Plasma Mobile

Doesn't Google have Voice Recognition APIs? Would it be alright to use their services in KDE?

Those are what the VDG has made so far, there have been UI mockups for a lot of software from the file manager to software center that are leaps and bounds better than what are in place, but the manpower in general KDE is really lacking to keep up with the VDG's mockups to actually implement them.

My corporation has a strict KDE/Qt-first policy for any linux apps, due to the useability and comfort that comes from a KDE desktop.

I still keep my users at KDE 4.x for now, because 5.x is not quite ready yet. but it's very close.

so, I guess this is me actually disagreeing with your actual disagreement.

http://plasma-active.org/

Some history, Plasma Active is about five years old now and its development coincided with a would-have-been-crowdfunded-toay tablet called Vivaldi that was supposed to be an open hardware device that never panned out because costs got out of control and interest waned.

It was based on Mer, rather than Kubuntu, and Qt4 rather than 5. Today it looks like a colossal wreck, and all the "Active" app UIs developed for it are all complete wastes of code and time today because QML was not mature enough when they made that "first attempt". Today there are common themed QML elements called the qt-quick-controls that everyone can use to build UIs that adapt to every devices native toolkit, while still supporting animations and flow elements and all the nice graphical perks hardware accelerated UIs allow.

Its been basically dead in the water for over two years, since the tablet project went belly up, and there even used to be a "Kubuntu-active" fork of Kubuntu that the project was maintaining as a way to try out the Plasma Active desktop on top of a Kubuntu core. The shell from Plasma Active did eventually see use in its adoption as the "netbook" interface found in Plasma 4 near the end of its lifecycle.

So you definitely want some post-mortum on the last KDE mobile attempt and you also want to consider how Plasma Mobile might succeed or fail in a similar vein.

Why did Plasma Active basically never do anything? Number one, no device support. It was not targeting phones at the time, and was instead looking towards tablets, but it never even ran properly on a Nexus 7 and those only turned up while it was on the decline. Without hardware, software is useless. This time around Plasma Mobile is shipping for the Nexus 5 out of the gate, a significant improvement. If they can make images for popular dev phones for a while, they will certainly have more potential devs than Active ever did, which pretty much amounted to building and running the shell in a child Kwin to tinker with, but never use on an actual device.

Number two, its UI was a wreck. This was before the KDE VDG, Breeze, and Material Design were a thing, so Plasma Active was built on ugly Oxygen and Active Apps were made with effectively prototype QML1 where you had to write your own button class. As a result, all the apps not made for Active (and most weren't, since it was dead in the water) were their desktop versions and unusable on small touch screens, and all the Active apps looked like you drew some rectangles on a white background you could click because that is literally how you made buttons in QML1.

Plasma Mobile does improve on this as well. Breeze, the new KDE theme, beats the pants out of their older defaults. They now have a KDE font, and a KDE icon theme meant to encompass everything, that are all consistent. they now have their visual design group making mockups and directing design for new apps and old, though there is an obvious schism between designers making mockups and anyone actually implementing them - there is an amazing Muon design that was never implemented into the final product which looks and feels like butt. And not a single core KDE app in the KDE Software releases is shipping a QML2 powered UI that is meant for mobile that uses a common design language like Material. It means all the core KDE apps - Kontact, Dolphin, whatever web browser you want to use (Rekonq is dead, Konqeror is a zombie, and there are several new projects out there - for mobile, you might be able to get away with a dumb qtwebengine wrapper), Amarok (which also has another amazing VDG UI floating around that is still unimplemented), Dragon Player, etc. They all need mobile UIs, none of them have them now, and you wouldn't even want to make a mobile UI, you would want to make adaptive UIs - you would want all these programs to be able to scale from phone size to desktop size and readjust their layouts accordingly, because that is what QML2 is really good at. Its also a complexity nightmare, if anyone has tried a responsive website, albeit QML2 does so much of a better job equipping you with the tools to do that its almost not comparable.

Lastly, Plasma Active never had an app store. At all. Muon was still in its infancy back then, so you were using... Apper. Which is basically a worse Synaptic package manager, and Synaptic does not belong in the same room as a mobile phone UI. The newer appstream infrastructure and the future xdg-app tech should be great when it lands, but like I said before Muon is still a mess, though maybe that responsive adaptive UI redesign will also bring the VDG designs to the forefront finally and make it really great.

But Plasma Mobile supporting Ubuntu Touch Apps, Sailfish Apps, Firefox OS HTML5 apps, and Android apps with slashtik will definitely solve the app problem. Security might become a problem, then, though. You are now mixing debs, click packages, apks, app manifests, and RPMs from Sailfish on one system. Sounds like a giant mess. So do you just containerize everything? Good luck with that limited mobile phone storage and memory size. How does apparmor integrate with that mess? How about how Muon integrates knewstuff content from kdelook and kdeapps - those are literally just zipfiles it extracts into predetermined locations.

A lot of questions, but as long as they don't go over their heads and stay within their manpower - which if Active is any indication is that you really just need a core base working great before trying to take over the world - it should have much more potential than the last KDE foray into mobile.

The dream, of course, is that Sailfish / Ubuntu Touch / Firefox OS all eventually rebase off Plasma Mobile and just provide their own UI shells over it, so that all the major open source mobile players are working in one ecosystem rather than working in their own trees at their own companies not trying to play ball with the competition. Good luck with that one.

I think the main reason that smartphones still work is that that most people use them in a really reduced way. One could argue for feature hiding, which is a good thing. However hiding I think means "moving it out of the way" and not "make sure nobody finds it".

I am not saying that Android and iOS are bad, but more that there is room for improvement. It's good to see new projects doing things a bit different. I always found it weird when people complained about copying. I don't think that's bad, but right now there doesn't seem to be a lot of progress. I like how KDE at least tries to do things a bit different instead of copying without understanding why someone designed their UX a certain way. Not every design works for every project/product.

I've also heard a lot good about Windows Phones, even though this caused fewer apps that really integrated. It's great when Microsoft tries new stuff, like in Windows 8. Yeah, it didn't work out, but the UX designes that sometimes are called innovative don't seem to be on par with the amount of research going into other fields of the big OSs. It may also be due to the market settling, but there is still room for improvements, so I hope it's not stagnating. It would be great to see other players.

I know different styles make thins harder for people doing cross-platform applications. Sticking to status quo makes things easier, but it also leads to stagnation and it feels like KDE4 and Windows 8 made a lot of people scared of trying out new stuff.