Mastodawn

Lyude🌹#BLM Jun 22

i am starting to think that we actually really do need a specification that defines a baseline of wayland protocols to try to get close to application developer's expectations.
I also like, really would like to see more expectations set around "please stop saying 'we're not trying to be windows'". i gotta be honest i'm starting to realize this is kind of a really dismissive response that ignores the fact that we have a protocol here that we are trying to get people to use, and people aren't really going to use something they're literally not able to develop for :\.

Lyude🌹#BLM Jun 22

People tell me this isn't an issue but then i see stuff like this:
https://www.kicad.org/blog/2025/06/KiCad-and-Wayland-Support/
that was posted -literally this month-. And it's frustrating because out of the issues listed there:
- cursor warping is and has been supported for a while across multiple compositors. It is "optional" as in "you probably won't have it on a car's in-vehicle-infotainment system.". like, the expectation is -really- not about compositors deciding not to implement it
- clipboard issues have been solved for ages now?? i literally remember which clipboard issue they're talking about because I had it on my own desktop lol

KiCad and Wayland Support

The KiCad development team frequently receives questions about our support for Wayland. Given that Fedora and Ubuntu are both planning to drop X11 support from their main desktop environments in the near future, we want to provide clear, transparent guidance to our users about the current state of Wayland support in KiCad. Current Status Is Functional but Degraded KiCad does run on Wayland systems, but with significant limitations and known issues that substantially degrade the user experience.

karolherbst 🐧 🦀Jun 22

@Lyude This is the moment where companies usually rebrand their product and then everybody is happy after the rebranding, because "this is much better, we can figure out the remaining issues"

Lyude🌹#BLM Jun 22

@karolherbst i genuinely think we need to sit down and have an actual "this is what you can expect from any wayland desktop compositor that's marked as being desktop compliant". like going into a lot of these points it's not particularly surprising that they're worried people won't have consistent support for them. we have this tiny barebones core wayland protocol and then basically whatever other protocols a compositor chooses to support and just this assumption that some optional protocols will always be supported without any clear indication to app developers

karolherbst 🐧 🦀Jun 22

@Lyude The issue is, that we don't want to enforce a trademark, and without that we have 0 leverage.

Like yes we could say "this and that is expected to be there", but if a desktop doesn't follow it, there is nothing we can do about it.

People talk about "spec body" and "you need a proper spec", but people forget those only work if you are actually willing to go to court over your trademark.

Lyude🌹#BLM Jun 22

@karolherbst then we just have the spec and have desktops like kde implement it and leave people out who don't want to play ball until they realize that no one is writing applications for their desktop.
i don't feel like this is an issue we can ignore when a lot of these applications don't even have linux as a primary userbase. unless we're ok with wine becoming the most popular API for application development on linux :\

Lyude🌹#BLM Jun 22

@karolherbst like, frankly, I do think having other compositors agree on a baseline would push things forward even if we have projects like gnome trying to avoid this. because at some point they'll still have to answer once most of their userbase is complaining that they can't even implement a basic standard everyone else agreed on. and if they don't, maybe at this point it's best we just leave them to their own devices until they can't ignore it anymore.

karolherbst 🐧 🦀

@Lyude Yeah.. it's a tough situation to be in.

The issue is, we are kinda one of the few FOSS projects who actually does something like that.

And we can't enforce anything. It's a social problem, and if desktops have different visions on what they want to achieve, then they'll just end up doing whatever they want anyway.

Like even the discussions around "veto"s. Not having them would change exactly nothing. And publicly shaming hasn't worked up until now either.

Lyude🌹#BLM Jun 22

@karolherbst it's less about enforcement and more about getting the compositors that want to play nice with eachother to just do so, and then just letting natural pressure from the userbase push other projects into starting to support it as well.
we can still say "this is what we define as Freedesktop's desktop compositor base". if kde, sway, whatever other compositors we get on board support it and projects like gnome don't, it's not that we need to force them into it. it's just we're not going to wait around

karolherbst 🐧 🦀Jun 22

@Lyude I think the biggest problem is, that people just hyperfocus on what gnome is or isn't doing.

Like why do people even care? I don't get it. Most of the rants really feel just like wanting to hate on the project for no reason besides hating on it.

People should just stop caring and do what they think is right. And if their app doesn't work on Gnome, then they can document it and just tell users to not use it on gnome and move on.

