AUA: We are the Plasma dev team. Ask Us Anything about Plasma 6, gear 24.02, Frameworks 6 and everything else in the upcoming Megarelease.
David, Nate, Josh, Marco, Carl, and Niccolò are here ready to answer all your questions on Plasma (all versions), Gear, Frameworks, Wayland (and how it affects KDE's software), and everything in between.
Fire away, Lemmy!
We were expecting to be done in an hour and we have past the 2-hour mark already! Time flies when you are having fun.
Thank you for all the questions and the welcoming and friendly atmosphere, but the devs must get back to making Plasma 6 great.
Please keep the conversation going and KDE contributors will continue to answer over the next days as time permits.
With Wayland becoming the "Default" for most distributions now, will KDE begin to integrate some Wayland only features that you're excited about?
I've seem some very interesting experiments for swapping desktop sessions (GNOME to KDE to Sway whole CSGO was running) all without losing state, and storing application state to disk.
GPU Hotplugging is one I'm pretty hyped for! Hybrid graphics laptops aren't too different from a GPU hotplugging scenario either, and they're usually quite janky on Linux getting the external displays to work usually means swapping from hybrid mode to dGPU mode (restarting the X session in the process). At least with Wayland that (eventually) be a lot more graceful!
on the side of our apps (or anything written with Qt) all of that (and most important seamless compositor restart/crash recovery with the application surviving) everyhting should be there for 6.0. with other toolkits the mileage might vary (depending when they include the required changes, when a given distribution packages them and so on)
Josh says: "Yes, we are always interested in making our applications mobile-ready and almost every new KDE application uses Kirigami our convergent framework. Some of our older applications such as Okular, Dolphin, etc need more work on mobile but this is something that's being worked on."
We sure do plan of moving more and more of our app to the new convergent ui toolkit made with QML and Kirigami, in the future more and more of our apps should become mobile ready
In addition to that Josh said, we have a list of mobile friendly first party applications here: https://plasma-mobile.org/ It's not completely up to date and is missing some newer additions.
is there any plans for tok to return? the need for a proper telegram client is a lot , telegram desktop cannot be as good as a native client on mobile screen.
Plasma Vaults! It's the best implementation of having a little encrypted bucket to put your important files in that I've ever used, on any platform. It's very well integrated into Plasma as a 1st-party supported feature, and it works wonderfully.
I use plasma vaults! Its great for homework folders and tax information!
One frustration with vaults though is that theres no clean way to make a portable vault on a USB stick or backed up to a cloud provider (nextcloud, google drive, etc) without digging into weird dot-folder paths and manually entering links to these in a text config file. FUSE-style integration would be rad.
EDIT: The primary use case for this would be to be able to carry sensitive information around like PII, tax, password vaults, family photos, documents, and so on, in such a way that you always have it on you (like on a keychain) or backed up elsewhere, and would be especially useful in cases of disaster - but if you drop and lose it somewhere, a malicious actor doesnt suddenly have your data.
It has not been developed in the last few years but always krunner!
A recent-ish feature that was already in Plasma 5 is to bind key presses to your extra mouse buttons or tablet tool buttons
Moving to Linux w/ a Logitech MX3 mouse, I legit spent 2-3 days troubleshooting across solaar and piper before I got logiops to work. I'll disable it and dig around the KDE settings to try for a more integrated solution that is easily editable.
KDE Connect is something I keep my eye on and check in every once in a while: Is there a dedicated page tracking updates specifically to KDE Connect? I’m really very much looking forward to a time where it is feature compete with Android compared to Apple’s continuity platform. I would absolutely love to move to Android+Linux as my daily drivers, but I feel like I’m giving up on too much by leaving the Apple ecosystem.
Does Connect use BTLE?
Could you guys implement an auto tethering option between phone/PC?
How instant are notifications synced? Do notifications disappear on one side or the other when viewed on one or the other platform?
Maybe implement a “link to KDE” notification toggle to mirror the “link to windows” functionality of Android?
Is there a dedicated page tracking updates specifically to KDE Connect?
It is part of KDE Gear, so generally in the gear release announcement. The last few releases were not that big in term of feature but the next one includes some goodies.
Does Connect use BTLE?
