Yes. It's all about transparency. I can see exactly what KDE is exporting, so I'm willing to help KDE. I cannot say the same for closed source software.
I really like gnome the software, but I've started considering moving away from it after a decade simply because of how toxic and difficult gnome the project can be.
I also really like GNOME the software but I moved away a few months ago because of this.
As is, the current GNOME is unusable to me without extensions because they refuse to implement support for appindicators. You literally cannot use applications that minimize to tray on vanilla GNOME right now.
They have been talking about adding their own protocol for years but that is of no use when things are broken right now.
Important features and bug fixes are always stuck in merge request limbo for years. VRR for Wayland got merged recently after 4 years and it's still experimental. DRM leasing is still missing on Wayland, KDE added it 3 years ago.
The final straw was when KDE announced HDR support last year I switched over because I knew GNOME would probably lag behind by months or even years.
As is, the current GNOME is unusable to me without extensions because they refuse to implement support for appindicators. You literally cannot use applications that minimize to tray on vanilla GNOME right now.
They have been talking about adding their own protocol for years but that is of no use when things are broken right now.
So what, just use the extension. Currently no cross-desktop API for systrays that doesn't suck in one or another way exists, so GNOME doesn't have support for them. If you care that much about not using an extension, implement it for yourself.
Yeah, just because the api is not perfect, to just not support it, is no solution.
With that argument you can just skip most interop api, as they all have pain points.
What PR? And what about the missing API that satisfies every/most desktops' needs?
And any GNOME user who needs that can use the extension. I don't really get the point, apart from philosophy, which doesn't really make sense here since nothing perfect exists yet, which GNOME seemingly doesn't like implementing. Maybe some work towards that would be good, but I'm just someone using software for free, without paying anything.
You can check this post post about why gnome has done away with appindicators. Basically everyone has their own and it's a mess, they're very much not bringing them back, appindicators are being replaced altogether by the notification system.
@imecth@cullmann@ohyran@UnityDevice@domi It wasn't true when Allan wrote that blog post, and it's still not true now. If you drop XEmbed and only support SNI (as Plasma did years ago), you have one way to handle it. As it is, Fedora Workstation has an open ticket about adding the appindicator extension because applications are broken without it and Ubuntu maintains and ships it to support a useful user experience.
Currently the ticket is deferred until we resolve updating the SNI spec.
I'm aware of their reason for dropping support but it's not sensible to drop a functioning system and replace it with nothing and then talk about how to do it better for years. That post is from 2017, it's 2024 now and there is still no replacement in sight.
You've missed the part where they have no intention of replacing it. It's bloat. And I agree with them.
Where relevant they've added stuff as a core part of the panel, like recently an indicator for VPN connections. If you want to use an application you can alt-tab to it, like we've done for decades. Everything else is relegated to media controls and notifications. Appindicators are legacy at this point, and they systematically get cut from modern designs like mobiles.
Notifications, you can have the app fire a notification when it's synced or disconnects for example. Gnome is working on better notifications right now. Tablets, chromebooks, cell phones... have been doing fine without appindicators; people just have a hard time changing their habits.
Notifications are more effective at displaying a change of status than a tiny icon turning red.
What's important to someone is gonna vary on a case by case basis, sometimes getting an email is an urgent notification, you can easily turn off the ones you don't care for or go into DND mode.
At least for us, notifications aren't something you can really glance at similarly to app indicators. They're usually text heavy, only really work for longer tasks for readability (which syncing usually isn't), and are always obscured behind another popup for persistent notifications. Persistent notifications also take up more space within the notifications popup, rather than a small icon that you can easily glance at to know what's happening.
As for programs not staying in the task manager, they usually take up less space if open as an app indicator, being able to be passively open but not take up as much space.
The problem is when you allow one developer its own applet, every application wants one, and suddenly you have 15 applets. Applications need to figure out alternative design patterns to achieve the same result or sidestep the problem.
There's this saying, out of sight, out of mind, do you really need to have a constant eye on every application? When there's an actual change you get a notification.
