 Thank you very much. So one of the for those who don't know me so well, I'm working on plasma Okay, so I work on plasma project and So getting community started working on plasmo project in 2015. It was introduced as a community project in 2015 academy The goal was the project is to full control over information of the mobile devices and communication and It should be inclusive to third party software and not like some closed-down systems like iOS and Also, same list cross-device experience or convergence as you call it and Also, we were we wanted to use the open standard switch So Today's my talk is about plasma mobile on the open devices. It doesn't actually focus much on the software side but actually more on the hardware and basically It doesn't even actually focus on the plasma mobile itself, but the the layers be below the plasma And so as a background I took liberty of copying two slides from my last year's presentation so last year's presentation I basically provided updates about my updates about work done in one year and at that time I listed out some External factors that affect the plasma by project and some of the challenges we were facing at that point so the External factors listed were physically the devices which are currently in the market have a really old Colonel wasn't it's not reasonable to support them For example, lots of people here will have a Nexus 5 which supports a 3.4 Colonel. So, yeah, that's a ancient Colonel and We also have a lack of open devices the devices where we don't have some kind of device schematics and Where you don't really know what's happening behind the scenes So that makes it really hard to put something totally different than than what manufacturer wanted to put on that device and Also, like those devices requires a closed-source binary blobs and PSP's to function and so this is a this was a External factors and Also, I listed out some of the open devices or the programs which were Which were known at that point The one is open devices program by Sony where they are basically trying to open source as much as they can other program is the It's a community effort to support as much as men line Do I as much as devices we can support in the main line Linux Colonel and There is also fair phone which provides open device schematics and and Basically, you can replace the parts of it yourself and Also, of course, it was a question mark at that point because it was merely a rumor that Purism will be building a Linux phone. But of course, it's not no longer a question mark. So that question mark should not be there So back to present So back to in this one year we had to work with the several projects and several community members and several corporations as well to To get To make it easier for open software to be present on the mobile devices It's entirely basically not structured effort like it. So basically different contributors working on totally different issues and trying to solve it so When it comes to the off the self devices, which are currently in the market Those comes with the old kernel version, but the Linux kernel have a But in theory it can be put it can be supported by the current main line kernel version just you need to do some work to add Socy system supporting the for example, Colonel if it's that you need to you need to add to hardware supports and so basically One of some of the Factors are DRM and GPU drivers. For example, there is a There is a free trainer project to which is reverse engineer driver for the adrino welcome devices There are Wi-Fi and Bluetooth drivers, which needs to be open sourced into mainline kernel And of course, that is the phone part, which is the modern and telephony So for that there is a project called Ophono, which we use in plus mobile. So more and more Drivers should be added into Ophono and Also, there is a Well every hardware in the mobile for example camera sensors and everything So well, I'm at that point it was the next in the past year we Had supported many Qualcomm devices with the mainline kernel You might have read my blog post in planet.qt.org so the Qualcomm devices basically The idea came to my mind when well, basically I was starting to work on plasma mobile That was a blog post by John Which is so that Nexus 7 2013 model was running 2013 model was running a mainline kernel and the Android So I thought like if it's a mainline kernel we will have an open drivers and in theory plasma mobile could work online and Of course, I had no access to Nexus 7 2013 model or that it was it was not easily available to It was not easily available in market anyway but I had an access fire device which was a reference device for plasma mobile at one point and And So I Got excited about this and like started reading up on various things and so I learned that Nexus 5 have a MSM8974 board and it was already supported to some level in into mainline kernel for So I try started by Like how can I add support for Nexus 5 into mainline kernel? Of course, that that that was a bigger technical challenge than I Did that was a really bigger technical challenge and then like it was a bigger bite than I could chew So well, I added some support Well, some support is basically you get a kernel log over the headphone jack and You can place the volume buttons, but that's of course not usable for plasma mobile So Well, due to lack of time and lack of proper knowledge. I basically serve that project but In this in past year one of the other community members managed to actually extend the work I have done and they actually managed to run a Fully mainline kernel and so with the graphics and Wi-Fi wireless and the support