KDE Connect will have the Bluetooth backend enabled by default with the next gear release (24.02). I'm not sure if this is BTLE or normal Bluetooth.
Could you guys implement an auto tethering option between phone/PC?
No idea :(
How instant are notifications synced?
For me it is pretty instant. I never miss my Bereal notifications thanks to it :)
Do notifications disappear on one side or the other when viewed on one or the other platform?
Yes
Maybe implement a “link to KDE” notification toggle to mirror the “link to windows” functionality of Android?
One of the most visible ones for me is that most common multimonitor workflows Just Work™ in the Wayland session now. There are still edge cases, but we've put a huge amount of effort into this.
To just test out without any risk and not touching any running system i would suggest a live image such as KDE Neon Unstable which has dailiy updated snapshots of our software stack
as moving to Qt6 from a developer POV, It has been remarkably uneventful. there are api changes for sure (Especially on the QML side of things) but the changes are not as great as say, Qt4 to Qt5.
What’s the best or recommended way to test out Plasma 6 RC2?
Neon Testing in a VM (or on bare metal if you're adventurous). Arch with the kde-unstable repo is good too, but that also includes a snapshot of the unreleased Qt 6.7 which introduces more bugs.
What has been the hardest problem to solve moving to Qt6?
Personally I'd have to say the large number of API and behavior changes in QtQuick that Qt 6 has brought. We use QtQuick very heavily throughout KDE, so this has required a lot of mandatory porting work, more than in our QtWidgets-based software. And there have even been changes between Qt 6.5, 6.6, and 6.7, so it's still a bit of a moving target
A near term plan is to make it so you don't have to use any Vulkan layer to make it work, by having gamescope use the protocol directly. After that, add support to all the other clients - Mesa, SDL, glfw, mpv, Krita, Chromium and so on
will there be an easy way to get HDR working with the Steam flatpak for the Plasma 6 launch?
You can run Steam in gamescope, doesn't matter if it's sandboxed or not.
Thanks for the response, excited to see more support being added!
You can run Steam in gamescope, doesn’t matter if it’s sandboxed or not.
So, installing your Vulkan layers and gamescope-git on the host and running ENABLE_HDR_WSI=1 gamescope --hdr-enabled --hdr-debug-force-output --steam -- env ENABLE_GAMESCOPE_WSI=1 DXVK_HDR=1 DISABLE_HDR_WSI=1 flatpak run com.valvesoftware.Steam -bigpicture should do it on Plasma 6?
There's been an undercurrent of discussion I've noticed (and tend to agree) that C++ is not the most friendly language for new/outside contributors, even the newer standard you use. Is there an effort to bring first class support/documentation for another higher level language as well? e.g dart/ruby/lua/python/javascript/C#/go (whatever is most feasible)
Something that's often not mentioned is that C++ with Qt is often a very different beast to use compared to C++ with the stdlib and other GUI frameworks. IMO Qt takes a lot of the pain out of C++, such that the criticism becomes blunted and mostly articulated by people looking in from outside who haven't tried it yet.
that suggests if nothing else, that point needs to be articulated more. But also then, C++ with QT is a very specific skill people now have to learn to help out, as opposed to stuff that would be more universally useful
I have some very, very, very, very WIP KDE widgets for Dart/Flutter in a project running. Currently it has almost nothing but it already automatically switches color theme when you switch color schemes in KDE for example 😄
Don't expect too much of that though, it's very unofficial and something I'm doing whenever I feel like it.
Some of the plasmoids are implemented in JavaScript I think, but personally I'd love to see Dart as a consideration for that "high level language" choice.
It already does a great job at gluing high level rendering abstractions with low level rendering libraries within the Flutter SDK, and supports a lot of useful paradigms in UI development such as first class functions, null safety, enhanced enumerators, etc.
I can see why it's given a wide berth from FOSS since it's associated with JavaScript AND Google (two controversial terms in programming to say the least...) but working with it daily really does remind me how well it lends itself to frontend development.
Thanks for all the great work on kde!
I was wondering if the KDE theme refresh, (brise) and icon (app and plasma icon) proposals are going to happen in plasma 6.1? Or is that not on the table?
Also, are there plans to have KDE connect support making your phone a webcam?