@imecth
Alternative design patterns like PUTTING A SINGLE ICON ON THE TASK BAR SO USERS CAN SEE AT ANY GIVEN MOMENT WHAT A SERVICE OF THEIR CHOICE IS DOING. Like Windows, MacOS, iPadOS, iPhoneOS, ReactOS, Kde Plasma, XFCE, and morr allow?
Do you honestly think an icon bar like this is a good thing? Look at the colors, the amount of them, how they fold because there's too many... And it's the same shit on windows too. It looks ugly, they're hard to click on, most of them don't serve any purpose... I agree appindicators do serve a purpose, but as it is, i prefer not having them at all.
To more directly answer, the only icons I would hide there would be the chrome and Pale Moon ones. However, not having seen those two before, I don't know what they are indicating. I would mouse over them or click on them to see what they are.
Now I am a KDE fanboy to the bone, a KDE eV member and past contributor to several projects ... so I am kinda biased :D so "yes, yes you should" THAT SAID I know a lot of awesome folks in the GNOME project. People who really really are brilliant and fantastic folks the issue is that there is a culture of "be loudest and most self-assured and you're the best" in certain aspects of the project and combined with the GNOME projects stated focus on just GNOME that creates an air of snobbery among some (sadly some of the people most outwardly visible) and a tendency to demand help from others but refusing to give it when asked.
Its a cycle of self-proclaimed victimhood too where they consider any disagreement as either "unprofessional" or just random hostility without reason when it comes from the outside.
Which sucks. Sucks amazingly. Specifically because there are so many great folks in the project doing awesome things for others and the GNOME project who seem doomed to obscurity because of their ability to work with others and not be blustering screaming malcontents due to the projects culture (in certain areas).
EDIT: just to hammer the point home. Amazing project, amazing people but for some reason a handful of people who from the outside look like random asshats have been actively promoted to the top. Perhaps within the project they don't appear as asshats? I don't know. I just know that I have a very very short list of people that I avoid and would leave a project if they where in it because I have seen what they do when in power. Three of that less-than-five list are from the GNOME projects leadership.
And they added back Tango as a fallback and the bug is fixed because now its FDO-compatible... I am pretending the snark at the end by Jakob isn't there and its all good
Absolutely, yep! I curse myself every time I just click “extract” forgetting that other file managers don’t do this, and end up with files all over the place
Just add HideDesktopPreviewBanner=true to the [General] section in the ~/.config/kdeglobals file and restart plasmashell with systemctl restart --user plasma-plasmashell.service or just log out and back in.
It would be nice if it were at least configurable to set as the default extract option. If I had to take a guess, it’d be that it’s not the default option because the amount of single files before needing a subfolder could vary between different people. Some folks may want only one, and others may be fine if it goes up to say 3. However, I suppose that could also just be a configurable option.
That being said, I’ve at the very least developed the muscle memory to always click that option no matter what. I can’t tell by your comment if you weren’t aware of the feature, but if not then hopefully it can be of use to you moving forward as well!
kscreen-doctor -o
Then adjust the below command accordingly : kscreen-doctor output.1.hdr.disable
Make sure you go open a bug report.
When you do, you can use hw-probe and run : sudo -E hw-probe -all -upload -dump-acpi -decode-acpi
And use the given link in your BR.
I see you got it fixed.
When you run hw-prope make sure your offending monitor is connected. hw-prope will help identify the monitor by hardware ID which is used in various areas like kernel space and will help with identifying the exact model of the problem monitor hopefully allowing for devs to reproduce the bug in a lab environment.
Looking through the probe logs, and seeing that your monitor is using EISA bus and works fine without HDR, there doesn't appear to be any issues on the Linux side of things. My guess is that they didn't implement HDR on the monitors side exactly to spec and that's where the problem resides. So, in this sense some monitor specific quirk fixup code is needed on the Linux side of things to get it working properly. If the devs ask any additional steps from you, be sure to do it and provide feedback.
No, you can run most KDE apps on other systems, including Windows and Mac. I use Kate as my text editor on my windows work machine.
I used to be a KDE dev. We were largely volunteers, unlike a lot of other FOSS projects that had hired coders. The KDE e.V. funding largely went to server maintenance and helping students attend the annual conference (travel expenses! I benefitted from this a few times). Not sure if it’s still like that. In my era, KDE could easily get by on less.
