues um that we still observe. So um so I'm not I'm not selling Lumiere as yeah use it as your production system but it's it's it's a big a good sneak preview. So I've been working uh with Lumiere preparing the stalk over the past few days. I've been using it then all day and um I'm I'm also using it all day uh when I'm at home from time to time but when I need stability when I'm have to be really sure that my session does not die on the way uh then I for example now I does not do not present on lumiri um so so we have these crashes and then we we have quite an issue with multimonitor support I'm currently working on uh a patch set that will enable this when you're logged in but the greeter For example, Lumiri comes with a greeter and uh an operation shell. Uh the greeter does not really support it yet. So we are working on that. Um we still have issues with a phonono and lumiri indicator network in Debian 13, but I hope that we can land this either before the release or before the first point release. Um the the onscreen keyboard um is well supported in Lumiri apps but uh at the moment we observe that it doesn't open when you click on text fields in in Wayland oh no in X wayand apps so in X11 apps that run on Lumier which is possible but they that doesn't interact with the onscreen keyboard so well um then we are so so for for actually for the Mobian people I know it is interesting to test Lumi But then like a fairphone 5 for example, it has a very high display resolution and everything is really really tiny. And in the MI one version that Ubuntu touch uses, you have fractional scaling support. In the MI 2 version, it only works partially. So this is also something that needs love. Um we have a bug with app window placement. Um I will come to something um later on where we hope to attract developers as some sort of a bounty program. Um so it would be nice if people actually would look into that for example. Um so and various UI glitches. So all the issues are documented on salsar on uh the canvan board of the reports team. So and um finally I have some impressions. So basically screenshots and minimal instructions how to uh lumiereize your Debian system. So get your Debbian 13, install it minimally, do an up install, Lumier, Chromium, and it you don't have to do the Chromium, but unfortunately in my screenshot, I had Chromium. So I added that because Chromium is not a native dependency, but it's it's good to have because of the morph browser, which is the the blue one with a wing, looks like a butterfly, because it's the the underlying Chromium engine is for some it's it's too old for some websites. So I always have both of them. Um so this is an easy command and you can test uh Lumiere with that. So, and if you want to brand it in Debian look and feel, then uh we have a a an extra package that does the so Lumier basically has a launcher on a dock on the left side where also the running applications appear and it has a launcher menu and um so and as you can see I I'm not I I I don't have too many applications installed here because this is something that that that um for example a bounty will be for. Um to actually come up with ideas of an a different style look and feel of this this this launcher menu so that you can also hand that you as a user can handle many many apps being installed because if you have too many apps installed it becomes really u a bit chaotic. So then um what I mentioned before Ayatana indicators top right corner. So um this is this is the daytime indicators. So if you have I that that is a blank device but if you have if you have the if you use the calendar software the events of the calendar will appear as previews in the daytime indicator. Uh you can access the the the clock app from there. You can access your alarms. they are viewed in there if you if you set alarms with your clock and so on. So and there are more indicators for network for sound for battery status uh for session management. Um so on on the right side you can see Deco which is the it's test m mainly so ignore any data you see it's just me. Um so and um this is Deco and Deco can is the the example app for convergence. So Deco can be full screen, it can be half of the screen, it can be just a little column. It always adjusts adjusts to the session and to its to its size, which is really nice to to observe. So um and then um it is possible to um to switch. Let me just see. I think I think I left out one of the screenshots. So you can if you go to the indicators and on the very right there's the system indicator and you can there's a button called computer mode and if you disable that you can enforce tablet mode uh for the device and this means that the um the the all apps go full screen but there is something that's called a side stage area. So you can have a full screen app, but on the very left there's space reserved for an app like a chat client or the email program or your two-actor authentication application. So everything that just comes with a list of information uh fits in there pretty well. The calendar, the weather app. So, and this is uh also Deco, but you can actually use Deco in this very slim space on the right and you can access the the the configuration menu and uh or the folder you can access the the mail folders and so on. So, you can you can then uh either with the mouse you move the mouse to the very right or you use on a touchcreen you just swipe. you can you can flick the pages in the side stage and if you flick more you can flick the stages uh the the the applications. So now I'm sort of moving in or moving the mouse pointer to the very left. Um you can you can sort of access the applications there. Um Lumier supports multiple workspaces. So um and also an external display is added as a workspace and and then doesn't have the EDP1 like like this one here. But then there's an HDMI one, HDMI one. Um, so um, and then you can you can use either your finger or the mouse to actually drop windows on another workspace. And most of this actually can be handled via the keyboard, which is also quite nice. So um, future prospects. So as every open source project, we