Wayland is a piece of cancer

>No support for lightweight toolkits including FLTK, FOX Toolkit >No support for most desktop environments except GNOME/KDE bloat and several i3 clones >No support for cwm, Fluxbox, FVWM, IceWM, JWM, or Window Maker >No official support for BSDs GTK4 and GNOME suck. Wayland is harmful. Wayland, systemd, GNOME, and GTK are destroying Linux modularity and compatibility. Red Hat is destroying Linux! >you will use our trash desktop and our trash init system and you will be happy

Honestly I don't even get, why people are willing to use wayland in the first place. Is the redhat/gnome propaganda so impressive? I mean, xorg already has everything, what your hart may possibly desire.

Wayland's design encourages modularity, and modularity is inherently bad for compatibility. That's why it's terrible. >doesn't provide multiple APIs for various crucial desktop features (global keyboard shortcuts, system tray, screenshotting, screencasting, and others) >doesn't provide a universal basic window compositor thus each desktop environment has to reinvent the wheel >don't have a universal method of storing and configuring display, session, keyboard and mouse settings There's nothing like Xlib and XCB. If they were trying to kill modularity, this is not the way to do it. >>63337 >xorg already has everything Except a standard font rendering API, standard graphical toolkit or standard IME implementation.

>>63337 These "people" are the new generation soy-"tech" lovers brought in by YouTubers(TM) and new LYnuxx hype. Most of them are not even real programmers or at best some kind of soy front end webdevs. The soy-geeks think it makes them cool like a l33t blue haired hacker with a twitter account but they are still a windows / mac platform lover. Linux is becoming mainstream, and hence dumbed down and platformized.

>>63339 I thought you cared most about licenses and the freedumb it gives. I guess it was actually about feeling smart the entire time.

>>63340 There is this one thing, tht I really don't understand about people like you. Are you a masochistic individual? I mean, does it feel good to you to not be able to do what you want? Or why do you oppose freedom of software? >I don't have anything to hide because I don't do anything bad First of all, we all do something bad. Second, if you don't do anything bad at home, do you like people looking through your windows?

>no support for <some toolkit> Just use xwayland >no support for <some window manager> It may come as a shock to you, but Wayland isn't intended to be an ABI-comaptible drop-in replacement for X. In the Wayland model the DE/WM is what implements the Wayland spec. This is different from X. Is it worse? You decide. >no official support for BSDs There's also nothing explicitly tying Wayland to Linux, just nobody has implemented a Wayland compositor for a BSD. >Wayland is destroying modularity Its literally more modular than X. >systemdicks and Wayland are similar No

>>63342 >Just use xwayland Why should I ever use this chimaera, while I still can just use plain xorg? wayland has literally no benefit over x and xorgs lack of standards for some stuff didn't even ever bother me. >Denying the fact, that systemd and wayland are both soykaf strongly backed by redhat Never ever trust redhat

>>63341 I want to be free from inconvenience. Windows doesn't do that all the time, there's room for improvement, but it's a lot better. I gain no pleasure from choosing between 10 soykafty desktop environments that all have the same problems, and I have no desire to finagle another soykafty one. >why do you oppose freedom of software? I don't. Wayland is no different to me than all of the other soykaf. It's just terrible in a different way. The license is fine though. It's still "free". You seem to automatically equate popularity with a lack of freedom. >>63342 >This is different from X. Is it worse? You decide. Yes. It's the exact opposite direction to the one needed to make something actually good.

Xorg has no GUI isolation. Any window can access another window and log sensitive information such as the root password. Wayland's GUI isolation is essential to security. That said, you should only use desktop environments that support Wayland.

>>63351 The few desktop environments that run on wayland are filled to the brim with bloat, spyware and just soykaf in general, just like wl itself. It's safer to live with the actually very small risks that come with xorg

>>63352 >very small risks Not supporting GUI isolation renders all sandboxing attempts useless. Firejail and bubblewrap cannot do anything on Xorg.

>>63353 I'm not saying that GUI isolation is bad, or that xorg is flawless. I just don't like wayland for not supporting anything good and for propably being spyware. Especially because of the second point I see xorg as a lesser risk. All that energy that is flowing into wayland could also have been used to help x. The fact that the wayland developers don't understand this simple beauty of free software also speaks for the project to be redhat spyware

>>63343 >Why should I ever use this chimaera Because its piss easy >wayland has literally no benefit over x Smaller code, many implementations, more modular design. >>Denying the fact, that systemd and wayland are both soykaf strongly backed by redhat Yeah, I'm a deceitful Jew who is trying to hide the fact that Wayland is backed by red hat. That's really the best you can come up with? >>63345 >Yes. It's the exact opposite direction to the one needed to make something actually good. Why is that? >>63353 That's not really true, firejail has the option to run an entire embedded xorg server per application. This is, that's a massive waste of resources.

>>63354 >All that energy that is flowing into wayland could also have been used to help x The best way to help X at this point is to kill it. Its a huge bloated pile of poorly designed software filled with unmaintained code. Every time people tried to update it to make it usable they just added to the pile. You could convince me Wayland is bad, but you cannot convince me X is good.

>>63355 >Why is that? Because everything in a system should be designed by the same group of people, who are organized, have a common goal, and have a responsibility to each other. When two things in a system don't interact properly, that's considered a problem with the system as a whole. Somebody is responsible for fixing it, whereas in a decentralized development model, nobody is considered responsible. Software also looks consistent with each other, and can continue working far into the future, without needing any maintenance. Things like a file picker, are non-issues, because all software uses the exact same one.

>>63336 Exactly. Wayland is a project catered to gnome's needs and i suspect redhat trying to make other DEs and WMs obsolete.

>>63356 It's been 30 years if the x11 protocol could be fixed then someone would have fixed it already. >>63337 >i know all the negatives and none of the positives >it must be everyone else who has been propagandized lol k >>63351 This.

>>63371 >if the x11 protocol could be fixed then someone would have fixed it already There's this little thing called money, and without it, motivation to fix things tends to be sparse.

>>63336 can we stop with this bloat meme? kde can be as little as you want it to be

>>63373 >kde can be as little as you want it to be You can't tell the devs to stop spreading their effort thin by having an excessive number of ux customization, and instead focus on stability, polishing defaults, and better accommodating small screen sizes.

>>63354 You are spreading misinformation and baseless conspiracy theories about Wayland. There is no evidence that it is spyware. Unless you provide reliable sources to back your claims, nobody will believe you.

>>63373 A meme isn't always "just a meme". The concept of bloat is a very serious issue, both in terms of efficiency and in terms of security. Perhaps the phrase "attack vector" means nothing to you, but it certain does to me and many others. I don't care what software other people use, but I do care about people spreading nonsense and misinformation and the suggestion of removing a blatantly solid argument against your software of choice is nothing short of dishonest and soykafty. >>63374 At the start, one could suggest `kde-base` to hold the absolute minimum and include `kde-[feature-name]` as a sort of addon.

>>63351 >That said, you should only use desktop environments that support Wayland. less bloated X server + ultrabloated DE vs bloated xorg + ultra lightweight WM. idk man, at least with xorg we can actually choose from a huge repertoire of WMs or DEs >>63342 >In the Wayland model the DE/WM is what implements the Wayland spec what an interesting way of saying "let the rest do the workload for me". If wayland came from a less known company/ group, that ideology would've been unviable and pretty much killed the project before even showing something usable to the potential userbase. >>63343 >Never ever trust redhat I will ask for sauce on that

>>63379 >less bloated X server + ultrabloated DE vs bloated xorg + ultra lightweight WM False dichotomy, I use sway which is a really comfy i3 clone but there are dozens of others, mostly because of the wlroots project which makes it very easy to implement a lightweight Wayland compositor. >what an interesting way of saying "let the rest do the workload for me". If wayland came from a less known company/ group, that ideology would've been unviable and pretty much killed the project before even showing something usable to the potential userbase. The Wayland group also publishes a reference compositor, and there are multiple lightweight libraries that people can use to avoid duplicating work (I must again mention wlroots). Unlike X, its so easy to implement the Wayland protocol that dozens of people have implemented it, not just one group.

>>63351 Hey, here's something for your security, how about you just don't run shady programs? I thought Linux users were supposed to be smarter than the average Windows user, or at least smart enough not to do what Windows users do all the time and that's downloading random binaries off of literally-who websites and compromising their system. But apparently not, apparently we need mobile OS style "security" where we have to constantly find workarounds to get basic soykaf done because apparently we need protection from ourselves. The idea behind Wayland has got to be the most condescending soykaf I've ever heard.

>>63353 >Firejail and bubblewrap cannot do anything on Xorg. xpra, xephyr, vnc There are things you can do to isolate windows from each other without wayland. It would be nice to have a display server that was designed properly to begin with though. >>63408 >just don't run untrusted code openbsd called they want their security strategy back. >>63378 >everything new and scary is """bloat""" You're not really adding anything to the discussion.

>>everything new and scary is """bloat""" >You're not really adding anything to the discussion. And yet, I am not even implying that "new", nor "scary" is bloat. Nor am I even trying to argue about wayland vs xorg, I don't care what you use. What I am saying is that unused features are simply a space, memory, and security issue. An unused feature is bloat, period. This concern is a valid arguement against any program. For example, I find `javascript` a pointless and unsafe feature in my PDF reader. Thus, I disable it, removing the "bloat". Unfortunately, the anti-bloat-as-a-vocabulary-word types fall under two categories. The "define bloat" types, of which I have satisfied this question to allow for real discussion (which is their primary concern), and the "bloat is not a thing" types, of which I always find to be "filtered" by any form of abstraction or above simple-syntax discussions. You, unfortunately, are falling under the second camp, and I do not feel particularly bad, I just feel you have no capacity at which to discuss this. Feel free to "prove me wrong" but note that I did just say that I am simply arguing that not using a program due to "bloat" is a valid reason to not use the program and defining "bloat" for those who may fail to understand its usage.

There was some discussion on Wayland in the old Desktop Thread: https://archive.is/KnUOj#90% || https://web.archive.org/web/20220818041858/https://www.lainchan.org/%CE%A9/res/57666.html#58654 From the old thread: The pros of Wayland * No screen tearing * Applications are properly isolated from each other (X11 has this problem https://theinvisiblethings.blogspot.com/2011/04/linux-security-circus-on-gui-isolation.html) * Xorg is in life support mode. The cons of Wayland * Portability (FreeDesktop.org thinks that supporting only Linux is good enough) * Fragmentation (which happened because FreeDesktop.org was dumb. wlroots should have existed from the beginning.) * You can't run 100% Wayland-only system in practice * Sway is the only Wayland compositor (= a WM but Wayland) that just works. Labwc (a clone of OpenBox) might be a good alternative in the future, though. * Some things may not works as easily as in Xorg (screen shots, Firefox may (?) require that use add "MOZ_ENABLE_WAYLAND=1" environment variable, etc.). Notes * There is no Xorg server equivalent for Wayland, instead a Wayland compositor implements the whole thing (but wlroots exists now). * You can secure Xorg by not running it as root and by disabling both the TCP socket and the Unix socket (https://wiki.archlinux.org/title/Xorg#Using_xinitrc & https://wiki.archlinux.org/title/xinit#xserverrc) * As far as I know, you can use Xephyr or Xpra? (or VMs) to isolate X11 applications from each other. You can further increase the security by using Bubblewrap. * Xplain: https://magcius.github.io/xplain/article/ * https://wayland-book.com/ * https://invidious.tiekoetter.com/watch?v=8E0SOWo-Gsg (a video on Hikari Wayland compositor) * A madman might port OpenBSD's Xenocara & Xenodm to GNU/Linux?