Lyude🌹#BLM Jun 22

@karolherbst It's because we're already involved in these communities such that we already know the details around all of these things. But this really isn't discoverable for most app developers, and we've had plenty of incidents that give us bad PR and in turn give people the impression that we're just not intent on making Wayland work for application developers.
Like: you and I know cursor warping is "optional" as in "you won't have it on a car's infotainment system but it's pretty much guaranteed everywhere else". But we know that because we've been following this development for ages, how are others supposed to discover this sort of thing if we don't even really have a consensus that says this is expected across wayland desktops?

karolherbst 🐧 🦀Jun 22

@Lyude normally there is https://wayland.app/ but it lists "no desktop" for "Pointer warp", but that's probably because nobody updated it 🙃

I do think we can document it, and maybe have something like "profiles" per use case, so I don't think it's necessarily a bad idea to make it easier to find what's supported across the desktop.

But I think it's mostly a documentation problem, because app developers need their own documentation (like https://wayland.app)

karolherbst 🐧 🦀Jun 22

@Lyude I was bringing up STA/STF funding on xf-bod already, maybe we should find some money to hire people writing good and proper documentation?

Janne Grunau Jun 22

@karolherbst @Lyude having a top level filter for compositors on https://wayland.app/ so that it only lists protocols supported by the selected compositors would be useful

Wayland Protocol Documentation | Wayland Explorer

A better way to read Wayland documentation

karolherbst 🐧 🦀Jun 22

@janne @Lyude yeah, sadly this project isn't affiliated with Freedesktop, sooo.. maybe check if they already have an issue filed on their github.

Though I _think_ some people were wondering if we could just move it under the wayland umbrella...

karolherbst 🐧 🦀Jun 22

@janne @Lyude Anyway, we need to hire people to write documentation for us, because that's probably not going to happen on its own.

Janne Grunau Jun 22

@karolherbst @Lyude there are few feature requests for that and https://absurdlysuspicious.github.io/wayland-protocols-table/ exists

Wayland protocols support table

karolherbst 🐧 🦀Jun 22

@janne oh, that's great. Now if we'd have a single source of truth for what's supported where...

Janne Grunau Jun 22

@karolherbst afaik it uses wayland explorer as data source

karolherbst 🐧 🦀Jun 22

@janne ah yeah.. that's wayland.app right...

We really should move it into Freedesktop or something like that 🙃

Or at least promote it better so it's kept up to date more quickly? Dunno how much of it is outdated though, maybe it's perfect, never really bothered checking.

Neal Gompa (ニール・ゴンパ)

@janne @karolherbst @Lyude This is amazing and going in my bookmarks, thanks! 😃

Matt Campbell Jun 22

@Lyude @karolherbst Since you mention a car infotainment system, I wonder if part of the problem is that, to the extent that there's actual money going into developing the Linux GUI stack, maybe it's primarily for applications like that, rather than conventional desktops. And I imagine there is especially not much funding going into developing the kind of conventional desktop environment that makes vocal power users happy.

karolherbst 🐧 🦀Jun 22

@matt @Lyude Well.. Red Hat pays a bunch of gnome developers at least.

But I am also working on trying to secure funding for Freedesktop as a hole we could maybe contract people with.

But that's something for the board to figure out if we are willing to do that and deal with the necessary paperwork.

I'd love to get STA/STF funding directly for Freedesktop, so we can do more "general desktop" things instead of it being focused on a single desktop.

Neal Gompa (ニール・ゴンパ)

@karolherbst @matt @Lyude There are a few companies directly investing in conventional desktop Linux:

* Techpaladin (KDE)
* Blue Systems (KDE)
* Red Hat, Canonical (GNOME)
* Igalia, Codethink, Collabora (GNOME)

But beyond that, I don't think there's much.

Adil Arif Jun 23

@neal There is one more.

* SUSE (GNOME)

Joan Torres may only work on GNOME part time while working at SUSE. However, that is still more than SUSE was contributing than 5 years ago 😀.

Niels De Graef Jun 23

@adilarif @neal he did join Red Hat recently jfyi :-)

Neal Gompa (ニール・ゴンパ)

@nielsdg @adilarif Yeah, I left SUSE out because I knew that he moved to Red Hat.

Adil Arif 6d ago

@neal @nielsdg Thanks for sharing guys. Red Hat got lucky hiring Joan. He is a rockstar.

Here is to hoping SUSE backfills that GNOME contributing position with someone else.

