Sebastian Kügler: KDE’s Plasma Mobile is running on Plasma 5 and Kubuntu

The KDE Community just announced its ambitious Plasma Mobile platform. To better understand the project I reached out to Sebastian Kügler, one of Plasma’s lead architects and developers, and Chief Operations at Blue Systems. Following is a detailed and exclusive interview about Plasma Mobile.

Sebastian Kügler

Swapnil: Is Plasma Mobile a decentralized project like many other KDE projects or is there a project leader to lead the development?

Sebastian Kügler: The project has been initiated by Blue Systems as internal project, and once we deemed it good enough and useful (basically beyond a proof-of-concept stage), we have moved all project assets to KDE infrastructure and will continue development from there on. It’s not a decentralized project, it’s supposed to be a project run by those who do the work and who have useful contributions, much in the sense of KDE’s moniker of “those who do the work decide”.

Swapnil: Is this the same codebase used by Seigo’s Plasma Tablet or is it different? 

SK: Much of the code is the same, but unlike the Vivaldi codebase, it’s running on KDE Frameworks 5 (which wasn’t available back then), it runs on Wayland (which also wasn’t an option due to maturity of Wayland itself and KDE’s implementation of the Wayland protocols). We’ve preserved, lifted, updated and improved much of the existing code to fit the new objectives.

Swapnil: Which OS is being used for Plasma Mobile? Is it Mer or Kubuntu?

SK: Our reference implementation is based on Kubuntu. An earlier proof-of-concept ran on top of Mer, but in a consolidation step, we’ve moved to Kubuntu. Blue Systems has been sponsoring a couple of developers in the Kubuntu team, and this move allowed us to tap into existing packaging, QA, system integration, deployment and testing processes. We do welcome and encourage others to create a Plasma Mobile implementation on top of other OSes, and we will actively help with that.

Swapnil: Mobile is all about apps; so what kind of apps are available for the platform?

SK: We haven’t focused on specific apps, but rather worked on the app support stack to support a wide variety of apps on Plasma Mobile. Perhaps the most interesting part in that are Android apps, which we are “fairly close” to have support for. So rather than creating everything from scratch, we want to tap into existing ecosystems.

We also have preliminary support for some apps for Ubuntu, and we’ve implemented support for any app using X11 (but obviously the app needs to be able to deal with touch input and screen size and resolution). This is what we mean with “inclusive”.

Of course, in order to achieve the highest possible degree of integration, KDE and Plasma apps can and will stand out. Our mobile version of the Okular document viewer, the Koko image gallery and some others can really shine, and we’re making them available.

We also want to give KDE developers a “home platform” for their applications, but we realize that we currently, due to market share, are not a very rewarding target in terms of users. Still, we need to get past a critical mass to appeal to application developers, and we hope to achieve that by allowing a wide variety of apps, so that we get important functionality from the get-go.

Swapnil: Many members of the KDE community have created successful commercial products, for example ownCloud. What is the commercial prospects of Plasma Mobile as any commercial incentive will help boost adoption?

SK: We want to get Plasma Mobile to a level where it’s close to market quality, and due to its Free nature, is an interesting platform for hardware vendors to ship on their devices. Right now, it’s too early to call it a commercial product, but it’s something we want to achieve in terms of quality, design, support for third party apps (opening up the Android app ecosystem is a biggie) and openness of processes. We know that many hardware vendors are currently at the mercy of Google’s Android, and that does come with risks to them. We want to offer an alternative for them, but are also open to suggestions and to collaboration projects.

Swapnil: KDE has been talking about ‘convergence’ much before Canonical, so will ‘traditional’ Linux/KDE apps, let’s say Kate or Konsole, run on Plasma Mobile? 

SK: Konsole already does (we have included it in our reference image since it can come in really handy during development), and I suppose Kate would run just as well. These two apps aren’t the best examples, though, since they don’t change their UI based on formfactors — it’s a bit like running LibreOffice in its current form on a phone: it runs, but it’s of limited use and not up to mobile OS UI standards.

Swapnil: As I can see Blue Systems initiated the project, is there any corporate involvement?

SK: As we’re only announcing the project now, outside of Blue Systems, there is no corporate involvement yet. We do want to bring Plasma Mobile to a point where it does make economical sense to get involved, and we welcome other companies to take part.

Swapnil: There was Bodega for Plasma Tablet, will there be a dedicated ‘App Store’ for Plasma Mobile – something more polished than KDE-apps? SK: We haven’t worked on this topic, but I’m sure this will come up during Akademy this week. It’s also an excellent starting point, should anybody want to get involved. That said, we ship a modified version of the Muon package manager, so the user doesn’t have to drop to the command line to install or remove an app.

Swapnil: How do you see this reaching out to masses? Are you planning to work with hardware vendors, the way Canonical does, to bring the OS to the market?

SK: Honestly, we have to do a lot more work first, we’ve just reached a point where we can prove technical viability. Bringing it to the market means that we have to bring it to product-level functionality and quality. If a hardware vendor showed up on my doorstep and wanted to work with us on shipping Plasma Mobile on a device, that would of course be most welcome, but we realize that the product is too raw and has too many rough edges currently to make that very likely. Perhaps in a year, this will look different.