on the plus mobile and Then with the help of the community members we We started building a tabian based system which can run along with this mainline kernel work done by the by the community members So it was essentially thanks to people who are working on it and the communities like post market OS and Tabian community which was possible Thanks to them and those developers have now managed to port More and more two voices. So which is Sony Xperia Z2 and there is also Serious for which I'm not exactly sure of the model number and There is actually there is a so basically now we have a experimental Experimental branch experimental root file system, which can be done on such devices which we have which are already mainline It uses a fully open source mass abstract instead of the hacked up live hybrids and Hacked up live hybrids So it's a great to have such a open devices in the market and Also like you might have you might know this already so a caddy community partner with the tourism to build a build a live room five and agreed upon to and We part the tourism computers rain a Campaign which was successful and The caddy community agreed to support plus mobile on Libram five So it's expected to see pin January 2019 and it it also uses a fully mainline Kernel and fully open source drivers and in addition to that it also supports a Kills which is concept which is essential to privacy and security Goal V as a caddy community have set up So I Talked about devices and here are some of the pictures So it's a Sony Xperia's head voice I talked about it's running a Mainline Kernel version and Plasma mobile on it Then there is a neck. There is another Sony device running a post market OS and Plasma mobile and Here is another device running Plasma mobile and nexus file. Sorry not nexus file So and of course, there is a prisms development board, which is already which is also running a Plasma mobile so When we started Plasma mobile project essentially we were using lip hybrid and we were we we didn't actually explore the Opportunities to at that point we didn't actually explore the ability Possibilities to have a open devices, but now we are we are now Due to community members and the corporations. We are actually in the both position to Build upon the open devices and have a plasma while running on the search devices One of the interesting fact about such devices is it makes a development so much easier because Who is tech is so who is tech is familiar? You have a same mess our stacks. You have a same drivers. You have a same wireless drivers and you don't have to deal with the Deal with the system which which basically Where you run a Android inside the Where you run minimal Android inside the container and talk to it through various IPC and you try to convert to Cis called to Cis calls to the to the Android by my neck base system and This will also make it so much easier to have Multiple distributions to support mobile devices Currently we just have a whole for example kidney neon based root file system on the Plasma mobile but If we are we are going to follow such open standards, which are already getting used on desktop devices It will make it very easier for other distributions like tabian or our Linux or open suzay to build upon and Of course, we need a help So free you can ask you can talk to us in metrics channel plus mobile metrics dot org and Mail English. There is a platform where we're taking on We need a help on various topics. For example, mainlining such a new devices and we also need a so far we have success in running mainland control on the only Qualcomm devices, but There is effort to have a Mali based reverse engineer graphics driver Lima So we need help with such devices as well There are a lot that are there is also a community which so which can help you get started on mainlining your devices and Also, we have a post marketer's project Who have a very very good card on how to mainline your device? And Yes, before I go to the questions This is the one of the very small Thanks or if you just, you know, physically trying to restrict the work that you're doing, and I think it's okay to get your company ready if you are committed to a lot of such kind of talking, like if you don't like the search, you can't do much more than what you can as well, and you can't proceed with your own work, or you can't as well proceed with your third party. Yeah, so, we don't have any questions. We have a question for you. If I recall correctly, the login image Google does actually has a space on a contract, so it can't be done without an option of an intro card, which don't have to be opened. Which don't force to use Google's search results. Yes, I would like to, but still, if you know that the system is equalized with the system directly from the Google itself, and the entire folks and folks of folks will be rewardless, and there's no question about the actual end processes, but the main ones meet each other at the same, if they want to. So, I think it's important to get an introduction. I'm a little bit confused about the main Plasma mobile, and maybe what's a lot I've seen today is, it sounds very low level in its stuff, not much about actual Plasma. What do you do with focus sizes also while working with Plasma? Well, as a Plasma worker, I was kind of hoping to see, you know, how does Plasma work on mobile devices currently. So, for the Plasma mobile, it's a basically a system, and set up a very nice way to do that, which basically allows us to, basically allows us to store the mobile, for example, off Google, and then there's the server, the server and the user interface server, which servers