Demi Marie Obenour 4d ago

@neal @karolherbst @matt @Lyude Also System76 (COSMIC)

Josh Simmons Jun 22

@karolherbst @Lyude you have to support gnome as an application developer. it's not negotiable to stop supporting the majority of the linux desktop users. (this is one primary point of contention that continuously re-occurs, application developers *must* deal with the total madness version of conflicting everything. even just limiting to the major players of gnome/kde and maybe sway is a pretty big support matrix)

karolherbst 🐧 🦀Jun 22

@dotstdy @Lyude well yeah, but then you need to accept what they provide to you.

Like nobody can force them, and ranting about isn't helping either. And apparently shit-posting also doesn't. Big surprise.

If they are so important, then you kinda need to accept what they provide.

Like yes, in the ideal world every desktop would support the same set of APIs and everybody would be happy, but it's not the world we live in and I don't see how that's ever going to change.

Josh Simmons Jun 22

@karolherbst @Lyude i mean that's fine, but then you have to accept that people will be upset because wayland is breaking applications. i don't think it's reasonable to say only wayland developers get to rant (and rant they certainly do)

karolherbst 🐧 🦀Jun 22

@dotstdy @Lyude It's not helping regardless of who is doing the ranting.

And it's totally fine to be upset about it. I totally understand users being frustrated, because they only want things to work.

It's just we don't have billion of dollars we can just spend on making the Linux desktop awesome and polished and everything.

Josh Simmons Jun 22

@karolherbst @Lyude hmm, i would not say that money is the major blocker for any of the core pain-points for wayland. maybe accessibility, but otherwise it's mostly the impact of inflexibility around design decisions. (and the fundamental thing that it requires people to re-write their applications to varying degrees)

Josh Simmons Jun 22

@karolherbst @Lyude e.g. you could say that wayland would accept absolute pointer positioning protocols tomorrow and i'm pretty sure somebody would implement them, and one major sticking point would be gone. so i wouldn't say that is a funding or resources problem, except that if you want a perfect sphere that makes everyone happy, including people who fundamentally don't want to see the protocol, then you need infinite funding.

karolherbst 🐧 🦀Jun 22

@dotstdy @Lyude yeah, but desktops can just implement it.

Like they could do so tomorrow. They don't need any permission from "wayland upstream".

karolherbst 🐧 🦀Jun 22

@dotstdy @Lyude Maybe there are other good reasons it hasn't happened yet.

Josh Simmons Jun 22

@karolherbst @Lyude i didn't say there weren't and a lot of these things might just be fundamentally bad ideas. but you gotta engage with the idea that people disagree, and a lot of those people that disagree are people writing applications for the platform? to me at least that's the core problem here.

karolherbst 🐧 🦀Jun 22

@dotstdy @Lyude I think using the windows API for GUI applications is fine in principle. Like if app developers just want something that works, then they could just keep their windows apps and go for it.

Maybe the integration could be better, but I don't see a reason why it shouldn't work out as well as for games long-term.

However, I do understand that "native apps" are better and I think it's great there are app developers who care and it's a shame to demotivate them like this.

Josh Simmons Jun 22

@karolherbst @Lyude I'm not even talking about games, games have very low requirements from desktop environments. I mean stuff like KiCad or blender or firefox or etc

karolherbst 🐧 🦀Jun 22

@dotstdy @Lyude oh sure, but I don't see a reason why it shouldn't work for them in principle either.

Though I think people haven't really focused on it, so maybe there is a lot of polishing to do.

Matt Campbell Jun 22

@dotstdy @karolherbst @Lyude What requirements does Firefox (or any other browser) have that aren't met by all the popular compositors?

Josh Simmons Jun 22

@matt @karolherbst @Lyude https://bugzilla.mozilla.org/show_bug.cgi?id=1621261#c72 was one I was pointed to recently from a firefox person. (and see for example of a joyful engagement with the desktop community... https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/1141#note_2380514)

1621261 - (wayland-pip) [Wayland] PIP video is not kept on foreground when switching tabs or windows (Workaround:right-click on PiP window and enable always-on-top)

NEW (nobody) in Core - Widget: Gtk. Last updated 2025-06-11.

karolherbst 🐧 🦀Jun 22

@dotstdy @matt @Lyude Mhh maybe they don't know that Firefox uses it or something? Might be worth a discussion again... but yeah...

karolherbst 🐧 🦀Jun 22

@dotstdy @matt @Lyude Anyway, I think those discussions usually go better if one asks for use cases instead of "pls implement this feature"

Josh Simmons Jun 22

@karolherbst @matt @Lyude there are use-cases in that gnome thread (custom launchers / docks), it's just upstream doesn't think they're important. which well, it's not how i'd approach it, but i guess that's a way. (also shutting it down like this isn't a "we don't have resources to implement it" it's a "we will not implement this nor accept contributions that implement this, go away". which is a pretty different vibe)

karolherbst 🐧 🦀Jun 22

@dotstdy @matt @Lyude Right, but ultimately their choice.