Probably a lot more stuff would get done in KDE with more money, no? Big things like moving away from Bugzilla, supporting more languages like Python and Rust for KDE app development, (way) better documentation, marketing, ads, fulltime employees (marketing, UX/UI designers, developers, sysadmins, devops, lawyers, etc.).
I’d love to work for KDE for example, but without having to first contribute to it for years, get recognised by some important community members, give talks, and then finally maybe see some money to work fullfime on a project. There are probably many, many developers who would rather write opensource code fulltime, remotely and be paid a livable wage instead of toiling away in some for-profit business writing proprietary code built on top of opensource and never contributing back to the greater good.
Is there any reason to move away from Bugzilla? Afaik the reason why they’re not using GitLab issues it’s missing some features they need, which Bugzilla has. Also afaict the language thing is more of a choice than anything. Qt already has excellent Python support, but having everything written in C++ and QML makes things easier. But yes, they definitely could use more money and more paid developers. KDE could really use more manpower.
And yeah, a ton of devs would much rather be working on open source software, but if it’s not directly profitable it’s not gonna generate a lot of jobs. You need a lot of donations just to hire one developer full time. There’s always going to be a lot more jobs in closed source software than in open source.
UI and UX are terrible. Filing a bug against the right application is really difficult for me. Imagine how hard it is for the average user. Account creation is not easy (it sends you back and forth through multiple domains with multiple emails), finding the right project to file a report against takes a lot of clicks and guessing (is badly detected monitor a graphics problem, a desktop problem, a software screen configurator problem, a distro problem, or a linux kernel problem?), then it is not clear how to find or add logs, and I have never been able to get KDE’s crash handler to send off a bug report.
Gitlab and Bugzilla are no the the only projects that exist for this. I don’t have suggestions right now as I’m just describing my experience.
Also afaict the language thing is more of a choice than anything
It’s a lack of choice. Qt’s support is not the issue, it’s KDE’s support for Python that’s the issue. Try writing a KDE widget or component in Python. Where is the KDE Python API documentation? github.com/KDE/pykde5 last commit is from 10 years ago. What is invent.kde.org/kdevelop/kdev-python ? The README just says “read installation instructions”.
KDE Frameworks only has documentation for “C++ with Qt and QML”. The API documentation does not have a single mention of Rust nor Python.
You need a lot of donations just to hire one developer full time. There’s always going to be a lot more jobs in closed source software than in open source.
What I’m saying is that KDE could learn from Thunderbird. Imagine a budget of 6.5 million dollars for KDE. The money is there, it’s in people’s pockets, but there’s a reason they don’t donate to KDE.
I donate periodically to KDE, but my major gripe is that I don’t know where the money is going. They have no financial reports that can be easily found, individual projects don’t have a donation button, there’s no public tracking of their income or expenditure like on opencollective, and it’s not easy to find KDE devs (aka who is actually on the KDE team) so that one could sponsor individual devs.
Although I trust KDE more than Mozilla (MZ pays their CEO 7 million/year and invests in anything but Firefox, their most known project), I would much much much rather prefer it to know where the money goes.
You misunderstood. I’m aware KDE doesn’t support languages other than QML and C++. I’m saying this is probably a choice on their behalf to make things easier to maintain. Whether or not this choice is good is up for debate, but I don’t think it is the result of a lack of funding.
As for where your money goes, their donation page has a section on this which also links to annual reports. Some larger KDE projects do have their own donation buttons (like Kdenlive and Krita), but most don’t. A lot are just too small for it to be worth it, but this is still an area that can be improved. As for the “KDE Team”, that’s a pretty nebulous concept. I don’t think there is one, officially. KDE is a community of mostly volunteers working on KDE software. Most of the ones paid to work on KDE aren’t paid by donations, but are hired by third parties to work on KDE, like Blue Systems or Valve. Right now I believe most of the money is spent on infrastructure and events. Afaik, most of the people paid directly by donations are Krita and Kdenlive devs, since that’s what they raised their own funds for. KDE wants to start hiring more developers tho and I imagine their successful Plasma 6 fundraiser will allow them to do this.