Yeah the only way to do it right now is using droidcam, and I took one look at the documentation and gave up, and continued using my bad webcam. Also afiak the new pipewire video portals should make that feature easier to implement.
Neon unstable/testing already contains the relevant branches. Neon user edition should receive it as normal when it releases.
When Neon will update its Ubuntnu base is a question for Neon developers.
Are there any big things that you couldn’t squeeze in time for the mega release, but can’t wait to have them implemented in Plasma 6.1 later this year?
Wouldn't that be changing the corresponding color entries in the colorscheme? At least all of Plasma 5's window decorations seem to respect those.
Stuff like selection color's already changed so it doesn't seem too difficult?
In regards to theming, there's one thing that sticks like a sore thumb, and it's the fact that most if not all themes for Qt currently use the Kvantum engine ( github.com/tsujan/Kvantum ), which sadly is not well integrated with KDE in things as basic as supporting Plasma's color schemes and dynamic icons. Do you have any plans to collaborate directly with Kvantum, or with any other particular theming team other than Breeze's?
A challenge for integrating with Kvantum is that it isn't great for theming QtQuick due to bugs in our existing creaky theming infrastructure, as well as development direction preferences on the part of Kvantum's maintainer.
One thing we have planned and want to work on during the Plasma 6 lifecycle is a new unified theming system that can apply to all KDE and Qt apps, GTK apps, and Plasma. The idea is to have a new theme that can be directly consumed by Qt's Qstyle (for QtWidgets apps), KDE's QtQuick desktop style (for desktop QtQuick apps), KDE's Plasma style (for Plasma), as well as KDE's GTK theme. Essentially we would end up with a new theming engine and each of the existing themes we have would consume those themes. This would replace the current approach where the C++ QStyle is the central source of truth and our QtQuick desktop style pulls content from it, while our Plasma and GTK themes are totally separate and have to be changed manually.
The new proposal is in fact not unlike how Kvantum already works, but it's not rally made for easy upstreaming and it also uses SVG as the basis for its themes. We'd like to build our own thing and investigate using CSS as the basis for themes.
Needless to say, this is not happening for Plasma 6.0. :) But I'm hoping we can get it done sometime in the next year or three.
Actually Plasma is generally more popular than GNOME every time surveys are conducted. However we have to keep in mind that the direct consumers of a DE are actually not the end users, but rather the distributors who package and distribute it. There are a number of historical reasons why many distributors ended up picking GNOME over Plasma including accessibility, corporate sponsorship, an easier packaging experience, and the rocky KDE 4 rollout burning a lot of trust. So what you end up with today is many distros shipping GNOME despite pent-up desire for Plasma. It's a great illustration of how you need to keep your direct users happy.
And I think that pent-up desire is being unleashed these days due to various changes in our ecosystem. Plasma is better than ever and version 5 had a much less painful release compared to 4, with us aiming to do even better in Plasma 6. We also see an increasing number of hardware vendors shipping devices with Plasma on it (https://kde.org/hardware/), who had a strong financial incentive to listen to their customers by picking Plasma over GNOME. In addition, KDE's accessibility game is ramping up hugely, and we have more robust corporate sponsorship than we used to with Valve and Blue Systems putting tons of resources into KDE. Finally, GNOME seems to be becoming more hostile to their downstreams, causing them to need to do more of their own development or else migrate to be a fork or skin of Plasma. Interesting developments.
We do aim to improve our design and usability further as much as possible, however, one of the nice things of free software is really this big choice.
There are different projects and one size never fits all, if some people find the software written by our friends over GNOME more suited with their needs, that's totally fine.
It would also be interesting hearing on the motivations for this choice tough, as it always help us improving
@Bro666 With the transition from QT5 to 6 and Plasma 5.27's implementation of wp-fractional-scale-v1, will most KDE apps now be able to fractionally scale without using the upscale-downscale method? (Or in other words, how QT apps fractionally scale under X11)
@davidre While KDE doesn't oversee LibreOffice development, since they also have a QT6 backend for their UI, might they also benefit from this? Currently fractional scaled LibreOffice under Wayland with the QT5 backend is extremely laggy, presumably due to the upscale-downscale method.
Thanks for the AUA and all the work on KDE, especially with plasma 6.