>>63343 >wayland has literally no benefit over x Except that keylogging Wayland is not as simple as writing "xinput test 8 >>keys.log" to someone's .xprofile. What does Wayland not do? So far it's just the FPS limiter and lack of SSH forwarding as far as I can see. I've used it for almost a year with no issues.

>>63434 The key logging thing is true, but nearly every distro doesn't have GTK configured to use Wayland instead of Xwayland properly, meaning nearly every application on the distro will still be vulnerable. Not a problem with Wayland, but a lot of people don't recognize that using Xwayland gives you all the downsides of just using X.

>>63453 >Not a problem with Wayland So who should be blamed? And how should this problem be solved? The distro maintainers probably have a good excuse. >In a decentralized development model, nobody is considered responsible

>>63454 Why is blame necessary? Everyone knew adoption would be slow. >>63453 This is definitely not true on arch, Wayland is the default backend for gtk as downloaded from the system repositories.

So, as I understand it, on Wayland every program that needs X11 gets it's own Xwayland process (which is stripped down version of Xorg). Doesn't that mean that all X11 programs are isolated from each other if you use a Wayland compositor (like Labwc or Sway)?

>>63460 https://wayland.freedesktop.org/xserver.html You can sandbox applications with separate X servers with things like firejail and Xephyr.

This thread is pretty fuarrrking garbage. Loosely, wayland is a policy-based protocol rather than a mechanism based one. The rest follows from that. Something that follows is that it integrates well with the graphics subsystems (mesa, drm). 1. Software supports/uses wayland, not the other way round. This is pretty self-explanatory. It's a set of protocols built around the wire protocol that libwayland implements. There are lots of these for different things, e.g. xdg-shell, for putting windows on the screen, and protocols for IMEs. On xorg, your best bet for IMEs is to use toolkit-specific IME interfaces via e.g. ibus. 2. "just improve xorg" is a dumb proposition that comes from a place of not understanding either project. 3. It's not "wayland is secure", it's "X can't be secure". Any program running as your user can do all the things you can do. It doesn't really matter about "GUI isolation" unless you have some other sandboxing method in place. https://gitlab.freedesktop.org/wayland/wayland-protocols/-/merge_requests/68 will provide a generic way for compositors to integrate with sandboxing systems (including flatpak), which is infinitely better than having a seperate X root for every GUI app. 4. Gnome is a bad actor politically, as far as wayland goes. They've been against everything good, and write misleading, terrible articles like https://blogs.gnome.org/tbernard/2018/01/26/csd-initiative/ On the bright side, wlroots is going well, and useful protocols are being written and implemented (e.g. secure screen locking with ext-session-lock-v1) And as far as KDE this is a promising project imo https://subdiff.org/blog/2022/more-wlroots-with-kwinft-524/ It's implied in the article that they plan to implement wayland screensharing/recording protocols, which means that gnome will become the only compositor where you have to use pipewire/dbus to do that. Incidentally, gnome's own arroagance and refusal to implement and standardise interfaces is why people think "wayland is a gnome/systemd thing!", as you can see by the language they used in the above blog post ("oh, by 'wayland' we actually meant 'gnome' lol!"). No wonder people get the wrong idea. 5. Wayland can't do teary pageflips, wayland doesn't have any colour management workflow, wayland doesn't have x, y, z Colour management is hard, and not many people care about it, so work is slow. https://gitlab.freedesktop.org/pq/color-and-hdr is where it's being worked on. Valve's interest in wayland as far as gamescope means that screen tearing will be possible. The thing is with this, is that "wayland can't x" means "oh, I have to write a new protocol and implement it", whereas "xorg can't x" means "this is literally impossible no matter how hard we try" (because of the assumptions made in the protocol that existing clients rely on). Wayland makes very few assumptions and is a relatively thin layer that doesn't expose implementation details in the same way X does. This is why applications can't choose where they are on the screen -- that's an implementation detail. Even so, you could write a protocol to do this! Any perceived limitation of wayland is not technical, and if the _policy_ you think is worthwhile doesn't exist, you can make a case for it. (layer-shell is a shining example of this (which gnome didn't implement because they suck). ------------------ So practically, >>63416 has the right idea. I disagree with some minor technical points but as a user this basically describes what it's like. Watch the hikari video that they linked. If you want to look at wayland, sway is probably your best bet. No one is forcing you to stop using X. I'm happy to answer technical questions about internals etc, or about who's doing what and what's going on politically with the interactions between various projects like browsers and so on. I'd rather not deal with the type of arguments that are going on in this thread though, where two people who understand neither X11 nor wayland argue over which is "less bloat" and more "modular".

>>63810 >"wayland is a gnome/systemd thing!" Failing to mention both are Redhat things. >layer-shell is a shining example of this (which gnome didn't implement because they suck Demonstrating how easy Wayland makes fragmentation. >Even so, you could write a protocol to do this! So write a feature that won't be used by anyone because even if you "make a case for it", it won't be added to the most common implementation.

>>63811 >Failing to mention both are Redhat things. It isn't? Look at XDC sponsors, they are all invested in this. Might as well say "it's a google thing" because chromeos uses wayland. >Demonstrating how easy Wayland makes fragmentation. Gnome choosing not being compatible is not "wayland making [sic] fragmentation", it's gnome choosing not to be compatible -- which happens all around linux desktop stuff, just look at gsettings and so on. Gnome being a special snowflake is bigger than "wayland fragmentation". There are two "fragments": gnome, and sane people who care about things working. >So write a feature that won't be used by anyone because even if you "make a case for it", it won't be added to the most common implementation. This is incoherent. I assume even if it wasn't it would miss the point judging by the above.

If X and wayland are both flawed, why can't a new one be made from scratch addressing these issues? Excuse my ignorance, but the open source community has made some incredible stuff yet for some reason we are stuck with X for 3 decades, and wayland still has compromises.

>>63812 How many people use Gnome compared to your "sane" DE?

>>63816 How many people use windows? It doesn't mean anything. Gnome are bad actors and a net negative for the project. Mutter isn't a good compositor, bugs arise when using correctly written wayland programs that aren't part of the "gnome ecosystem" (e.g. foot). >>63814 It would require rewriting the part of every program that puts pixels into a window, and writing new programs that put windows onto the screen (compositors, and yes xorg is a compositor, the xcomposite extension doesn't define what the word means). So, it would be just as much work as wayland people have been doing since the project became a serious thing (~2014?). Given that wayland is extensible there's not much point in completely ripping everything up. Making new thing from scratch which addresses the issues is more work than just addressing the issues as they are. Which seems like it's being done, which I'm happy about. The things romangg, pekka paalanen, and simon ser are working on are good. And re: "wayland is a $company project", https://gitlab.freedesktop.org/wayland/wayland-protocols/-/blob/main/MEMBERS.md is a list of the people who make the decisions. Make your own judgements on whether they are in the pocket of "big hat" or not.

>>63820 >It doesn't mean anything. So you might as well use Plan9 and write programs to work with rio, since you think adoption doesn't matter.

>>63823 Not him but that would be idealistic if everything was interoperable >>63820 Multi monitor support seems to be an issue when you throw Xwayland in the mix which feels more like a bandaid than a proper fix. Xwayland seems to just undermine waylands existence in the first place, no?

>>63824 >Multi monitor support seems to be an issue when you throw Xwayland in the mix In what sense? I have a multi-monitor setup at work and everything Just Werks.

>>63823 This also doesn't mean anything. It's completely orthogonal. What point are you even trying to make? Why didn't you respond to any of the other points I made? You're just randomly responding to a few words of previous posts like it's the final word in a discussion, as if you want to make a *drops mic* killer argument and impress somebody. Except, everyone can still see the whole of the previous discussion. This board isn't fast enough for spamming one liners. What are you trying to accomplish overall? Deliberately misinterpret people online so that you can have petty arguments about nothing without having to actually know anything? >>63824 >Xwayland seems to just undermine waylands existence in the first place, no? Not really. Compatibility layers are important, otherwise there's no step between "using A" and "using B". Something like "oh, I want to use B, but I still need A for this one thing..." is pretty common, with e.g. linux and wine, wayland and xwayland, etc. I mean if you're running *everything* through xwayland then I'd question why you're using it. But this isn't usually the case, most "normal" GUI apps have wayland backends now. I can think of some that don't, e.g. blender, wine.

>>63833 >This also doesn't mean anything If the most common implementation of Wayland doesn't support a feature, developers won't use that feature, even if less popular implementations do support it. How hard is it to understand that? Why do I have to spell it out? Anything I don't respond to, I don't have anything to say about. I'm not going to do research to prove Redhat has a disproportionately large influence on Wayland.

>>63833 >But this isn't usually the case, most "normal" GUI apps have wayland backends now. Indeed but some of them, especially proprietary ones, like to bundle their own GUI toolkits which either have buggy or nonexistent Wayland support.

>>63834 >If the most common implementation of Wayland doesn't support a feature, developers won't use that feature. No, it doesn't work like that. For an app like mpv, the idle-inhibit protocol means that it can stop the DE going to sleep when a video is playing. This doesn't work on Gnome. So on gnome, everything stays the same, as if they hadn't written any code, but on _every other compositor_ it's an improvement. There's no loss by implementing the feature, it's just a "bonus" for the compositors that do implement it. It doesn't stop the app working. So that's one type of program. The other type is something like e.g. a taskbar. The people who use or develop for Gnome have no overlap with the people who use or want to develop indepedent taskbars. This is why people write layer-shell programs, even though gnome doesn't implement layer-shell. That is to say, programs that wouldn't be useful/wanted on gnome in the first place. The same goes for screen lockers. There's a screen locking protocol (ext-session-lock) that fails safely (i.e. crashing the screensaves isn't a security problem, re: jwz "I told you so"). Can you use it on gnome? No, but no one cares, because the people developing standalone screensavers aren't Gnome users or gnome devs.

Libliftoff is a cool project https://gitlab.freedesktop.org/emersion/libliftoff You can see a talk about KMS planes by the author here https://video.fosdem.org/2020/K.4.401/kms_planes.webm It will be cool to have direct scanout on more than just fullscreen windows in wlroots compositors. Gamescope https://github.com/Plagman/gamescope (comes with valve's new handheld) is using it.

>>63835 How would you even 'bully' proprietary and even some open source apps to support wayland? It seems like a chicken and egg situation.

>>63810 How the fuarrrk am I supposed to move windows around without a title bar? fuarrrking Chrome and all it's tabs. What autistic soykaf is this?

>>63846 Press alt and you can drag a window with your mouse anywhere you click on it. Or use a tiling window manager.

>>No official support for BSDs And do you use a BSD as main system?

>>63846 You want to move your windows without having to dodge the 30 different buttons we put there? What about the GNOME LOOK? Don't you care about the GNOME LOOK?

>>63851 BSD is irrelevant. OpenBSD and FreeBSD are dead.

The whole BSD thing isn't even true. Multiple compositors work on FreeBSD (at the very least sway and hikari), and I have no idea what "official" support means.

>>63858 They have only been growing? Although modest growth >>63860 There isn't such thing as 'official' support in anything FOSS unless you're talking about a specific distro I suppose

>>63858 Both FreeBSD and OpenBSD are great operating systems, they might be forgotten, but as in operating system design, and sanity, BSD is better than linux.