I mean, if people would leave it at that, it would be fine, but then some people decide to start hate campaigns over it and that's a huge problem.

So I can kinda understand why Gnome developers react like this, because to be fair, they are treated unfairly from time to time and I fully understand if they are not in the mood for discussions because of that.

Josh Simmons Jun 22

@karolherbst @matt @Lyude I mean that's all true, but it's not a de-escalation move and it won't de-escalate the situation. /shrug they're free to do whatever of course, but again, it affects the general perception of wayland.

Josh Simmons Jun 22

@karolherbst @matt @Lyude yeah i mean probably it would be good if firefox was involved if they want to use something, but otoh that exact flow is basically how an external developer looks at the whole thing. they go "oh i need picture in picture how do i do that in wayland" -> "oh... nevermind then i guess it'll stay broken". not everyone has the pull of firefox :')

Paul Khuong Jun 22

@karolherbst @dotstdy @matt @Lyude Makes you wonder about that "99% of users" figure.

karolherbst 🐧 🦀Jun 22

@pkhuong @dotstdy @matt @Lyude well it's not used yet...

Josh Simmons Jun 22

@matt @karolherbst @Lyude btw since i had some links open here's some more.

standard restore windows issues: https://bugzilla.mozilla.org/show_bug.cgi?id=1795531 https://notes.nickdiego.dev/chromium/wayland-session-management#Supporting+xdg-session-management

issues with popup placement https://bugzilla.mozilla.org/show_bug.cgi?id=1941237

issues with focus stealing https://bugzilla.mozilla.org/show_bug.cgi?id=1896457

classic obscured window refresh rate issue (should be fixable with current protocols): https://bugzilla.mozilla.org/show_bug.cgi?id=1854920

drag-drop issues (should be fixable with current protocols) https://bugzilla.mozilla.org/show_bug.cgi?id=1711465

1795531 - [Wayland] Restore windows to their original virtual desktop/workspace

NEW (nobody) in Core - Widget: Gtk. Last updated 2025-06-03.

Josh Simmons Jun 22

@karolherbst @Lyude perhaps this is a perspective thing but, they're the same people? in order to implement a widely deployed protocol, you need buy-in from the community. which means you likely need to appease at least kde+gnome. so, true you could write a whole new platform and implement whatever you like, but that's also not really a tractable approach for anybody. if you want to split hairs about whose fault things are then it gets a lot more complex, but that's not really outsider-facing

karolherbst 🐧 🦀Jun 22

@dotstdy @Lyude Yeah.. and I understand that it sucks for everybody.

But I do think that wayland developers have a point in focusing on use cases, not technical solutions.

But maybe the FOSS Linux desktop is also the wrong place for such grand "experiments" and it's rather something Apple or Windows should have started instead 🙃

But it is what it is and I think what's missing is a bit more good faith and open minds from both sides of this story.

Josh Simmons Jun 22

@karolherbst @Lyude yeah i mean honestly that's all i really think is missing as well, 100% agreed there. as always, social problems are hard, not technical problems.

Demi Marie Obenour Jun 22

@karolherbst @dotstdy @Lyude Good faith and open minds from all sides are absolutely critical. I hope I have shown that. If I ever don't show them, I hope someone tells me, because that means I made a mistake and need to fix it.

karolherbst 🐧 🦀Jun 22

@dotstdy @Lyude We are literally "competing" with companies having hundreds of billion of dollars and if they have things that doesn't work, they hire 100 devs and fix it.

Sure, it's not that simple, but I think people also have to cut us some slack, because the entire game is already rigged against us from the start.

It's actually incredible that we have not just one, but multiple working desktops without that sort of funding.

Sebastian Wick Jun 22

@Lyude @karolherbst

You're implying that...

