One of the key points of Plasma is while giving a simple default desktop experience, not limiting the user to that single, pre-packed one size fits all UI.
Its strength is to be flexible to greatly different user needs, “Simple by default, powerful when needed”.
Several years ago, the Visual Design Group had the idea of making easy to build and share desktop layouts to make easy to test wildly different user interfaces, see this old post by Thomas on the topic.
Since then, work on it has been going on, mostly on the infrastructure needed to make it a reality, and in Plasma 5.8 the first pieces are there, tough still far from the complete experience we want to offer.
The support for Look and Feel packages is there since a while (5.3 or so) that’s what one of those package can do:
Provide a default layout for when Plasma starts for the first time, it was used for distributions to personalise their UI, but now is easier for users as well.
Provide some default look options, like what color scheme to use, what icon theme etc
(advanced) provide the actual implementation of some UI, such as KRunner, the Alt+Tab window switcher dialog, the lock screen
So far the default Plasma layout provided by the Look and Feel theme was used only when starting up for the first time, on a clean user home (therefore very useful for distributions) but sice Plasma 5.8, in the Workspace theme -> Look & Feel section of system settings there is an option to load the new layout when switching the look and feel theme. Not as default as is a destructive action that will remove your current Desktop setup.
The other component is a tiny little application shipped in the “Plasma Sdk” package that’s called lookandfeelexplorer.
With this applciation you can:
Create a new Look and Feel theme
Edit the metadata and thumbnail of a locally created/installed theme
Create a defaults file based upon your current setup as well, such as color scheme and icon theme
The last two are the central part of sharing your idea of “the perfect desktop” with others, linked with the integration between the Look & Feel systemsetting module and the KDE store, also new in Plasma 5.8.
It’s still a preliminary feature, as ideally in the future if your shared Look & Feel theme depends for instance from a particular icon theme or a particular 3rd party plasmoid, the store integration will download those dependencies as well.
In the screenshot below, the desktop/Breeze-looking scrollbar it’s actually a control coming from QtQuickControls2.
Since one of the goals of Kirigami is to support also desktop applications, while working on Kirigami 2 (that is mostly porting it to be based on top of QtQuickControls2) I had the need for it to continue to integrate with desktop applications as well.
Unfortunately, desktops are not the primary target for QtQuickcontrls2, and while in Qt 5.9 some much needed desktop-related features, such as mouseover effects are getting back in, being as unrecognizable as possible with QWidget-based applications is definitely not in the roadmap.
Luckily, theming QQC2 is easy, so I’ve started a QtQuickControls2 style that integrates with desktop QStyle-based widgets (actually, based upon the QtQuickControls1 QStyle painter machinery).
If we are serious in using QtQuickControls2 in applications shipped together Plasma, we also must make sure they integrate well with the Plasma desktop’s Breeze look and feel.
I started by theming the scrollbar because I felt it was one of the hardest controls to do.
in QtQuickControls2 the scrollbars are supposed to be an attached property of flickables..
This is pretty simple for the mobile use case, turns out it’s a nightmare to create scrollbars that behave like we are used to in the desktop (unless you’re on mac, which insists to look more and more like iOS), which need to be
have optional little arrow buttons for increment/decrenent
be besides the flickable, not overlapping the contents, because again, they’re always visible
use the global desktop theme
I’ve been able to do all of the following, tough it’s necessary to reintroduce the QtQuickcontrols1 component called ScrollView (which imo was a mistake to remove in QtQuickControls2) which can correctly lay out and size the Flickable relative to its ScrollBar.
Hopefully, you’ll be soon able to build an application with QtQuickcontrols2 and Kirigami2 that feels right at home in a Plasma desktop.
What about Plasma-themed QQC2 controls for plasmoids? that will come too, of course!
Today the first feature update of Kirigami has been released.
We have a lot of bug fixes and some cool new features:
The Menu class features some changes and fixes which give greater control over the action triggered by submenus and leaf nodes in the menu tree. Submenus now know which entry is their parent, and allow the submenu’s view to be reset when the application needs it to.
The OverlaySheet now allows to embed ListView and GridView instances in it as well.
The Drawer width now is standardized so all applications look coherent from one another and the title now elides if it doesn’t fit. We also introduced the GlobalDrawer.bannerClicked signal to let applications react to banner interaction.
SwipeListItem has been polished to make sure its contents can fit to the space they have and we introduced the Separator component.
Desktop Kirigami applications support the “quit” shortcut (such as Ctrl+Q) now.