would like to have more uh more people that that feel, hey, this is really my cup of tea. I would like to chime in and and become part of this. So we are looking forward to anyone who's getting interested in contributions and bug reports or translators. So what what such a project can have. Yeah. Um so and then we would like to get all the stuff we have been staging in Debian now get that into Ubuntu touch. So Mia 2 and the boutio sync framework and we really need to port over to QD6 also upstream because upstream will at some point jump to Ubuntu 2604 and they so 244 will come this year 2604 as a as an operating system base will will drop QD5 that's at least the announcement I heard. So um the QD6 is a thing. QD6 will be way faster we expect compared to QD5 because that's what happened in KDE. So I assume we will experience the same. Um so for Debian 13 and also prospect for Debian 14, we would like to lend fixes of severe UI glitches over the Debian 13 release cycle. So whenever we find a patch for something that uh reduces a usability problem, we will bring this uh at least as long as Debian 14 is not released. We will try to backport, not to backport to bring this into Debian 13 and convince the Debian stable release team to actually have those patches. So I hope this works. Um then we want to update um Lumiri components in Debian unstable targeting the Debian Fortune release and for them some of these components for example for the indicators we plan to actually have the newer versions as long as it's possible compatibility wise wise we want to have them in Trixie backports so that you can actually mix your your Debian Trixie system with Trixie backports and then you get more previews of what will be in Debian 14 [snorts] and of course um at the moment we two Debian developers who have who do uploads for Lumier of course. Yeah, if you feel like that and also I've been in in rough contact or in in in not really um uh committed contact but sort of interesting and promising contact with the Mobian people around Debian. So I would really love to see a Mobian that could be run with Lumiere and so we could actually uh get ahead with that because that would really be um it would be a competitive to to Ubuntouch but uh it would be good for the Lumiere ecosystem. So um if you're not on Debian which I'm assume you are mostly um we have some other contributors who also help with this project like it was not just the Debian packaging it was also um uh so we had one guy from Nix uh sorry uh one one one woman from Nixo Kosima and um and she contributed um a lot a lot of CMake work to make actually Lumier components not just build and install on a FHS based system but on Nexos so in whatever paths they use. Um then we have um Alpine because the people from Postmarket OS are deeply involved in in getting Lumiere onto Postmarket OS because they have it they have gotten it. We have someone who has done it for Arch Linux. We have oh we have a typo. We have a a a um an interest group in Fedora. Um um and they they are sort of at the very beginning of packaging the the Lumiri stack and ah the top one. So of course Debian unstable packages will trickle through to Debian uh to to Ubuntu and so and from what I yesterday checked checked yesterday that the current development version of Ubuntu so targeting 2510 is having will have the same package versions as we currently have in Debian Trixie. So I haven't tested it, but I'm pretty sure that 2510 will be quite comparable to what we have in in Debian 13. So and so last but not least, if you if you if you want to try that or for whatever reason you want to test that, we have backported all the Lumier packages that we maintain for unstable. We have uh sucked them over into a launchpad PPA and rebuild them against uh Ubuntu 244 for LTS and you can install you can install Ubuntu 244 uh server a minimal system add PPA and say up install Lumiri and you reboot and you're ready to go. So this is also something that was partly achieved by this project. So last but not least, I would like to mention um a cooperation with OSSCI. They are we we are planning a training program and it was very it's similar to the talk that was here before from a university in Brazil where they where they taught uh where they yeah work with students to actually contribute to open source projects. We would like to achieve the same and have Lumiere as a starting point. So um attract people to open source development to Lumiri that is OSCI and via OCI we will also offer a bounty program and if you're interested in that um yeah take a photo of the link on screen or note it somewhere somehow else. Um so that's that and of course I'm just giving this talk. I'm I haven't done this on my own. So it's just the team of it's a mix of upstream developers in Lumier. Um let me mention Rajinan who is the Ubuntu touch release manager and um so and he's he's just the greatest meticulous code reviewer I've ever met and it's so he's actually the Mr. Quality in the project which is really good to have but all the others are really awesome as well and it's good to work with them. So this is via video a big thanks to the team and um and now I'm through with the talk and I can open up the room for questions, comments, feedback would be nice. >> Thanks so far. Please stand up. >> Hi, I have a Libra M5. I want to have Deb on it. maybe Mobian and the next layer would be Lemorei >> Lumiri >> but that's the way to go. >> Um >> that's a stack I have to build. >> So so basically are you are you a mobian developer? >> No. >> No. Okay. So so maybe let me let me get in touch with one guy who's who was stuck at the airport. Um Heii and he already built a mobian image with Lumier for the Fairphone 5 not for the Libra M. And I think I think it has to be handled in the mobian project and then you get images. But I'm not a mobian developer. Evangelos, you should know probably. But we can discuss this later. Yeah. >> Acknowledge. >> But it's it's it's a plan. >> Uh it's about Chromium. Why did you choose Chromium not sock or Firefox? Oh, you mean in when I showed the installation commands? Um, so the