1. not wanting to add a feature to a standard and thus our project isn't equally valid as wanting to add a feature
2. GNOME isn't playing nice with others
3. You (for whatever meaning of you) want to force GNOME (which includes me) to do what you want

How about you stop trying to influence things you do not work on even in the slightest? It's already awful enough having to deal with Gompas and Klumpps who jump from project to project without involvement, influencing standard bodies and telling the people who actually do things what they actually should do. Now you're joining them. You also want to join their youtube linux influcencers friends who then rally up the masses to bully us in the bug trackers because we're not doing what you want?

I'm really fucking mad that CoCC people are behaving like this. You're supposed to protect us from this shit, not be the cause.

Sebastian Wick Jun 22

@Lyude @karolherbst Did you read https://tesk.page/2025/06/18/its-true-we-dont-care-about-accessibility-on-linux/? Do you see the parallels? We're putting and did put an insane amount of resources into wayland and implementing wayland protocols. And now you're coming along saying "we should exclude GNOME because they are not implementing all the protocols I, as a bystander, like".

It’s True, “We” Don’t Care About Accessibility on Linux

What do virtue-signalers and privileged people without disabilities sharing content about accessibility on Linux being trash have in common? They don’t actually really care about the group they’re defending; they just exploit these victims’ unfortunate situation to either fuel hate against groups and projects actually trying to make the world a better place. I never thought I’d be this upset to a point I’d be writing an article about something this sensitive with a clickbait-y title. It’s simultaneously demotivating, unproductive, and infuriating. I’m here writing this post fully knowing that I could have been working on accessibility in GNOME, but really, I’m so tired of having my mood ruined because of privileged people spending at most 5 minutes to write erroneous posts and then pretending to be oblivious when confronted while it takes us 5 months of unpaid work to get a quarter of recognition, let alone acknowledgment, without accounting for the time “wasted” addressing these accusations.

TheEvilSkeleton

karolherbst 🐧 🦀Jun 22

@swick @Lyude In case I did come around the wrong way. I highly appreciate the work Gnome developers put into all of that, and the accessibility work is highly appreciated.

Anyway, it's fine to have technical disagreements and different visions. And I'd wished that people would be less forcing the matter overall here.

And yes, the random gnome hater riling up people are a massive problem, no denying that.

Sebastian Wick Jun 22

@Lyude @karolherbst Oh yeah, and if you want to force us to implement wayland protocols, are you going to force wlroots to implement portals?

Like, it's always the same shit. If you want, you can find everyone drawing a hard line in the sand somewhere. Some people just get all the shit in the world for it, and others don't.

Lyude🌹#BLM Jun 22

@swick @karolherbst I'm going to dip out of this conversation now but I want something to be perfectly clear here:
i have actually spent pretty huge amounts of time talking people down and actually defending decisions that gnome people have made. i literally have lost count of how many times I've taken gnome's sides on discussions. because frankly most of those discussions I don't mention here.

Sebastian Wick Jun 22

@Lyude @karolherbst Not relevant. What you said here is absolutely not acceptable and has consequences on my well being.

Demi Marie Obenour Jun 22

@swick @Lyude @karolherbst It is completely valid for GNOME to not want to add a feature. It is equally valid for application authors to only support compositors that have that feature.

Are there specific protocols that GNOME does not implement that are needed by specific applications to work properly? If so, GNOME's developers have two equally valid options:

They can decide that this is a situation they are okay with, in which case certain applications will not work properly (or at all) under GNOME.

They can implement the missing protocols so that the applications work properly under GNOME.

Matthias Klumpp Jun 23

@swick @Lyude @karolherbst Hey! 🙂
I have been a GNOME foundation member for ages (there is a near-100% chance you are using code I wrote for GNOME). I currently contribute less to it compared to the 2010s because I am busy with maintaining Freedesktop and related projects. Rest assured, I care about GNOME as much as you do.
So, as fellow developers, let's please assume good faith, we all want GNOME to be great! 😃 - And cross-desktop collaboration helps all desktops in the long term 😉

Pekka Paalanen 6d ago

@Lyude How is the union of wp and xdg protocol extension namespaces not what you ask for?

hwentland 4d ago

@pq @Lyude @karolherbst enforcing things is always hard/impossible. Would a site that documents compositor support of protocols, akin to drmdb.emersion.fr, be helpful here?

karolherbst 🐧 🦀4d ago

@hwentland @pq @Lyude https://absurdlysuspicious.github.io/wayland-protocols-table/

Wayland protocols support table