>>63858 >OpenBSD is dead I can refute this, but I'd rather hear your perspective. Perhaps you mean in a rhetorical sense. Would you kindly elaborate?

>>63853 Honestly I kinda like that, I always use keybinds to move, resize, etc windows on Xfce. My problems with gnome are everything else, like that useless top bar and the constant breakage. If you are going to "take inspiration" from macOS at least do it properly. Ok you can't for some reason use the top bar for the menus, then do away with it completely. Put the clock, notifications and calendar on the dock and give me my pixels back.

>>63336 >preventing further linux development because new = bad

>>63861 There is if the reference compositor uses Linux specific APIs. Same way that systemd is tied to Linux and glibc.

>>63846 Sway and KDE both support server-side title bars (via zxdg_decoration_manager_v1), it's just foot that doesn't support it. I assume Wayland apps are supposed to implement their own topbar, and that's what foot's CSD is supposed to be about -- but any other functioning compositor should give the protocol to clients that just want a title bar provided to them. >>63810 Got a scanner question: The enum wl_shell_surface::resize is marked as a bitfield, though its member fields have values 1, 2, 4, 5, 6, 8, 9, 10. If bitfield enums can have members with more than one bit set, how am I supposed to interpet what "bitfield" means when generating stubs? AFAIK this is the only enum from a well-known protocol that does this. >>64055 If you can open a BSD socket, and you can make shm objects (POSIX does this), and you can send file descriptors (POSIX does this too), you can speak Wayland. How a BSD compositor supports each protocol is up to them, but having the reference compositor do something Linux-specific behind the scenes shouldn't stop BSD from the same thing.

>>64181 <enum name="resize_edge"> <description summary="edge values for resizing"> These values are used to indicate which edge of a surface is being dragged in a resize operation. </description> <entry name="none" value="0"/> <entry name="top" value="1"/> <entry name="bottom" value="2"/> <entry name="left" value="4"/> <entry name="top_left" value="5"/> <entry name="bottom_left" value="6"/> <entry name="right" value="8"/> <entry name="top_right" value="9"/> <entry name="bottom_right" value="10"/> </enum> If you mean this, then it's not really seperate members, it's just showing how the data is encoded in the enum. Maybe it would be a little more correct to describe it differently, but the word "bitfield" means "bits have specific meanings" I guess, and it still does here, since top left is just top + left and so on. The word "bitfield" is just metadata and clarifies intent behind the actual numbers in the enum, I guess. I don't think it matters if you're writing a scanner implementation. The only thing I can see in wayland-scanner that actually does something based on whether it's marked as a bitfield is that it has to be uint.

Dear lord you guys are still going, if you don’t like it just don’t use it. I don’t like Gnome and Chrome. I don’t use Gnome and Chrome. It’s that simple.

>>64215 yes they aren't using it, why do people always say this stupid thing? people talk about how bad things are so others can know

>>64428 >people talk about how bad things are so others can know Nah they are just being autistic or arguing for the sake of arguing. If you think some software is bad for reason X, you simply state X and move on with your life. People here don’t even have a X to state, they decided wayland is bad and now are trying to find out why.

I'm currently trying out Wayland by writing an offline protocol browser directly on-top of Wayland (I'm using cairo and pango for rendering). Sidebar shows the index and you can select the protocol you're viewing by toggling a drop-down menu from the protocol name. I just have basic scrolling beyond what you see here. No text selection or hyperlinks to other sections.

Just don't use a graphical interface, I myself just use the TTY, using cmus if I want to listen to music, elinks for web browsing, and mplayer writing to framebuffer for video.

>>63336 Red Hat soykaf is used in a lot of distros, that doesn't mean you can't use alternatives. The only Red Hat thing that seems truly difficult to escape from is udev, since the xorg input drivers seem to depend on it as far as I can tell. You can use xorg instead of Wayland, OpenRC Runit etc. instead of Systemd, literally any other DE or WM instead of GNOME and qt (or preferably applications that require nether gtk or qt) instead of GTK.

>>65143 it has x11/wayland compatibility layers, though i don't know how well they work >>65145 almost certainly one

>>63338 just pick one of the available ones... oh wait, nvm, theyre all soykaf. Just install winblows to be honest

Btw does anyone know why pinch zooming don't work on x11?

>>63416 Update: I have ran Sway (a Wayland compositor) for about 3 months now. It just works for the most part and Sway is the best/most complete/most stable Wayland compositor. But... FireFox is not that stable under Wayland. Arranging bookmarks on the bookmark bar is hard (it doesn't work nearly as well as it does on X11 or even Botnet 10). But what made me strongly consider stopping using Wayland is the fact that arranging browser tabs made FireFox crash! It has never happened on X11 (I have used both Xorg and Xenocara). LabWC is even worse. It might crash when you view bookmarks on bookmarks bar and you use the overflow menu (where the extra bookmarks that do not fit the bar go). If you use Wayland, I highly recommend you write your soykafposts using your text editor of choice. I might go back to StumpWM (I prefer the way it works but I wanted to use Wayland instead of X11).

>>67520 Ok, I was wrong. My childish argument point still stands.

>>67517 The software ecosystem is still pretty unstable and there are giant gaps in needed companion protocols/programs. You cannot just run any X program in XWayland or SDL program in Wayland and assume it will work. There are big crashes and regressions you are ignoring or are unaware of. On top of that, Wayland is implicitly synchronized whereas the modern approach in Windows/MacOS/Vulkan/DX12 is now explicit synchronized. They have just taken too long with the project and now the ecosystem is no longer on the cutting edge, but unstable at the same time. It should have been reworked ages ago but now this will be even harder since they have some fledgling ecosystem to support. On top of all that, the attempt at security isn't really used by anyone except containerized OSes or stuff like flatpaks, the latter only applying to the flatpak programs and not the rest of the system. You don't really get isolation otherwise.

>>63454 the issue was already fixed a decade ago by the openbsd team and xenocara

>>68010 wlroots has the following dependencies: meson wayland wayland-protocols EGL and GLESv2 (optional, for the GLES2 renderer) Vulkan loader, headers and glslang (optional, for the Vulkan renderer) libdrm GBM (optional, for the GBM allocator) libinput (optional, for the libinput backend) xkbcommon udev (optional, for the session) pixman libseat (optional, for the session) hwdata (optional, for the DRM backend) libdisplay-info (optional, for the DRM backend) libliftoff (optional, for the DRM backend) None of these are particularly huge. And yes, you can use wlroots with eudev, I currently run that and it works great. Wayland itself doesn't seem to have any explicit dependencies, it is implied that they expect everything to be done with system libraries.

Tried using sway on a raspberry pi 4b running the linux-rpi kernel (never tried a non-rpi arm64 kernel build). Performance was dogshit. Tried every improvement anyone anywhere ever recommended, including all the ones that looked like they'd never work. They didn't work. Tried X in Enlightenment-DE on the same hardware. Performance was acceptable despite Enlightenment being built primarily for wayland. When you talk about user experience I need to hear your specs before I can make anything of it. If I need to restrict myself to a specific set of cards and compositors to have an invisible and frictionless experience then describe the machine it just werks on or I won't believe you, will begin to assume bad faith on your part, and in assuming bad faith will respond in worse faith and defame you as an IBM wage slave trying to promote an embrace-extend-extinguish replacement of programmers by watsonx chatbots. >>63371 >It's been 30 years if the x11 protocol could be fixed then someone would have fixed it already. 2008 was 15 years ago. If the wayland protocol could be implemented adequately then the world would look very different. For one thing the wayland-flagship, wayland-savior Steam deck wouldn't run a "heavily customized" completely nonstandard version of the protocol built to turn team wayland's sacred cows into hamburger. >65441 >Your comparing a protocol that's basically in its beta to a software that's been around for 30 years. He's comparing a protocol that has been consigned to beta by user stories and reputation based on poor experiences despite 15 years of expert development and public releases, to a software stack that was considered unworkably obsolete when it was 10 years old and has received comparatively little hard labor since then. Being outpaced by a 30 year old workhorse who was scheduled for the glue factory back in 2004 means your adult and wearily aging clydesdale does not have the blood of champions and shouldn't be given the first position in the stable even if X11 wheezes and is even longer in the tooth.

>>72647 Things taking time to come together does not correlate to poor quality. For example your mom spent 9 months germinating your stupid ass. You sound like those customers who call tech support whenever a new popup shows up that they didn't bother reading and then complain that the new system is too complex.

>>72662 these are criticisms that don't match reality as someone who is literally on wayland right now and runs everything on it from AAA games released this week to Gorky 17. I also work on this system as a senior software engineer. I have my i3wm/xorg configs saved in case i need to go back, i have not felt the need to go back because the problems i had 5 years ago have been adressed. What makes you sound like someone who calls tech support whenever a popup shows up is the fact that you don't bother with understanding reality, you want reality to match your expectation without disturbing you in any way. A project with active development has active development. Things are not set in stone, problems that existed in 2017 are not problems that exist in 2023. Some problems are not universal, and require effort from you to seek out the necessary flags or configurations so that your system works as you expect it to. Wayland is not xorg, linux is not windows, water is fuarrrking water. And don't get me wrong, criticism is not what makes me upset with your post. Stupid criticism based on nothing but a literal skill issue is.

>>72677 >has already been fixed "This technically has a workaround" is not a counter argument to, "this poor design decision has repercussions". I'm not going to bother finding problems with Wayland nobody else has ever mentioned. I don't use it because I haven't been sold on it. >What do you want They should have copied how Microsoft or Apple does things, instead of doing their own snowflake, "minimalist" bullshit.

>>72680 >"This technically has a workaround" What are you even talking about? Have you even clicked on the very simple queries indicating that these changes are implementations on the protocol itself and not workarounds? A fundamental change on the tool you use to allow for a feature that was not possible before is the definition of a fix. Are you that incapable of reading that you think i sent you a link to a blog post talking about how to fix wayland renders with a different tool or something, you stupid cunt? Unless you think using a flag in a piece of software is a "workaround", in which case holy fuarrrk do i have news to you about X or any linux tool whatsoever. > "this poor design decision has repercussions". What poor designs, the ones already addressed, fixed on the main branch and left to the user to choose how their system behaves? >I'm not going to bother finding problems with Wayland nobody else has ever mentioned. I don't use it because I haven't been sold on it. That's fine, you don't have to, but if you are criticizing, you should criticize reality and not the monsters in your thick head you dumb cunt. >They should have copied how Microsoft or Apple does things, instead of doing their own snowflake, "minimalist" bullshit. at this point you're just an asylum patient without a nurse, go ask for coins on the street or something.

>>72698 >u have no friends, unlike me, a giga life ruler oh man you guys are really scraping the bottom of the barrel eh holy fuarrrk my asshole

>>72749 >>72780 an argument >is not an argument I need a feature >you don't need the feature ah right, thanks for figuring this out guys. Just like any other Linux-related discussion. Now that you owned the anon epic style you can keep on going with your miserable existence by raging and being offended that people point out most obvious and basic flaws in a piece of software. Someone gotta defend the controversial thing, and be firm on their opinions about this thing (not arguments because we don't have any - we will green quote and distort words instead).