Plasma 5.8 will depend from Kirigami 1.1, so if you are planning to write a Kirigami-based application, it will work by default and nicely integrate in the Plasma 5.8 desktop.
Plasma 5.8 also has a new big user for Kirigami 1.1, that is Discover: the application to search and install for new software, has a brand new user interface, based upon Kirigami.
This is problably the last feature release based upon QtQuickControls 1, QtQuickControls 2 version is on the way at an experimental stage. The port will have way simpler code (and smaller memory footprint) but this is an entry for another day 🙂
Yesterday I did lead my first 30 meters pitch (Girotondo, 5c in Finale Ligure) that’s quite another level of scary compared with the ~15 meters stuff I was used to, and quite literally “pushing your comfort zone”.
When you see the last quickdraw you placed 3 meters below you (and the ground 20-something) and the next bolt just a couple of moves away.. just a couple too much, all fibers in your body tell you “give up”, fingers are starting to slip, they want to slip.
But you really rather not want to give up, because again, the last quickdraw you placed is 3 meters below you, so the fall will be very long (and scary, and potentially painful), and because what the hell, you came all the way up there, just to give up?
You just say fuck pain, fuck the fear, and just keep moving forward, one slow move at a time, just try to clip a quickdraw to one last bolt, and then I’ll maybe give up there, after all even if I fail, yes the fall will be long and scary and potentially painful, but at least I know I failed trying. All the safety precautions were taken (is a surprisingly safe thing to do, once you go manically through the checklist, it’s a *very* security conscious community) so while accidents do happen, usually the “worst thing that can happen” is “not very much”. In that controlled environment you learn to replace pretty quickly the “should do this”, “should have done that when I could” with the “just go for it”.
Breeze icons are very simple SVG files, especially the ones used for actions that are mostly monochromatic are very simple, and that’s part of their appeal.
Since some time, Plasma themes that are SVG files as well have the capability of being colored with system themes.
So, why not doing this for every icon as well?
One problem with monochromatic icons is that they can lose contrast in particular situation: when the application uses another color scheme or in places such as in menu items uder the mouse, that get a blue background, giving a not too visible dark gray on dark blue.
That’s one of those “last mile” polishing issues that may be small, but have quite a big impact on the perceived quality of the finished product.
Starting with Plasma 5.7 Icons will behave the same way as Plasma themes: they can have an internal stylesheet which colors will be replaced at runtime with the colors from the system theme.
Here, Dolphin with the colors theme “Wonton Soup” and all the breeze icons following the text color of the theme:
Some applications, like Gwenview can use a completely custom color scheme, in the case of GwenView, it switches to a dark color scheme when fullscreen, regardless of the normal system color theme:
Icons that are in a “selected” state such as the menu item under the mouse or the current dolphin sidebar item change their color, just like the text does too:
Here with some custom colors for the highlight areas:
How to create a compatible icon?
First of all, Big kudos to Andreas for updating the whole breeze theme to the standard described below 😀
This is a quite minimal SVG file that supports colors from the system theme (Plasma Svg themes follow the same convention):
The colors defined in Text, Background, Highlight etc will be replaced with the corresponding colors from the system theme (look at the Systemsettings module to configure colors to see what colors they actually are)
This also defines a very minimal palette of “semantic” colors: you have the foreground and background colors for purely monochromatic shapes, and other colors for small accents that make the icon slightly more expressive than purely monochromatic such as Highlight, positive, negative (an “x” icon to close will be usually of “negative” color)
The other important part is the actual definition of the shape:
The rectangle has class=”ColorScheme-Text” that will make the stylesheet match to the class definition of ColorScheme-Text, that defines a color.
In order to actually apply that color, you can see the attribute style=”fill:currentColor”. It’s important no other colors are defined in the style attribute.
This post wanders quite off topic… but has a lot of pretty pictures indeed 😉
Sometimes I ask myself why I am a software developer, the answer in the end is that I always enjoyed creating things, whatever it is going from being sketched out totally in my imagination to finally seeing it in the flesh. Writing software, especially graphical software can be very satisfying exactly because of this mental process of seeing the thing you thought about slowly forming and starting to actually working, with the gap between the mental image and the real thing narrowing more and more (yeah, I’m one of those heavy visual thinkers that can think almost exclusively by images).
But sometimes nothing can replace the satisfaction to create an actual, beautiful object, and I also feel manual skills are something that we should cultivate much more, I feel more complete if every now and then I do something that a) I don’t know anything about when I start and b) it’s a difficult manual skill to craft.