most of the 90% of all the Plasma that is a shared protocol, and then we also have a Kiribati, which is sort of, which is sort of, you know, for any part of your applications, which can be used even on the end of your iOS, or even on the desktop. So, for Plasma mobile, it's basically a Plasma OS, so you know how to use your own desktop and it doesn't actually change very much. It's just a question, you know, and you can use this time of the month. Yeah, right. All right, do we have any more questions? I'm wondering if two things, actually, maybe you meant the Sony Xperia X device, which with the unknown set, because for that one, the hardware is pretty open. Sony documented a lot of stuff, and I'm also wondering, do you think about joining efforts with the Sadelfish people? About joining efforts with the Selfish West people, I think in the past, I mean, like it's a thing of past year, we started a collaborative effort with the Selfish West and Nemo Mobile and Ubuntu Touch people, which is called Halium. It's a separate to Plasma Mobile. So, it works on a solution to use... So, it works on a solution to use Plasma Mobile on basically this old hardware, where the only Android can be used. So, yes, we are basically collaborating with the Selfish West, but as far as I am aware, they are not putting any effort in the mainlining devices, basically. So, my current presentation was about the mainlining device. So, basically, you get to use very recent, kind of like 4.18 or 4.19, and not the vendor provided 4.4 or 4.9. Are we on time or...? So, there's a reason why he talks a lot about Linux low-level stuff and why he's in charge of that and why there's this sort of weird separation of where does the Plasma stop and where does the Mobile end, right? So, I hack on Plasma. The way Plasma is architected is that we have a single executable called Plasma Shell, which loads the rest of the UI from a package, which means if you have a system that has Plasma Shell installed on it, as a third-party, you can install an entire different UI into that system, right? So, this is then why we call Plasma Desktop and Plasma Mobile, not because we like lots of keystrokes, but because in our minds, those are kind of separate. Plasma is the runtime that enables you to write UI targeting a specific form factor, and Desktop, for example, is a project that implements a desktop UI for that, and Plasma Mobile is two things. It's also a project to implement a device-specific user interface, and also our effort to bring the Plasma runtime to mobile devices and all of the steps that go into that, like device image creation and so on, which he is the expert for, which is kind of why he focuses on that. Now, under your eyes, side of things, some of the things we've done in the last year are we had an optimization sprint where we worked very hard on making this runtime more performant on low-powered ARM hardware. For example, the target device we started with at the start of that sprint, which was for a week, from pressing the Power button to getting the UI at the time, we took 34 seconds. At the end of the sprint, it was down to 12 or something like that, and we cut memory usage by, I think, 150 megabytes and stuff like that. So on the runtime side of things, we've worked very hard to target a lower-power envelope. Then a lot of the work that we're doing right now are in Kirigami. Kirigami is a KDE framework. It's a user interface toolkit that extends Qt and Qt Quick. What Kirigami provides is stuff to make your applications convergent, which means that with one code base you can target both a desktop and a mobile form factor. And we've been working, for example, on porting KDE's system settings application to Kirigami and all the modules in there to make them available on Plasma Mobile. We also see more and more of our KDE applications port to Kirigami or add Kirigami-based interfaces. So as time goes on, more and more applications pop up on the Plasma Mobile device image and become more usable on the mobile form factor. As for the shell specifically, there has been some polish. It needs much more work, for sure. So the way it works is that Plasma is very, very plug-in oriented. Everything that you see on screen is in fact a plug-in. Even the notion that you have a desktop, for example, on the desktop or that you have a panel is provided by a plug-in. The plug-ins are made aware of their constraints and they can adapt to them. This gives you the ability to do something like use one code base to provide desktop icons but also have a file pop-up on your panel, right? So what we do is we reuse some of the existing UI bits we have on the desktop in the mobile shell, like, for example, the network manager UI to pick your Wi-Fi or pick your connection is reused in the mobile shell to the form factor a bit. What this has given us is a usable basic system that shares a lot of code. It needs a lot more work to be polished specifically at form factor, which the framework allows for. You can, on a file basis and on a sort of per UI screen basis, you can swap in form factor-specific URL files. And that's work that is ongoing as well. So that's like a short overview of what's going on on the UI side. And I think we're very excited to see more and more apps work on the mobile form factor. And as they pick up Kirigami to do that, Kirigami also gets better, which then creates this fortuitous cycle where it becomes more and more appealing to port your application to Kirigami. Okay, thank you very much. It was very interesting in my mind. We have a plasmobile buff on the Monday. So if you want, you can stop by. It's at room 122 at 10.30.