>>72782 >color management ye, it isn't possible today as far as i'm concerned, if that's a dealbreaker to you, you should not use wayland. But alongside HDR it's a priority for KDE and probably valve: https://zamundaaa.github.io/wayland/2023/05/18/hdr-and-color-management-in-kwin.html >push to talk while you have to be focused on the window in order to activate or deactivate push to talk, most software can receive calls to activate and deactivate the mic, this would require a small keybind implementation on your part such as: https://github.com/mumble-voip/mumble/pull/3675 any other issues your favorite blog poster made this week you want to introduce us to? Because, although you are a defensive little bitch at the very least you are bringing real issues that exist as opposed to nonissues from youtube raised earlier. Issues raised in the most effortless and superficial manner that display the tech savvy of a grandmother with alzheimers? Yes. But at least real issues. And besides, we can't expect much from people like you, now can we?

>>72782 if colour management was an essential feature then the windows laptops at [computer store] wouldn't have the saturation blown out to showcase the most V I V I D 4k parrot photos and orange stock photo people on their display units. more generally I'm sick of this dumbass single issue concern trolling. it's old. nobody cares. bugs will get fixed or they won't, but this piss and moan dialectic is useless. If you care so badly find a dev working on it and send them a nice message and top up their ko.fi/liberapay/whatever.

>>72784 I see that you are very much offended, sorry for wayland being that big a part of your personality that defines who you are. After all, this is not just a piece of software - this is something bigger than yourself. If wayland was a human I'm sure it would suck you off if it could see just how much you love it. >>72787 >If you care so badly find a dev working on it You know very well this does not work like that. Wayland has actual proposals and resolutions of the long haunting issues, with the only reason of them not being accepted being that developers are very firm on their opinions how the protocol should work. In fact, they decided that it's not them who works on most of the parts of the protocol, it's the job of the compositor or even the end user, because, obviously, time of RedHat employees is more valuable than yours. Wayland is surrounded by drama, it's not being controversial just because. So it's not Wayland developers who resolve most of the Wayland issues, it's the compositor developers, and each does it differently. So then it happens that the developers decide to restrict my freedom over some exaggerated security concerns, like keylogging. Which is a valid concern, much like concerns raised by the governments around the world when they decide that you lost your privilege to walk and lock you down instead of giving a proper solution. Fortunately, things like fractional scaling and HDR get resolved and color management is incoming. Last time I checked there was a draft. Yay!!! Just took them 15 years to figure out that users actually need them. And the other issues like hardware cursors (https://github.com/swaywm/sway/issues/4763 https://bugs.kde.org/show_bug.cgi?id=452117) will probably take forever to resolve. Just how much time is 15 years to implement something that already existed or could exist in X11? Nothing for people who don't value their time at all. If only they weren't forcefully pushing this piece of incomplete crap down my throat, maybe I could ignore its existence.

>>72794 >I see that you are very much offended Yes! You finally got something correct. I am VERY upset because you see, i really, really like lainchan at it's best, highly intelligent technical folks getting together to discuss something they actively care about, usually technology, sometimes drugs. But what i get out of lainchan that i don't get in other imageboards is people who know what they are talking about. And you don't. You are not curious, you are not intelligent, you contribute nothing but you want to appear Correct. You conflate technical reasons with moral reasons, you want to say "i don't like this because it doesn't do what i want it to", and sure, you are free to open a thread and say that, that would at the very least lead to some honest discussion about something, as opposed to this superficial discussion about nothing which is what this thread is. This discussion could be about Windows 11, your lack of curiosity but at the same time desire to be correct and sound authoritative would very much upset me to the core. And yes, while you are within your right to be a stupid motherfucker, i also am in my right to point out your stupidity, imagine you're a gorilla in a park, i'm happy to put a warning sign around you so that people with less knowledge don't mistake you for something you are not.

>>72794 Why do you bother talking to brainwashed redhat bots? If somebody is too stupid to see that redhat, much like every corporation that ever got into linux, is evil, it's his problem and his delusion. You're throwing peas at a full metal wall.

I swear to god, why are you motherfuckers still here? This whole thread amounts to >Wayland can't do this and this and that at the moment, so if you don't want to switch to Wayland because of that, that's understandable. >That's the work of satanic RedHat! Like, nobody is ever going to reach a consensus here. No one takes anyone seriously here at this point, so why bother? OP is also crappy, retarded and a conspiratoid mentioning RedHat like it matters in any way. Everyone here seems to agree that Xorg is full of cruft, bloated, badly written, unnecessarily complicated. Some people wanna do something different, and escape X's grip on the Linux desktop. Some people don't care, and will just keep using X, or feel the need to call you a RedHat wage slave. There's nothing more to it than that. Go do something more productive with your life, for fuarrrk sake. This whole """debate""" is so insignificant in the grand scheme of things, you guys have no idea. I use Wayland myself, and Wayland users in this thread have brought good points, even though I don't care about any of this, because I can simply close this tab, ignore this stupid thread, and just keep on using Wayland. The reason why I'm posting this though is because you motherfuckers deserve better, and I hate to see people wasting their time with stupidity.

>>72795 >>72798 Same fаgging? Really?

>>72796 I like red hat. Dave Airlie and Karol Herbst are two of the most talented software engineers on the planet and the work they've put into vulkan and opencl respectively and open source hardware enablement graphics and ai generally is interesting in the way Minimal Custom Dekstop original tiling WM rice donut steel reddit soykaf will never be. >>72794 so many errors, for starters wayland devs publish a reference compositor called Weston that implements the protocol, second the process is managed through freedesktop, not redhat, and the process of ratifying these rfcs into standards takes so long because they aren't just jamming corporate decisions down everyone else's throat. you could even get into this process, the easiest way would be to start submitting patches to wlroots the project representing window manager stakeholders, but that might cut into the free time you have to post unhinged rants.

>>72795 I read that post imagining a gape mouth sоyjak.

>>72817 Not the same person. I want Truth not peace, i don't give a soykaf about being productive. I want you to have real arguments as opposed to whatever the fuarrrk this >>72823 is. As i repeated many times, there IS a lot to criticize but you spend your time acting like a high school clique, you want to feel authoritative and Right as opposed to spending the time to do meaningful research on the subject and give real critiques. You can not care about it, but you should be able to tell that critiquing something you don't care about to understand is just gossip. Something i would not expect of lainchan. I will keep hammering your stupidity as long as you continue being a stupid cunt pretending to know what the fuarrrk they are talking about. And "you" applies to everyone with this behavior, not you you. I know the concept of anonymous imageboards is a bit hard to grasp to your twitter discussion social clique brain, such a thing as criticizing the contents of the argument as opposed to the physical person behind the keyboard must be wild levels of complexity for such a dog soykaf stupid cunt like you.

>>72826 You sure do think a lot about yourself.

>>72827 Maybe. I can't see how that is relevant to the subject at hand. Which is the entire point.

>>72822 >I like red hat. Ahh, I was in the process of writing a wall, but then I suddenly felt futile, so fuarrrk it.

probonod, is that you? some more text to make the post long enough

>>63336 > blaming red hat for other people fleeing the dumpster fire that is X11 > how could you all leave behind my precious niche toolkits???? > what do you mean you dont like tearing and assloads of architecturally superfluous and buggy code????? none of this is getting forced on you. you are just mad about soykaf changing. dont let that stop you from yelling at clouds though.

Here's something I don't understand: >wayland is supposed to prevent keylogging and screen grabbing >this means screenshot software doesn't work >so they make a protocol for it >i install a program called grim and take a screenshot So now what prevents any malicious program from grabbing the screen using the same protocol or calling grim and accessing ~/Pictures/? I've never seen any kind of access control for the protocols. Does any compositor care about this? Does anyone care about restricting personal folders to only trusted programs? I feel like anything short of QubesOS tier isolation is a security theater unfortunately. Shit feels too leaky. Another thing I don't understand is how did they think hardcoding forced vsync (every frame is perfect™) was a genius idea? This gives me the impression the designers were bunch of noobs who never played an FPS game in their life. Hell, even in GUI, 60 FPS with vsync on vs off has noticeable smoothness and latency difference. Apparently they're now adding disabling vsync after all these years. Better late then never I guess. I wonder if they would have even allowed such a thing without a very based elephant called Valve being in the room now.

>>74734 we need to write our own thing finally that would have an elaborate security architecture and integrate with another, system wide security architecture that would be deeply rooted into the system and will tell apart trusted and untrusted parts of the system without compromising the overall simplicity of UNIX too much. but we'll never do this because bathing in soykaf is the normal state of humanity. things will never be good.

(I'm lanon from >>72798) >>72817 LMAO not at all, my guy. But I can see why you thought that. >>74734 The whole security argument from wayland is relevant, but then you have all of these kinds of complications because of that same mentality. I say fuarrrk all of this soykaf, and just embrace the fact that there's no such thing as perfect security. >>74736 >things will never be good. "If not you, then who? If not now, then when?" Take matters into your own hands, lanon, and things will be good in no time.

>>63336 I likely won't ever use Wayland because every single time someone mentions having issues with it it turns into victim blaming, gaslighting or literally being told that "you don't need X feature", which is something I find very aggravating when random internet people pretend to know what exactly I need. I don't really give a damn how pixels appear on my screen, be it through X11, Wayland or something esoteric that uses OpenGL or whatever. I care whether it works and X11 has worked well enough for many years. I am also of the belief that if something works it is best to just leave it alone and at most just strip out whatever functionality is practically useless in $current_year.

>>74745 i have claws instead of hands i can't do intelligible work i will never write software even bad. i'm not the kind to change the world. we need a leader. like Stallman once has been

i have a feeling linux will replace windows at the price of becoming windows and the og linux users will remain a niche using "obscure" distros to avoid all the windofication changes. So just brandname stolen

>GTK4 and GNOME suck. Wayland is harmful. Wayland, systemd, GNOME, and GTK are destroying Linux modularity and compatibility. Red Hat is destroying Linux! true, true, but the terminal and x11 both are destroying linux as well

>>74750 No, you don't need a leader. A leader telling you what to do won't change you for the better, they will just make you think more and force you to get busy, which is never a good thing. Do stuff because you want to, not because Big Boy Stallman wants, or worse, "needs" you to. The former is much more empowering than the latter. >i'm not the kind to change the world This isn't a matter of changing the world, this is just about a single software project that fits your needs. You might not be able to write a whole different display server, but I'm sure that that mentality of "i have claws instead of hands" affects other aspects of your life. I mean, I'm getting into /hum/ territory now, so I'll stop here. I don't even know what your personal situation is, so I can't give correct advice (assuming you even wanted advice. I apologize if you didn't want that).

>>No official support for BSDs FreeBSD has had ton of up-to-date Wayland stuff in its repositories for quite a long time now, and it is quite a bit easier to setup than a Linux distribution like Void. OpenBSD has a few Wayland packages (compositors, terminals, ..) in its ports system and it is easily compileable and working. What is "official support"? What are the BSDs you are referencing? >>63342 > just nobody has implemented a Wayland compositor for a BSD. This is probably the worst thread I have seen on Ω (although I'm not on here that often), why even check if what you're writing is true. Also, "implemented" is a terrible word for that and makes it sound much more complicated than it is. https://cvsweb.openbsd.org/ports/wayland/

They should have called it X12

>>74889 i wish the good successor will be called O server

Ah yes, the great "new" 15 year old thing that's entire design philosophy is "not my problem, go talk to the compositor developer instead". Fractional scaling works by super-sampling by requirement, because end-user resources are there to be wasted. Took so long to add a standard global keyboard shortcut protocol that we now have fragmentation in terms of how to handle user input, which is impressive at the very least. In general, "targeting Wayland" ends up being more akin to targeting every single individual compositor, since they all do things differently in order to monkey-patch the missing features of Wayland and wlroots into their software. Oh, and not to mention Waylands extremely poor OpenGL and Vulkan support, only supporting a small subset of modes and configurations compared to X11s Mesa. Doesn't surprise me to see how unconstructive and aggressive this discussion has been. People decree "works on my machine" and simply assault the other side, ignoring the missing features or fragmentation that makes targeting Wayland an asylum worthy task. Ah you see, color management being missing isn't a valid criticism because who uses it anyway! Say what you will about Windows, but the WinAPI providing a centralized way to construct GUI applications is one of the reasons it succeeded like it did, and it was and is something X11 severely lacked. The answer isn't to further fragment and complicate the creation of GUI apps.