That’s also why before drawing the svg for the Kirigami banner
I “had” to make some experiments of an actual kirigami…
But this post is not about that.
Almost a year ago, a friend of mine told me that he wanted to learn a bit to hack on some simple Arduino stuff, and you know what? I wanted too.
I have this stubborn quality that makes me to go quite overboard when I decide to do something (especially if is not for myself) and not to stop until is done, actually useful and pretty, so if we are doing some Arduino project, let’s do something that has an use and that will be pretty… and that’s how the project “Bagnur” started (means watering can in Piedmontese language).
Of course it needed a logo :p
The project is one that is seen again and again on the interwebs, so is just a remix of existing ideas rather than something truly innovative: the Arduino takes data from a soil moisture sensor (different humidity in the soil changes current conductivity), to figure out how much water the soil of a potted plant has.
When is dry enough it opens a solenoid based valve to pour enough water in, and the moisture sampling goes on and on, hopefully stopping you from killing those poor potted plants from thirst after forgetting watering them for days 😉
The final hand made steampunk-like package
The moisture sensor is based on the LM393 chip, the solenoid valve is similar to this one (hilariously overspec for this project, love this kind of overkill).
So we know the Arduino output pins have a too weak current to operate the valve, it will have to be powered separatedly: the Arduino will close its circuit with a transistor, I had just salvaged an E13007 NPN transistor that had useful charateristics (low base-emitter saturation voltage, resists quite heavy loads) from a broken ATX power supply.
This makes things a bit more interesting, luckily in the end a single power source was enough to power the Arduino and the valve in parallel, probably not particularly recomended, but cheap and compact (having 2 different power bricks for such a sillyness wouldn’t have been particularly fun for day to day use).
To make the project a bit more interesting, we have a potentiometer that will regulate how much water the plant needs, different plants, different needs and one of those pretty RGB LEDs, that will be the “output UI” of the thing.
The state of the led will be:
Fading from pure red (soil bone dry) to pure green (soil soaking wet) with all the values in between
Fading to blue when the valve is open, and the plant is being watered
The LED will stay usually at a very low power, the sensor will do a reading every 10 seconds or so, when this happens the LED will fade to full power with a nice animation
For the final thing, I decided to use an Arduino Nano compatible board, based as well on the ATMega 328 chip, since I don’t need many pins or performance, it just has to be as small, low power and cheap as possible.
In this Github repo there are both the source code of the program running in the arduino and a couple of schematics drawn with Fritzing, so it should be of easy replication and improvement… if someone really wants to 😉
The final board will be in an hand-made case made of wood that gives it a cool almost “steampunk” look: that was honestly for me the most fun part of the project of all, probably because was the farthest from what I usually do, the most low-tech, hand skill demanding part (and slightly dangerous… yep, it involves rotating blades :p).
Video of the thing in action
This videos shows some assembling steps and the thing in action, both opening the valve automatically based on the moisture sensor values and manually with the button on top.
Let’s start from the usual prototyping with the classic Arduino Uno/breadboard combination (here still with different colored leds instead of the final single RGB one)
After the breadboard prototype is done, let’s start to lay out the components on a perfboard, to make it look more like a final prototype, usable day to day
First components attached to the perfboard. The Arduino Nano will connect it as a daughter board.
Yep!, i definitely need to improve my soldering skills, but I swear, even if is kinda ugly, it works like a charm 😉
Still some components missing, already attached to the wooden base
As I mentioned, for me the most interesting part was to build a wooden case from scratch, from a raw plank of wood, trying to master some of the classic woodwork tecniques.
The plank is made thinner and smoother via a thickness planer; after this step the setup of the machine changes becoming a surface planer for a more precise and smoother retouching. The process repeats until the wood reaches the desired thickness.
Did i mention, ROTATING BLADES?
All the sides of the box are done
After the box has been glued, it has to be tied very, very tightly as the glue dries
The joints are not very precise …yet
…But some aggressive sandpapering for sure helps, best if done with a Sander
The lid of the box with a translucent, water proof paint, attached to the top are the moisture sensor, the RGB LED and a button to manually open the valve.
The complete device, ready to be closed. The wires on the left control the RGB LED, the blue and green wires in the middle control the button, the wires on the right control the moisture sensor.
The completed little, mean machine.
It has been an insanely fun project, and I am sure my technique can still improve in all areas (designing an electronics board, soldering, woodwork…). If I’ll keep trying to improve with new projects I don’t know yet, but the recomendation I can make is get out of the comfort zone of your day-to-day work and experiment: you have only to gain, if only for the act of doing things wrong, without which there is no learning.