Also, keep in mind the fact that Mozilla software including Thunderbird is just significantly more popular than pretty much anything KDE makes. Yes, KDE is infinitely larger in scope than Thunderbird is, but Thunderbird sees a lot more mainstream professional use than any piece of KDE software and that’s what gets you money.
@parens@troyunrau Exactly, it took me 5 years from my first contribution to finally getting hired full time to work on some KDE and KDE related open source code. And I'm very prolific in term of contributions (https://invent.kde.org/carlschwan).
It's quite an issue to get money, the current budget KDE e.V. has is around 200 000€ a year and that pays for a few part time employees. I'm lucky that I found a company interested in developing a commercial open source project based on KDE (gnupg/g10 code).
Most KDE apps are Linux-only. More popular apps will often have some support for other operating systems, but that support is only good for the most popular apps, like Kate, Krita, Okular or Kdenlive.
There used to be (in the 4.x days) a general installer which allowed pretty much who whole KDE ecosystem to be installed on windows. Does that not exist anymore? I used to use Okteta on windows this way :)
I’ve heard of this before, but as far as I know that no longer exists. Most of the KDE apps available on other operating systems can be found here: binary-factory.kde.org. Okteta is indeed one of them, available on both macOS and Windows. There are definitely a fair amount, but many of these ports are in rough shape and this is still a small portion of total number of KDE apps available on Linux.
They have KDE connect and some other tools but I’m not aware of how popular they are even for Linux users. A lot of it is duplication of other already popular tools available. And their bittorrent client sucks in my humble opinion.
I would expect for most users KDE is just another window manager for Linux. Secondly it’s not even the most popular. So I would expect its user numbers to be much smaller.
Yeah that’s a fair point. I’m currently using kubuntu and enjoy it more than other distros I’ve tried. But I’ve never felt a strong desire to adopt all the tools. It took me a long while to realize that the themes store is more or less dead and theme authors expect you to install a 3rd party tool (Kvantum) to get the real theme. I wish they would make that more obvious, cause when you’re browsing themes in KDE none of that is obvious and you left wondering why it didn’t do much.
That is strange. It is resizing for me depending on the payment option I choose, showing all the info without the need of scrolling. I am using the desktop version of Firefox.
Krita has been downloaded in the millions. Kdenlive is getting there, as is KDE Connect since they now support Windows and macOS. GCompris is used by 10s of millions of kids in schools and homes all over the world.
What’s wrong with KTorrent? Also KDE has a ton of apps (like 200 of them last time I counted). Many of them are certainly the most popular tool for what they do on Linux. On other operating systems, they’re usually not very popular, tho. Krita is probably the most popular of the bunch and might even be more popular than Plasma. It raises its own funds independently from the rest of KDE, tho. It’s also probably still nowhere near Thunderbird’s popularity.
It is difficult to get hard figures about these things, as KDE software is downloaded from many different sources, many of which don’t track details because of privacy concerns (we don’t track personal data of any kind either, by the way).
But we can infer indirectly what the most used (and probably most downloaded) package is and that would be GCompris by a very long way. We know this because
we have the confirmation from the educational department of one of the most populated provinces in India (specifically Kerala) where GCompris used daily by approximately 4 million pupils. We also know it is used extensively in other areas of India.
it is the software that pops up organically more often on non-FOSS social media platforms where teachers hang out (for example, Instagram). Through these interactions, we know it is used extensively in schools and by parents in most of the world.
Second in the list would be Krita, again by inferring from several sources, like the number of downloads from the Microsoft Store (most s Krita users are on Windows), Google Play Store, and familiarity with the software in design-themed social media groups.
Most people who use these apps are not aware that KDE exists and are not familiar with the concept of Open Source. For them it is just some software they found or was recommended that happens to be free.
Plasma is not at the top of the list, but usage is probably in the millions thanks to things like the Steam Deck and large installation in public institutions (including throughout NASA, CERN and again the computer labs of many schools).
But, then again, it makes sense that determined apps be used more than the desktop environment: installing and using an app for a specific purpose (to play educational activities or paint a pictures, for example) is much, much easier and familiar as a task for end users than installing a whole new OS and desktop environment on top.
KDE
Top