>>74904 linux will never be good but i wish it stayed bad enough to keep retards at bay. yes we would have to put up with X for another century (still may have to, lol) but at least we would do so at peace with ourselves. look itt. almost no real linuxoids. most of these idiots are reddit desktop thread dummies. disgusting. also X is soykafty as well, don't just forget it. remember that citation? "i have never seen anything fill up vacuum so fast and still suck". but yet it has much more brain to it than wayland.

>>74908 >look itt. almost no real linuxoids. most of these idiots are reddit desktop thread dummies. disgusting. No kidding. The lack of any real information on either side of this thread has been absurd to see. It's a quagmire in here lol >also X is soykafty as well, don't just forget it. remember that citation? "i have never seen anything fill up vacuum so fast and still suck". but yet it has much more brain to it than wayland. I think that's the part of it I hate most. X11 is not good, it's a complete mess, but its very well featured and reasonably robust. If it had a unified GUI widget system in it, it would at least actually compete somewhat with WinAPI, which is also a complete mess. But also GUI programming in general is a complete mess, and it's why we've seen a massive shift over to declarative GUI libraries (JavaFX, WPF, UWP, hell the Electron and NodeJS hellscape even). GUI applications are tightly coupled to themselves and have to do a lot of things. They are the antithesis of the ill defined "unix philosophy", and yet Wayland is convinced this modular approach, where they standardise basic features 15 years after inception, is acceptable or robust in this landscape. I don't expect Wayland to ever really compete effectively with any other windowing solution, not even X11. It's gotten itself stuck in the politics of a soykafty and malformed design philosophy and as a result it relies on the work of everybody else (SwayWM, GNOME, KDE, Qt, etc) to take up the slack. It's extremely disappointing.

>>74751 Thats literally what already happened, ubuntu is nothing like artix or other non-systemd distros. The flatpak and other soykaf make it a weird windows-y mess.

what the fuarrrk is a wayland? tooshorttooshorttooshort

>>74911 >the hell is a gayland? From what I've understood it's a replacement for xorg that a bunch of trannies are ideologically adamant about. My guess is it's part of Silicon Valley's covert embrace, extend and extinguish strategy to soykaf up foss and open source projects, due to its odd design decisions that make it "work" if you basically just work around the intended API. And granted, xorg is three decades old and full of tech debt and way overdue to be superseded. However it doesn't look like wayland is it, which introduces an architecture where applications no longer are in control of their redraw loops (unless you force redraws to just be spammed) and it's such a breaking change that it's now splitting compatibility in the linux space/community. >b..but xorg is keylogger I haven't anyone seen bother about leaky clipboards either and that's potentially a bigger issue since anyone with a higher threat level (sys admin etc.) is going to use a password manager. >b..but we don't need to support your use case My use case is things working. Not having to deal with yet another swath of regression bugs and glitches on things that worked perfectly before after each update. >b..but wayland is still in alpha As it has been for 15 years. And I just don't have the patience to have your experiments affect my work flow and wait for another 15 years until it matures until the next shiny thing comes along. Sometimes you just have to cut your losses and learn from your mistakes instead of doubling down and make things worse for everybody.

>>74912 >a bunch of trannies are ideologically adamant about what? >and it's such a breaking change that it's now splitting compatibility in the linux space/community. eh???

unix was never good gui was a mistake ncurses were cursed to fail we should have stuck to command line modern tech is brain damaging and evil lets all love LAIN

>>74911 Wayland basics Wayland is a protocol that is similar to X11. But there is no Wayland equivalent of Xorg (X server) but there is a reference implementation of a Wayland compositor called "Weston" (NOTE: Weston is not suitable for regular use!). There is also wlroots library that can be used to build a Wayland compositor. A Wayland compositor is a WM but for Wayland (but it needs to do more work than a X11 WM because there isn't equivalent Xorg Server). There is also Xwayland that can be used to run X11 programs under Wayland (it's basically a stripped-down version of Xorg). Wayland is intended to be successor of Xorg. To try Wayland on Arch Linux, you need to install the following packages: sudo pacman -S sway swaybg xorg-xwayland bemenu foot #you can also use alacritty instead of foot After this, you can launch Sway by typing sway in Linux TTY vconsole (read the Arch Wiki and Sway docs for more info). You need Slurp and Grim to be able to take screenshots. Common Wayland compositors include: * Sway - i3(-gaps) clone that just works * Wayfire - Compiz clone with cool effects (it was chosen as the official GUI for Raspberry Pi OS) * Labwc - OpenBox clone for Wayland * DWL - DWM clone for Wayland * Hyprland - a dynamic Wayland compositor * Cagebreak - a Ratpoison clone for Wayland Pick a Wayland compositor and try it. Then decide if Wayland is good or not. I recommend that you try Sway first because it just works. Links * https://wiki.gentoo.org/wiki/List_of_software_for_Wayland * https://wiki.archlinux.org/title/Wayland Good posts to read: >>63810 && >>63416

I really don't get what you people are so angry about... technology changes overtime, that's not a bad thing. Wayland has blatant superiorities over xorg, any feature disparity will eventually be implemented "the wayland way". If you absolutely refuse to use wayland, guess what? Linux is modular, you can just use an older DE or WM that uses xorg, which I assume most of you already do anyway. If people thought this way 20 years ago, we wouldn't have pipewire, systemd, or proton.

>>74923 >Wayland has blatant superiorities over xorg And far, far more inferiorities which make it unsuitable for many. But two more decades and I might consider using it. Maybe. Further gaslighting won't convince me of your cause. Nor will >git gud do you any favors.

>>74923 Nooooo but other people are spending effort making something new and good and also they're TRANNIES!!!!!! I need to use X.org and SysVInit to win the culture wars! I would stick around to argue endlessly about what wayland was like five years ago (because Imagine updating your talking points, lul) but I'm off to send death threats to Lennart Poettering.

Wayland + KDE doesn't work with my nvidia card, but Xorg does. In the loonix world, people will endlessly tell you what they use is great, but every time you give what they rave about a chance, it ends up being nothing but a massive headache.

What is with these people being emotionally attached to a barely functioning WM and implying that people here of all places think that software needs to remain static? Change for change's sake is usually more trouble than things remaining stagnant however. Has nothing to do with anger but annoyance that a system is being pushed on an already quite fractured environment which looks like a still work in progress patchwork of minimal and private protocols and a whole bag of different composers. It took valve to get us to a point where one could pretend wayland could be adapted everywhere while still dismissing any deficits with this weird cultish attitude. Makes me automatically distrustful of you waylandgelicals. >it's just a protocol So is X. What makes it better? >it stops keylogging https://github.com/Aishou/wayland-keylogger >it will be ready soon Sure, I'll wait another 15 years for it to reach the maturity of xorg. Although I hope that wayland and xorg are going to be replaced by something way before that which isn't carried solely on the ego of its developers but an actual elegant solutions. >there wouldn't be no pipewire if people had concerns about functionality The adoption of pipewire over pulse audio went pretty smooth from what I can tell since that happened once it was already usable. Sure, maybe not all people involved with wayland are trannies but so far this is the style of argumentation I've seen so far from its proponents. Again, maybe just cut your losses at this point and get back to the drawing board.

>>74929 how does arguing with trannies make not another tranny just from the other side? asking for a friend

>>74927 Unfortunately I think this is just becoming more and more common. The modern Linux landscape has come to feel so much more trouble to use than it was even a decade ago. Sure, gaming is easier but now you're juggling multiple package managers even as a non-dev, so so much half baked unready software (wayland for example), disparate and usually undocumented configuration. Starting to feel a whole lot like old Windows. Think I'm finally over it, and will likely be making a move towards BSDs again. While far from perfect, they actually feel like a Unix system and not a mad max car most of the time.

>>74939 first and foremost Slackware also, elaborate really what do you mean if you don't want to put up with modern soykaf these glitterboys are 'designing' well you still have the choice to avoid them even on very popular distros like debian because hey everything's packaged anyway! well, true, systemd isn't gonna ever feel like true unix, but still is usable for end users. just don't adopt the soykaf you don't want to put up with. you still can choose which packages to install and until it's true you can stay with linux. also been seeing people complaining linux outperforms even greatly tunes bsds on servers out of the box. and hardware support is greater also, Slackware, or did i mention it already?

s/tunes/tuned DDDDDDDDDDDDDDDDDDD:

>>63416 >No screen tearing literally my #1 reason for using wayland. it's absurd to me that this was is just considered normal and expected behavior on x, none of the supposed fixes ever worked for me and even simple youtube videos had noticeable screen tearing. Wayland on the other hand seems to at least be prioritizing a livable UX, even if it's still rough and experimental around the edges. You can't make me go back to x for anything, I actually like using my computer instead of just philosophizing about how superior I am for using one subsystem over another.

>>74946 >none of the supposed fixes ever worked which ones did you try? if you used something like kde on low spec machine then yeah expect tearings. i have lower-mid machine with i3 and picom and no tearings with no configuration at all even 2K video which is afaik the best my cpu can do without gpu accel. some people do --vsync with picom. some use x.conf: Option "TearFree" "true" in Device section

>>63336 It's those tools that have to support Wayland, not vice-versa. > you will use our [...] trash init system and you will be happy I can see you're just another nagger.

>>75012 is it so hard to stay on reddit and not come here? honestly if you want to post SO much go to 4chan there's always a place for >>>>YOU <<<<

>>75012 >Everyone has to cater to X! >Why does nobody use X?! I genuinely can't tell if this is mere trolling or genuine delusion due to how frequently this exact argument comes up, let alone after 15 years of (non-)use. >inb4 you don't have to use Wayland I know. Then again, it is inconsistent with the many, many claims that Xorg WILL be removed from whatever distro I use in favor of forcibly ramming Wayland down my throat anyway. t. using Devuan. I figure you know why.

>>74946 >i need wayland to make videos play properly literal retard confirmed

its funny because you can probably get screen tearing while playing vids on wayland since video rendering is often done via an escape hatch and has nothing at all to do with the desktop

I was researching some xorg question of mine and stumbled on this: https://askubuntu.com/questions/1174798/xrandr-how-to-shift-output-that-is-lower-resolution-than-the-display-native-res Essentially the OP lost the right 25% of his monitor and wanted to set xorg to output a smaller resolution and shift the image to the remaining 75%. The accepted answer explains with pictures how this can be achieved. Since this seems like a very rare usecase I was wondering if wayland can do this and how. If this is a compositor/implementation specific thing, then how many of the existing implementations can do this?