It's not really a question, more of a request, but is it possible you all could make the kwrite/kate situation less complicated? I don't mind having options, but from what I understand, you can only have them both installed (or none), not one or the other. Also, if I want Kwrite to be the default editor, I have to go into every supported mime type (file association) and switch the default (instead of just once for all of them).
My preference would be that there's only one text editor (to cause less confusion; maybe just have a toggle button to switch between basic and advanced modes), and to make it easier to change file associations for common file types (like text files, video files, music, images, etc.); even if it's technically a different file format, e.g. .mp4 vs .mkv for videos, I should only have to set the default app once.
You can have only one installed, and Fedora KDE ships this way. If your distro doesn’t offer that, it seems like a distro packaging issue.
Thanks for the info; it does look like that's how my distro packages it.
make it easier to change file associations for common file types
That’s in Plasma 6. :)
That's awesome to hear! It's a problem I've had for a while, so I'm glad it's being worked on.
Anyway, I just want to say I really appreciate all the work y'all have done over the years; I've been using plasma for the past 6-7 years, and it's a desktop environment like no other.
In addition to the obvious answer of "because our software is really good!", IMO an under-appreciated reason is that KDE truly is an anarchic and largely volunteer-run community. As long as there are passionate volunteers, there will be KDE; you don't have to worry about it just dying one day should some big corporation pull the plug for some reason. We've all become so accustomed these days to software being disposable, but KDE really does give you a measure of longevity and continuity that you're unlikely to get elsewhere, especially without paying a lot of money for it.
I think everyone has different preferences on how they use their machine. However I would invite to just try Plasma out. One thing where it stands out is in my opinion that you can tailor it to your wishes and use case.
Thanks for the AUA! Really excited about the Mega Release and even more for the future after!
My question is regarding Nate's original proposal for the KDE goals "Professionalise KDE" if I remember well. Now, I know that the goal was changed and the scope reduced to "Automatization and Systematization" but I truly believe the original proposal's spirit could have a huge impact on the whole ecosystem.
For example I remember the proposal contained the plan to hire a person working "full time" on actively finding fundings and grants and applying to them.
We got the Mega Release fundraiser but for example on the forum I found https://discuss.kde.org/t/formalities-for-stf-funding/9916 and it looks like a waste of potential to not apply for grants because no one has the time to do so.
Is there anyone at KDE, maybe among the Board, who is still trying to make the "Professionalise KDE" happen?
P.S. I know you guys just want to reach the biggest amount of people but maybe we could start having the AUAs/AMAs and similar, only on Lemmy and a post on Reddit saying that the event will be hosted on Lemmy. This would make the Lemmy community the official one and could convince people to make the switch for the "exclusive content". It's not a critic, just want to see Lemmy prosper.
As a Board member now, I'll answer this one. To a certain extent I ran for a seat on the Board because I realized that proposing this via a Goal was the wrong place to do it, and the more effective way to push for that change was by being on the Board. And now that I've been elected there, yes I do still want to do it and push for it internally where possible. However now that I have a fuller picture of the e.V.'s situation, I realize that there are budgetary concerns that must be met before we do more hiring, and KDE e.V. operates under extremely strict German nonprofit rules that make it not as simple as it might appear. Now, in an ideal world this person would be effectively self-funding, but we need to make sure we can afford them in the first place! It's a bit of a chicken-egg situation, really.
Note that we've already done a certain amount of professionalizing KDE inasmuch as it means "KDE e.V. hires professionals to work on KDE." KDE e.V. does indeed now employ multiple technical engineers to work on various areas of the software stack, and I want to see this grow even more. But, we need more money to make it happen, and until we do have that employed professional fundraiser, it's up to existing members of the Board and the community to improve the revenue side of the equation so that it becomes possible! That's why I'm so happy with the progress of the current fundraiser. 700 new members means at least 70,000€ of new recurring yearly income, which is enormous for KDE e.V. If we can keep up this kind of fundraising performance, we can do so much cool stuff in the next few years.
I'm someone that uses sway because I really like to keep my workflow as keyboard driven as possible. However, I do not necessarily need the minimalism of sway and sometimes miss the more polished experience of a full DE. Are there ways to configure Plasma to give me a sway-like keyboard-driven flow while still getting some of the other benefits of Plasma?