Over the second week of March I’ve been at the sprint at CERN.
It has been an amazing experience seeing those very big toys, where the cutting edge research is done (noted with satisfaction the presence of Plasma desktops in the CMS control room)
On my side, some interesting little things happened:
All new systemtray finished and merged
During the sprint I’ve merged a thing i was working since a while: the system tray of Plasma was one of the most complicated plasmoids out there due to the very peculiar things it does.
Its code was really showing its age (it surviced at least 3 portings across different technologies) and even tough the old Xembed-based systray icon protocol was dropped, its architecture was still decidedly all
It has now been completely rewritten, its code is now way simpler, using less layers of proxymodels and went from ~2000 locs of C++ to ~300
While completely new, the users shouldn’t even notice any UI change, the only noticeable change should be less bugs and working better 😉
During the sprint, a new repository was born.
What was Plasma Mobile components is now residing in a separate git repository: https://quickgit.kde.org/?p=kirigami.git Kirigami (The names comes from a Japanese paper folding craft similar to Origami, but unlike Origami cutting the paper is allowed) is a set of QtQuick components at the moment targeted for mobile use (in the future desktop as well) targeting both Plasma Mobile and Android. It’s not a whole set of components, all the “Primitive” ones like buttons and textboxes are a job for QtQuickControls (soon QtQuickContrls2) but it’s a set of high level components to make as easy as possible making applications that look gorgeous on mobile devices that follow the Visual Design Group UI guidelines.
The target of those components is anybody that wants to do an application using QtQuick as its main UI, especially if targeting a mobile platform, without adding many dependencies. They work both in Plasma Mobile and Android.
It will eventually become a Tier-1 KDE Framework.
While I was refining the components, it turns out a piece of desktop software just has its first release of its Android port, it is already using a tech preview of the Kirigami components: it’s Subsurface a dive log software started some years ago by Linus Torvalds (in GTK+) and recently ported over Qt (here a talk by one of its main developers Dirk Hohndel about the porting process https://www.youtube.com/watch?v=ON0A1dsQOV0)
It’s awesome having already an early adopter (which has been a pleasure to work with) for the components and also means we are getting a ton of feedback on it.
An informational tooltip in Plasma is an item that shows extra informations for items such as task items, and is a single entity, moving along what you want to know more of, rather than magically appear out of nowhere (that’s the kind of “magic” the human brain doesn’t like).
A problem with Plasma tooltips was that they tried to animate themselves, that is usually not a good idea on X11, due to its old async architecture.
But KWin to the rescue! if we want a smooth animation of both the position and the size of the tooltip, the compositor is the place where to do it.
You can see it in this video taskbar tooltips animating and resizing/morphing in a similar way Windows 7 does.
Btw, the animations in he video looks way less smooth than are in reality, due how terrible screencasting is under X11, but for that:
The Plasma theme system had a feature (since many years, actually) in which SVG elements done in a certain way can be recolored with colors coming from a theme file.
The Breeze Plasma theme (and now all the monochrome Breeze icons too) was all done in this way, in part to prepare what I’m, presenting today:
If the colors in the SVG can follow a color scheme defined in the theme, they can follow also a system wide color theme no?
For Plasma 5.6, (as a feature that was requested really a lot) the default Breeze theme, while by looking familiar, it will change color following the applications scheme.
However, if you prefer to maintain a clear distinction between the workspace and the applications (And I’m definitely among them), there are still available the themes “Breeze Light” and “Breeze Dark”, just as before (Oxygen and Air also received some nice visual updates).
Let’s look at some screenshot:
So far so normal, typical Breeze theme we had so far.
Quoting from the announcement:
“KDE Neon is the intersection of these needs using a stable Ubuntu long-term release as its core, packaging the hottest software fresh from the KDE Community ovens. Compute knowing you have a solid foundation and enjoy the features you experience in the world’s most customisable desktop.”
This is pretty exciting for anyone who wants a stable core system with a setup of KDE Plasma software on to as recent as possible, setted-up and configured as good as possible, with hopefully less issues like “distro X has a slightly outdated version of kibrary Y which is know that makes app Z crash”.
Hell, I was so excited that during a sleepless night, I has been completely possessed by the Muse and I had to do this artwork based on a riff of its logo:
NEON written as a neon sign, how original :p It’s pretty tacky but it’s intended to be, here for your viewing pleasure at a typical wallpaper resolution.