>>63337 the most frustrating part is that progress of the loonix desktop has screeched to a fuarrrking halt just so wayland can play catchup with xorg for the next like 5 years im considering using bsd over linux as it seems like it's not going to be enshittified anytime soon due to it's relative obscurity and rejection of the antichrists (soystemD, rust, wayland, pulseaudio, etc)

>>75026 >if you can't get our archaic display server to work properly you're a literal retard OP probably forgot to install an X compositor for his i3

>>74751 It's definitely going to happen at the advent of so many guys going into computer science. I actually think it's going to be more men/boys using Linux and women staying on proprietary software with their social media - especially now that it's not some new hot thing anymore. >>74910 I use Ubuntu and I don't have snap, and hardly ever use Flatshit. Debian is about 7 years older than Arch, you know. I was considering switching to Arch but I'm highly skeptical of the AUR and all the uh, let's just say "eccentric" software maintainers. Why should I switch to something niche like Artix when I can go full Debian and compile most of my software from source?

>>>>74751 i still don't understand ever having to use snaps or flatpaks almost all standard package managers are better

>>74949 Have tried that, of course. Did not succeed. >>75874 I've used xcompmgr or compton maybe, both of which did not solve the tearing issue but have introduced more performance issues on my i7-1280P. I've migrated to KDE on Wayland and have never been more satisfied with how my laptop runs. No more obscure i3wm tomfoolery for me, I've had enough.

>>>76460 Might want to run something proprietary or not accessible in your repos. Also sandboxing can be neat but mainly it is for normal people and corpos

>>75398 nope, there's like two (and gnomes one I think) incompatible unstable and non-standard protocols for output management and they don't support that

>>76500 i understand it in corporate and servers, as sandboxing can be useful but for a normal end user i don't see much of a benefit sandboxing, in the rare case of it not being in your repo or broken in it: flatpaks can be useful but mint forces it as the default in the app store, and flatpaks can be pretty slow and it's bloat, so it just ruins the experience for many

The worst part is that even when someone tries to improve wayland it getd WONFIX'd by gnome devs.

GNOME is halfway through killing the linux ecosystem. It seems the only reason it is still going is because certain parties pay.

>>76948 Gnome isn't all bad... GTK is really fast and responsive on my system. I think Gnome turned bad sometime around 2013, because a lot of their oldest software is pretty good. Evince, the pdf viewer, is really good and I use it as my main reader for anything that's not epubs (Calibre for that). I'm really conflicted about Gnome, because they seem like a bunch of dunning kruger dickheads in discussions, but their desktop environment is relatively stable.

I accepted Gnome's rigid design choices that can't easily be changed and gave it a good go for a while but it just carries on being shockingly disappointing the more I used it. I shouldn't really have bothered with a DE that doesn't even by default allow the menu bar on all screens. Then I need holding super to just resize instead of give a context menu, I need to avoid activities binding to super by doing a custom binding to get rid of it, I need extensions akin to a browser that I'm worried will stop being supported, GTK4 made it harder for people to get slimmer title bars, I had an instance of gnome completely freezing probably due to an extension, it's just really not good. I think to have all the features built in and to be able to get on with using your actual PROGRAMs, you somewhat have to accept how your graphical shell wants you to use it. But Gnome is too torturing. KDE is fine, if you're like me it will STILL upset you on what you can and can't do to an extent but it's about as good as it gets. Maybe I should've just stayed on Sway because it literally never upset me ever lol, I just want to be modern and try instantly having way more features.

tried gnome wayland on my 15 yo macbook for the lulz and it keeps crashing doing random tasks among other things... I used to run debian and dwm on that system and it worked for years without any problem whatsoever. Literally who asked for a new display server? Shit just werks, leave it. Sure nvidia users have to edit the occasional nomodeset when setting up your OS but how is that enough of a justification for wayland? >but we need a better display manager for muh HDR and 4K hardware acceleration support in chromium browsers No. What everyone wanted was a solution for GUI frameworks, some sort of a standard and ideally cross platform way of developing GUI's. And now we're all stuck with electron and chromium. But hey, those window animations look really neat tho :o

All the actual information and real discussion was already done in the old thread. There is no more new information being posted, in fact this thread is rife with misinformation. The sole purpose of this thread is "culture war" BS. If you want to participate, go for it, but just know that what you're doing is more like ritualized combat than standing up for the truth.

>>77070 What is the truth then? This thread is nearly 2 years old. You don't think information changes in that timeframe?

I want to explore other options. Is kiwmi or DWL any good? I would like to use CageBreak (because I love Ratpoison) but it lacks some of the features, like window selection with C-b ' and the ability to remove just 1 split (currently you have to remove all splits with the only command. Ratpoison also has the : command the help system). Normally I just use Ratpoison and if I need floating windows, I will just use tmpwm or newwm Ratpoison command to switch to another WM (FVWM, CWM and *Box works well).

both wayland and xorg suffer from the real floss cancer freedesktop.org

>>80684 fuarrrking true. Hundreds of posts in multiple threads to argue over Wayland vs xorg when both are pretty much neutral in the face of actual cancer (freedesktop, systemd, pulseaudio)

>>80685 Can't imagine why people would do such a thing ;)

>>80685 wayland doesn't stand aside freeborkstop. Gnome affected it alot. Idk how much wlroots fixed the gnome issue. Probably wayland is going to be overtook by the same agenda. I don't know. I don't use it. I wish redhat died together with all their festering crap.

>>80673 You missed: >xwayland is an often-dysfunctional bandaid that also has been slapped on everything and is used as an excuse to both avoid porting and to avoid developing native functionality in wayland. This is a sore spot for both Wayland fans and xorg enjoyers because they both get to blame each other for it being both broken and essential, since it would either be fixed or nixed if the X install base wasn't so large and wayland development so languid.

>>63336 this thread is full of /g/tourists

>>63336 Arcan is the only hope against Wayland. Xorg has been taken over by Wayland activists and there's only so much that Enrico Weigelt can do to slow its planned death.

>>63336 damn and still better xorg

>>63336 I tried Wayland multiple times in the last few years; things did not work as I wanted them to work, so I always ended up going back to X11. Screaming about how X11 is old and outdated soykaf while Wayland is the new, shiny cure for everything is pointless when only one of these things is doing what I want it to do. >B-B-But systemd! I never cared about systemd, and the new sound-related stuff has not broken itself for me yet, so both are irrelevant. And Gnome would be fine if Ebussy would stop acting like a retard and realize that making things different for the sake of it is just fuarrrking stupid.

>>85748 The issue is that Wayland's security is not coming from it being better but from the fact that the so-called "developers" of it just cut away plenty of crucial elements from their work and expect it to be implemented somewhere else... which may or may not still end with the same or worse security issues being introduced to the system, but the fuarrrkwits from the Wayland development team will be able to point out that it's not their problem. The security of Wayland is all just an illusion, a meme.

>>78831 >Wayland is the future and it always will be. "fifteen more years bro, I swear, it'll become good, just wait fifteen more years pleeaaasse" >and the future is now typical gayland supporter, literally fails to understand the concept of past, present, and future. sad. >The following have moved or are moving to wayland >The following are being strong-armed by Guhnome (RedHat) into wayland, or be "left behind", because GTK5 will not support the most common used display server by all the UNIXes here, fixed it for you :)

>>85875 ugh, you can spot a 4chan refugee from an airplane wayland has already won, xorg is dead and depricated wayland is technically superior, you talk about simplicity all day long yet in this instance contradict yourselves, wayland is simpler, trims off useless features (server-client architecture for example) and is more secure

>>86079 >xorg is dead and depricated works on my machine >wayland is technically superior then why can't firefox remember the Picture-in-Picture window position and have it set to remain on top by default? >you talk about simplicity all day long where? >useless features >server-client architecture child >more secure any program can still access the whole filesystem, retard, it secures nothing

>>86080 >works do you know what deprecated means? windows xp works too... >it doesn't have feature X I can claim that Windows is technically superior for having a certain feature X, and do the exact same thing for Linux and MacOS, it doesn't disprove that overall it is superior >where you as in "your kind", your kind as in "the chronic 4chan user", you need more details? >child I know your self-hatred and low self-esteem has gotten you into the habit of calling people names when you are proven wrong, but we're not running mainframes, it is reality, grow up >filesystem are you serious LOL it's not the compositor's job, use a sandboxing software it's clear that I'm talking to a larper, I will no longer respond

Client server architecture was superior and correct. Wayland is another cancer designed by illiterate glitterboys same as the entire freedesktop thing. Stop killing open source retards!

>>86085 >Client server architecture was superior and correct Client server architecture was never meant for desktop, X11 on the desktop is a hack, it was meant for labs in the 80s. >as the entire freedesktop thing X11 is under freedesktop.

>>86083 >windows xp works too with all modern software, unmodified? please show the rest of the class >I can claim that Windows is technically superior for having a certain feature X do it. which feature? go on. can windows have multiple distinct keyboards in one session? can it have the screen rotated at an arbitrary angle? can wayland? if it were up to you gnome "people", you'd probably remove pipes from posix because "the use case is not clear" >habit of calling people names when you are proven wrong >we're not running mainframes you are either a child or have never worked in tech, and it shows >it's not the compositor's job, use a sandboxing software here we go again with "wayland is juts a protocol" and "it's not supposed to do that"; stfu >I will no longer respond you will

>>86091 >Client server architecture was never meant for desktop Maybe??? It's cool in a way, I can just connect to /tmp/.X11-unix/X0 (or $XDG_RUNTIME_DIR/wayland-0 iirc) and open a window and draw something without loading a bunch of dynamic libs. And I'm sure the socket is at this same path on most Unices. If the client-server model is not the "right thing", what is? Only one app at a time can access the video memory (or any hardware) (it depends). Eventually some single program has to send the frames, in order (dvi, hdmi, vga, displayport, usb are all serial).

>>86102 I agree that it's cool, but still doesn't mean it was meant for it, nor that it is superior nor correct as the anon said. I just wanted to point out that the post I was responding to is wrong, and it is. That aside anyone should use whatever software they want, doesn't really matter. Something not being meant for a certain purpose doesn't mean it can't be used for it, and be good at it too, either. Unix, for instance, was never meant to be used for desktops, they weren't even really a thing when it was created, but it is a good OS for it IMO, with problems and limitations, sure, but still good.

>>86234 bruh, he was and still is broke. the man lives off of donations and events, for fvck's sake. making good software isn't profitable; companies must pump out new soykaf, and so quickly (therefore resulting the program being soykaf to begin with), otherwise investors and shareholders get all uppity and throw a hussy fit. In order to make a good system, one must already be rich, but you don't get rich by making a good system. It's an artificial chicken and egg scenario: it can be resolved, broken; all that needs to happen is some worthless billionaire spend his wealth funding something better than what's currently being pushed, instead of sitting on his wealth and jerking off to the cash mountains. I will never happen tho, because billionaires are absolutely retarded; and not one of them even know what the words "window manager" mean. all they know is how to manage other managers into hiring marketers in order to get richer, by spying on normal humans using even more bloated and invasive soykafware, thus perpetuating the software equivalent of the human centipede

>>86251 The Foundation is beholden to its board and principal chairs, who represent major stakeholders. "The public" is not considered a major stakeholder. Platinum members are: Meta, Ericson, Intel, Fujitsu, Huawei, Microsoft, Hitachi, Samsung, NEC, IBM, Baidu, Oracle, Qualcomm. These are the members who will decide on the appointment of the next Benevolent Dictator For Life if any. Linus himself has not appointed a successor. They will also decide on the direction of the project. Brodie Robertson made at least one video on the subject of having asked the Linux Foundation to give some support to desktop-oriented projects with a fraction of the money the Foundation currently invests into cryptocurrency and AI development, to which a Twitter foundation rep reacted by saying Brodie was turning into Bryan Lunduke and that no member of the Linux Foundation is currently interested in linux for desktop.