first reason is because it was in the screenshot by accident and I realized later. So I added the text, not the screenshot. Um, so um, personally it's it's maybe a personal preference because I I I'm quite more used to working with Chromium. So I have it installed on my systems that I use for daily work. Um and yeah so from here on it's a personal opinion about Chromo and Firefox and how bloated Firefox has become but let me stop here. Okay good. >> Uh can Lumiere apps applications be used in other desktops or part of them? Which ones? >> Yes and no. So so the the overall answer is yes except from some functionalities. So for example um so one one example is um we don't have a telegram client as a dev package in Debian 13 anymore because the telegram desktop has been removed. I use telegram we use telegram in the UB ports community for interaction. So and I'm not a fan of having something installed as a snap or as a flat pack if I don't really have to. So we have a native uh telegram client in Numeiri which is called teleports and it can be installed and I use it on a different desktop environment. But so the problem is if you want to add an attachment and send a photo into the chat or something that only works on Lumiri because the interaction with other Lumier apps is not fluent. It can be amended but yeah so so yes and no. Hi, question from my side um slightly um biased of course um how do you do your graphical testing at this moment and if not so we can talk later on >> yeah so um um keep the mic please um so are you addressing something like open QA or so >> for example >> yeah we don't have that okay >> so there was so in at canonical times there was something that's called um autop is it Right. Yeah, I think autopilot and which was a bit like that. So it fired up the application and it would sort of do stuff automatically in the background and then it would check is it actually looking as it's supposed to look but we are not having that. So open QA would be a good thing but there is mostly is actually that um that we so in with the upstream hat on um we have contributors we let them send screenshots we sort of test the stuff if it's supposed if it looks like it's supposed to look but it's not automized. >> Okay thank you but we can talk later. >> Yeah. Are there any further questions on the mobile device? getting uh data plan activated network connectivity making it possible where has the work to be be be done at the lumeies layers asking the kernel or is it uh more far away to the telecom provider allowing >> yeah so you you're referring to why it's not working on Debian so so honorable with the touch phones. Uh we have a working telephone stack completely. You can do phone calls, you can send text messages, MMS and you can of course do mobile data connections but in Debian we have a different and the phonono and and I have different hardware. So I have these mbim compliant modems in so I have a Lenovo device for testing. So one thing is the modem does not have the telephoneonyy part. So I can't I won't be able to do phone calls but I'd like to be able to to connect via mobile data but doesn't have the support for this MBM stuff yet. They are working on it and we have one contributor in from Ubuntu touch who is actually contributing to upstream to land that but these patches are still in the makings and once if they appear I'll try to get them through to Debian 13 but I can't promise that. Maybe it's then also a phono from Trixie backports that will support that and okay and the stack is Lumier indicator network it's the UI underneath is a mix of a phonono and network manager and network manager controlling a phonono so it's partially directly from the network indicator to a phonono partially via network manager and uh and then yeah it's like you know from other desktops it's network manager at the end >> uh Okay. So, no obstruction from the telecom uh providers. Okay. Thanks. >> Mhm. >> Um in the near or far future, would you see Lumier as a competitor to Gnome and KDA? >> Definitely. Okay. Actually, I I I'd rather go further. Um, so I have tried using recent Gnome on tablets and I'm not so happy with that compared to how happy I'm with Lumri on Abuntu touch devices or on on tablets based on Lumri and Debian. So um and I think also like with every desktop environment it becomes a matter of taste. So yeah >> and uh will there be some support uh in the few future maybe far future for model manager instead of for phono? >> This is not planned at the moment. Yeah it so the thing is it was it was started with a phonono and we are really you know heavily tied into a phone at the moment. Um and the the so getting a phonono fixed is is is is the bigger goal actually to to have this as as an option and actually we have a problem here because if you install motor manager and a phonono on the same system they will silently conflict and uh so and suddenly aono won't do anything and modem manager will also be some sort of inhibited so um so we actually added a conflict into the oppono package that whenever you install a phonono get rid of modem manager because unexplainable errors happen then they both try to sort of sit on top of the modem. That's a problem. Yeah. So, so while the mic is being moved, um, so just let me let me mention if people are here who are experienced in QD5 or on video Q5, QT6 transitions of from QD5 to QD6 or QML development, we are really looking for developers and we are happy to have them. And as I said, the Lumiere project is financially backed. So there is probably also a possibility to get freelancing set up. Um just one comment for the people who run Morbian uh with the Debian Trixie release you can just app get installer and then it will appear on the login screen. >> You can select it and it more or less works. Yeah. >> Awesome. Great. Thanks Andre. >> So will it remove model manager? Certainly yes because of the dependency conflict that we added otherwise it won't work at all. Thank you for the very very insightful and great talk. Um we can continue in the post. Um see you in the next talks. >> Thanks for being around. >> [applause]