>>86266 >logind, gvfsd, dbus, etc etc. These are all a part of creating open source malware ecosystem. Open source software that despite having a free license does not actually provide any freedom by sheer complexity. dbus serves no purpose at all, but retards don't know how to use a socket library lmao. These are not desktop. These are Microsoft Linux.

>>86274 Well, in a 20 day uptime dbus has used 4 seconds of CPU and is using 2.2Mb of memory and elogind has used less than a second of CPU and is using 1Mb of memory, I'm not defending that Wayland soft requires them but it's not *that* bad

>>86292 dbus is wrong because it doesn't need to exist and serves no purpose, a higher level library for unix sockets would do everything it does without requiring a separate daemon with retarded semantics. logind is wrong because of its tight(tm) integration with abominations such as polkit/dbus/etc. also it dynamically creates and removes /run/user/$UID on login/logout which is fuarrrking retarded and interrupts my workflow. ultimately, it does such obscene things as handling acpi events. i tell you it ought to be 10x simpler. the login manager mustn't do anything except tracking logins and exposing an unix socket interface to query these logins. seat managements must be a separate daemon. the same for acpi events. things like /run/user/$UID must be allocated statically at boot time and never touched again. holy fuarrrk even if you have tens of thousands users doing things at the same time 50M tmpfs on /run/user would be able to contain everything, why the fuarrrk does it need to be allocated/deallocated dynamically at all?

>>86316 did you, by any chance, forget to take your medication?

wayland will just werk when we have nuclear fusion

waylanders can I do this yet?? https://wiki.archlinux.org/title/Chroot#Run_graphical_applications_from_chroot

Wayland will just werk when we get this whole perpetual motion thing figured out

>>86533 Good friends, sweet friends, let me not stir you up to such a sudden flood of mutiny. They that have done this deed are honorable. What private griefs they have, alas, I know not, that made them do it. They are wise and honorable and will no doubt with reasons answer you. I come not, friends, to steal away your hearts. I am no youtuber, as Brodie is, but, as you know me all, a plain blunt man that love my friend, and that they know full well that gave me public leave to speak of him.

>>86560 You are gay .

>>86251 >International finance has ensured these institutions are rotten from the top down. Interest-carrying loans and lack of periodic debt forgiveness... anything else or does this cut at the root?

>>86251 >I read someplace that the Linux foundation only spends about 3% of its donations on maintaining the kernel. Last I heard, the linux foundation only serves to protect the name "linux". It's not involved in the development.

>>86592 Same way Facebook consolidated social media. Instagram was running laps around them with their photo app, Facebook bought it. WhatsApp was obsoleting Facebook's texting with a simple and effective business plan, Facebook bought them too for billions and killed privacy. FAGMAN are one with the government, it's impossible to disrupt them because they'll buy up, kill, bury any superior new competition.

>>86630 >x11multimonitor setup is broken and will never be fixed. Are you mad? Literally what's your problem? Just use xrandr to set different resolutions, it works out of the box.

>>86630 if you don't know how to use the xrandr cli you can instead use arandr, it's actually more intuitive then the windows monitor setup

>>86630 Waylanders can't actually point out any meaningful advantage of the soykafty excuse for a display manager and are forced to lie and manipulate. Nothing new, nothing new.

>>86668 >You arent paying them I'm not, but ibm is paying redhat and canonical is paying ubuntu and they're both the main parties pushing wayland onto everyone else, I dont think you understand the dynamic at play here >no one is forcing you to work on wayland. Objectively wrong. You're a bug tester when you use wayland, I ran into a lot of issues with it recently but most of them were thanks to the stupid forced vsync and abysmal nvidia support. The kwin compositor also runs like dogshit and drops frames in the 720p desktop on a 980 Every DE dev also has had their update pipeline uprooted by the messy move to wayland, leaving bug fixes and feature adds by the wayside just to make sure their DE doesnt get obsoleted when every mainline distro moves to the shiny new toy. XFCE has a multitude of decade+ old bugs that make it unusable for me and I'm sure converting their very X11 reliant DE to wayland has taken up a lot of time and resources, potentially worsening performance thanks to the usage of a less optimized compositor if they decided to go with kwin or whatever >At this point majority of linux users are on wayland systems. Majority of windows users are on windows 11 systems, people using mainline distros arent in control of what canonical or redhat pushes onto them >Modern features are regularly implemented You mean feature? Waylands only feature is HDR, everything else is easily done in ARANDR. I saw a bait post on reddit asking what the features of wayland were over X11, got 300+ replies and the only features listed were >HDR >VRR "done in X11 easily" >Muh security >X11 is old 10+ years of this soykaf for HDR? Really? >hardware vendors have actually started contributing to linux Wayland had nothing to do with that, some hardware vendors are worse with wayland! Nvidia's drivers on wayland are soykaf and need config tweaks just to get GPU accel, the control panel is missing 3/4ths of its options and temperature/fan control is missing. Yes this is with the latest drivers and latest kernel, the average user doesnt care about the cope so dont tell me its nvidias fault because I really dont care >You have no basis to call their work a waste Yes I do! The move to wayland has been a catastrophic soykafshow for basically 0 gain, all this buzz and compatibility breaks over something as transparent and featureless as a display manager, at least systemd makes it easier for noobs to string along startup processes. Wayland brings you nothing, please think about it for a minute.

>>63336 https://github.com/X11Libre/xserver Wayland devs are seething, coping, malding, and attempting to damage control, but it seems that the sole person that attempted to patch up X11 for the last few years decided to just fork it instead of hoping that Red Hat drones will push his commits. I'm not sure if in the current, corporate-controlled state of Linux this will bring any change, but hey, kudos to this guy for trying to stand against this bullshit.

>>86720 >I dont think you understand the dynamic at play here I think you're being too kind here. This guy is shilling Wayland hard, I doubt he even understands that or even why he is. He likely does not care about what is happening. It's the same for a lot of these unnatural changes happening in this sphere.

>>86738 I know and accept that this guy is kind of a nutcase, but the fact that there's at least some kind of resistance against Wayland is better than nothing. I doubt that it will succeed with just him being involved, but maybe it will open up the eyes of other people. I agree that the rhetoric that he's using is absolutely retarded, but I also agree with his disdain and hate towards the corporate side of Linux.

Nobody needed to replace X11 until the linux youtubers told them they do

>>86756 I didn't love X11/Xorg until I watched that Tom Scott video about making an emoji keyboard, and then learned how trivial it is to implement on linux; only then to find out about that this "new" thing called wayland - that is even more limited than Windows - is being aggressively pushed because "HDR" (the feature that 99% of users actually neither have nor need), at the expense of screen recording, clipboards, multiple distinct input devices, rotating the display, firefox's picture-in-picture window not staying on top by default, windows forgetting their past positions, and so on... all for a meme technology, of which half the users are using low brightness (less that 3000 nit) monitors, and are therefore not seeing a difference. And I see people arguing that it's more "secure". How? How could it ever be? All programs still have access to everything, every single file, the whole enchilada if you will, just not each others' display output. They cry tirelessly about "wayland is the future, you can't stop it"; but not once have they sat and wondered why people actually want to stop it. If anything, wayland pushers and worshipers remind me of the last couple phrases of this story https://clarkesworldmagazine.com/watts_01_10/ It's so tiresome -_-

>>86738 There is also OpenBSD's xenocara fork. I heard that Hyperbola GNU/Linux-libre developers have ported it to their repos. >https://www.hyperbola.info/packages/extra/x86_64/xenocara-server/

>>86760 accept yours and then kill yourself. Arcan is starting to gain traction. Linux will survive

>>86776 They all say that. "work in progress" its been "worked on" for nearly SEVENTEEN YEARS! and it is STILL WORSE THAN X11! Give it up man, the only thing wayland has to itself is the plethora of bugs, userspace breaks, gargantuan amount of compatibility loss, lowering the amount of available compositors to FOUR aaaaand HDR, something basically nobody uses or notices All you idiots just point to "X11 old, need new toy" what? Software doesnt age like a car or milk or a person, if it works and is something as transparent and featureless as a display manager protocol then it just fuarrrking works and doesnt need to be reiterated upon! I'm not attacking the inherent underlying fundamental basis of technology here, all I'm trying to ask is why we have to blindly accept this new display manager protocol hodgepodge that is clearly inferior to what came before? Am I a Luddite because I dont automatically believe that any new proposed protocol is magically gonna be better than the one that came before purely because its younger? fuarrrk no! I dont know where this unconditional belief in "new = better" but you people need to actually begin to question if uprooting every bit of the linux desktop software set every 10 years is a good idea. The bugs and compatibility loss of this practice is what has given linux its bad reputation among normal people. It needs to stop

>>86781 i was telling you that new x is wip are you retaded

>>86781 >>86782 he may be retarded, but he's got the right spirit at least :)

>>86782 NTA but your post in >>86776 looks like a baby's soykafty diaper exploded on the keyboard. It is you who is retarded for wondering why nobody understands you when you talk that way.

>>86784 i might be retarded but i couldn't care less because you come across as somebody who just wants to soykafpost

On one of my machines X runs fine so I just use that, however on my laptop it always had screen tearing running it due to some buggy integrated graphics drivers. Using Wayland feels better than using picom in my usecase so I use it, I don't care however about running native wayland programs, I just use whatever I use elsewhere. The debate is meaningless to me because wayland runs X programs just fine, it's like arguing about whereas you should run programs within a terminal or through a launcher, like just use whatever is better for the usecase, it doesn't matter much.

>>86735 This person has the worst optics in existence. Apparently he claimed that WW2 was forced on Germany... He also put "make xorg great again" in the project description while also claiming that project is politics free.

>>86790 maaaaybe everything YOU use on Gnomes wayland session works, unfortunately EVERY wayland implementation is UNIQUE and has its own bugs and incompatibilities because NOTHING IS UNIFORM OR STANDARD!!! Gnomes wayland may work with gtkpod, KDE's might not, i3's might not, hyprlands might not, cinnamons might not, whatever the hell Raspberry Pi OS has implemented might not, XFCE's cobbled together at the last minute implementation definitely wont. It's all a crapshoot and for what?? HDR?? Wayland has made life hell for DE devs due to the need for EVERY DE dev to rewrite everything to retrofit either gnome, kde, i3 or cinnamons compositor to their DE among so many other things in order to get their DE somewhat functional in a somewhat reasonable time with a soykafty implementation of wayland and not get obsoleted by wayland only mainstream normal people distros like ubuntu, fedora and redhat who afaik have dropped X entirely in their latest releases The post even stated wayland has pretty much killed off every independent DE project. >No support for lightweight toolkits including FLTK, FOX Toolkit >No support for most desktop environments except GNOME/KDE bloat and several i3 clones >No support for cwm, Fluxbox, FVWM, IceWM, TDE, JWM, or Window Maker >No official support for BSDs I'm telling you, wayland has broken so much and has forced everyone else to clean up their soykaf! This has wasted so much dev time and has left the linux desktop stuck in 2018 and regressing, its yet another layer of broken compatibility that'll end up breaking heaps of soykaf throughout the years As for X11's utterly impossible to fix tearing problem for intel Add Option "TearFree" "false" Option "TripleBuffer" "false" Option "SwapbuffersWait" "false" to your /etc/X11/xorg.conf.d/20-intel.conf Fixed, I can have a tear free experience on X11 by doing an almost identical process on my 17 year old budget amd GPU. You dont need wayland!

When I first entered this site there were a lot of "I use only the command line and the framebuffer" types, now it's all the same soykaf all over again but way lamer, can't wait to see you glitterboys defend Wayland when a newer Linux graphical system comes out.

>>86807 >defend Wayland name one (1) attribute that can be defended in that heaping pile of soykaf

>>86807 If the corpos somehow manage to uproot the graphical system for a third time I'm genuinely moving to BSD or Morph because it'll be so fuarrrking over for linux at that point

>>86813 Linux noob here. What were the second and first times?

>>86813 What is Morph? Never heard of that before

>>86825 why not? we currently have multiple. If the cancer grows too much then you have to do that. That aside the trivial solution is simply stay on an old version. Im doing it right now cause one update broke my software.

wayland remains barely functional and the obvious sabotage of continued x11 development is appalling. Xlibre looks great and the maintainer has his head in the right place. Good luck, you're up against the sjw corpo rats

>>86821 MorphOS is a semi continuation of Amiga for PowerPC devices, its goal is to be as fast as possible and retain backwards compatibility with most old programs Unfortunately it is closed source and costs money but its being maintained by people who actually know how to program

>>86793 There is no way he isn't a Neo Nazi thinking Germany innocently got pulled into WWII. Chuds tend to be awful at maintaining projects it seems so I dont have high expectations.

>>86830 I agree. Metux can be better at communicating but I really wish the best for this project. The fork was overdue.

>>86835 Regardless of his politics it's good that it's bringing light to Red Hat's stranglehold on Linux. >>86818 It's far from over, we're at least several years out from the start of that happening; the recent xlibre fork is a very good start in countering Red Hat (even though it will probably be short lived). >rust didn't the person who was putting rust in the kernel get booted out? the rust hype in general seems to be dying out from what I've seen.

>>86850 don't add junk and your big programm will be just fine. anyway x11 is mess simply becauss lizards did it purposefully. wayland isn't any better though, ironically. doesn't solve anything either. x.org had to forbid idiots to contribute from the start and there would be none of those issues. making software """modular""" doesn't magically make it easier to write software that's not soykaf. wayland is soykaf. GNOME and KDE are soykaf. everything freedekstop is soykaf. you're a soykafhead too

>>86850 >Have you ever seen a brilliant but mislead programmer put EVERYTHING into one big program? That becomes an issue over time, it becomes hard to change anything without breaking something else, compile times increase, more tests need to be passed to make a release. To be blunt, usually this is an issue of programmer incompetence. I have worked on massive spaghetti systems, but I've also worked on even larger systems that were beautifully constructed and properly modular and flexible. The difference was the average competence of the programming teams with access to the system. Everyone cannot program and the death of computation as a science killed any chance for technology to improve. >...so it doesn't have much to do with Wayland itself. So Wayland refused to support any unified protocol for a majority of basic features and told downstream maintainers to get fuarrrked and do it themselves? Sounds like it has everything to do with Wayland. >The main benefit is that every frame is perfect What defines perfect? This is just Wayland propaganda. Is a movie skipping frames due to Wayland not asking for a new frame on time "perfect"? What about low latency system hitching because something stole time away from the compositor? Perfect is a useless word and it should not be used unless you're trying to advertize to troglodytes.

>>86850 damn retard, you really wrote all that just to to say: no use case for wayland >but muh developers that, I'm afraid, sound like a you problem. Enjoy your fragmentation :)

>>86850 >every frame is perfect Hitomi (Hidden manually)

>>86850 >Input configuration is now done by the compositor, including XKB options and device configuration. Configuring specific devices is therefore compositor dependent how quaint; I gave you a command, and you give me theories. show me how to set my second physical keyboard to arabic under kwin, mutter and/or wlroots. >The main benefit is that every frame is perfect I can turn on vsync just fine, so every frame I see already is perfect. Quite literally gained nothing. >the compositor is not a separate process another deficiency: the compositor can now take down the entire session if/when it crashes. Doesn't happen on Xorg. Seems like a regression, because it is >you can implement various things that were not possible with a non-composited display server again, example(s) or kill yourself >and have a guarantee that your users can all use them. already debunked here >>86533 >Wayland to not be easily leaking windows and such here we go again with you security theater that you yourself don't even understand

All the time spent typing in this thread that could have gone into programming to improve X11 and Wayland and make a viable third option

>>86872 >>86873 >>86874 Chanminers generating slop AI responses that account for nothing, ignore these users

>>86874 nobody said its complex, my point was modularity is automatically better because it makes mistakes easier to fix as it keeps them localized as for external api's i would always use a interface module to shield internal code from external influence. The stable api promise might get broken or you might decide to change api and then its a lot more work if you didn't separate your own codebase from the external api with an interface.

>>86876 >The stable api promise might get broken or you might decide to change api Well you pay for being idiot don't you

>>86877 less work to prepare for changes than stalk devs for possible change announcements. You get extra time with deprecated anyway.

>>86878 >i can't maintain integrity of design you take the Linus meme too seriously. integrity of design in Linux is impossible because it would take far more effort than corporate glitterboy would ever put into the project, so Linus accepted that and went on. and he pays for it, since Linux is a huge mess now and bugs will last to its last days, there's so many of them.

>>86886 >more bugs IT got trendy and now halfwits push for entry. Current freshman now need help with downloading and installing an IDE - im not joking here, friend complained to me about this as he works freshman tutor. Many of these are also trying to contribute to foss for their resume with disastrous results. That aside i stand by the protect your code from external api advice not for linux but coding in general because i started programming at 14 and by now a lot of the libs i used stopped working / changed. Even the stable libs that still work are simply allowing you to use the legacy calls but tell you if you want the last decades improvements use these new calls. So tldr its really better to prepare for change when you plan your project to live more than ~4years

>>86890 you can extend and improve internal modules without breaking their old APIs. you would still need to clean up once every 10 years (maybe) but it would be far easier than some external glitterboy library breaking compatibility. also when you separate internal subsystems and you need to remove an old function you at least know that you only break code that uses this function in other subsystems. you seem to think i make it sound easy what i'm telling though is that separating one codebase into several doesn't magically solve any of the issues of legacy code.

>>86863 >how quaint; I gave you a command, and you give me theories. show me how to set my second physical keyboard to arabic under kwin, mutter and/or wlroots. I'm not a manual. Go and figure it out yourself. >again, example(s) or kill yourself You don't need them, a non-composited display server cannot do certain things by its nature, it needs a compositor, or it needs to be a compositor to do those things. >I can turn on vsync just fine, so every frame I see already is perfect. That is not supported on every hardware platform, and it's an implementation burden elsewhere. >already debunked here >>86533 The users of Z compositor can be guaranteed of Z's capabilities once they're properly implemented. >here we go again with you security theater that you yourself don't even understand There's new possibilities as a result that you obviously don't understand. >>86864 >Nobody cares about the cope, nvidia is the largest gpu oem and their pre 2015 cards need to work! It's 2025, those cards are over a decade old now, and aren't far from entering the scrapyard. It would be a complete charity case for them to be redoing the drivers to work on Wayland. Without a source code release they're destined to be bricks. Given another 5 to 10 years they will be completely irrelevant and your complaints will be forgotten anyway. I'd strongly recommend choosing your GPU better next time. >Name one that actually breaks userspace and cannot be fixed, I keep hearing about these mystical impossible to fix failings of X11 but nobody can provide me with an example X cannot be composited, it needs some other process to do that. X is stuck on a broken multi monitor model. Your monitors are actually just one big screen to it. This requires a bunch of hacks and sometimes some programs will just be launched inbetween screens anyway. X requires pointless extra IPC steps that don't actually help you. X requires a config file that, although usually autogenerated nowadays, is hard to understand for new users and is full of X anachronisms. X's network transparency was not made for the modern use case of GLX and big bitmaps and so on. It's touted by a minority only who could be moved to external programs, and it increases attack surface for everyone. X as a standalone display server uses an antique driver system that requires more maintenance, that is forced to be needed but modern hardware no longer needs it. X is a giant roadblock to making use of jails/containers/VMs/hypervisors in concert with graphical programs. It even makes the basic kiosk use case harder for no good reason, you need all this extra stuff that adds unwanted functionality and attack surface. X doesn't scale DPI/windows beyond very rudimentary "set the DPI value and forget it" type configuration. X cannot remove things that make it X11 (X11 is a standard), therefore these things can never change. It makes X inflexible and brittle. Oh yeah, and X seems to have artifacts with TearFree enabled that I don't have on Wayland. Eliminating tearing doesn't fix all the artifacts, and it's not a failing GPU either. >The display server handling heaps of stuff is a GOOD THING, less work for downstream devs is a brilliant thing that lowers the skill and time barrier for devs! Why on earth would you want a server that does less? A display server is not a car, hacking away "useless" features wont make it go any better The compositor implementer can choose what protocols to implement, there are many protocols adding a lot of functionality into the compositor process. See: https://wayland.app/protocols/ >>86880 The software backing your hardware is not fine. The company you bought it from doesn't support your platform properly and the drivers are in a legacy state.

>>86912 they recently had a hotfix to a hotfix on windows. nvidia drivers are simply soykaf on both and the OS itself cant fix that.

>>86940 >Half of the people I know use intel dual core laptops, granted they're 2nd gen-5th gen i3/i5's but still in the same ballpark of the Merom Core 2's from 06. It depends on who you're around I guess. It became so cheap to get a quad core (even in laptops there's many generations of them to choose from) in most places at some point that you either have an extremely low power use case, truly just web browse with high tolerance for performance hiccups and hardware decoded video and have zero other use for computers, or are just very poor. >Maybe Americans are just picky about computers? Is it somehow specifically American to want to do more than just watch YouTube/TikTok/Bilibili and read web pages? Artists, engineers, videographers/video editors, photographers, gamers, game developers, AI enthusiasts, some portion of programmers and scientists, they're all using more than the bottom of the barrel computer. You don't have any in your country? The extremely poor/retiree with zero technology related hobbies/teenager/etc bracket is choosing between mobile devices and whatever OS comes with the computer, they're unlikely to care about Linux at all. All iOS and Android devices can watch videos and use a web browser so they directly compete for this bracket. Not much room for Linux to be relevant anymore in that area. >AMDGPU does not work on GCN1 Incorrect, I've done it. You need a kernel parameter to enable it. Of course, behavior on different models may vary and it's experimental support. GCN 1.0 is EOL at this point anyway. > AMDGPU would cause it to not boot I remember certain outputs being buggy or nonfunctional, it may have been a HDMI issue. I had good results on certain outputs. Again it was never officially supported on amdgpu, your experience will vary. >W10 IoT Fair enough, but this sort of LTS Windows distribution becomes a niche and not the mainstream. >From a linux nerds perspective its unreasonable but from a normal persons perspective its pretty dumb, 8 in every 10 computers on this planet use an nvidia card so it should be elementary to support their older and still incredibly popular line of GPU's Again it's simply not possible unless NVIDIA does it. If their drivers are only compatible with X for older models then that was their decision, unless they release the source code there's nothing the community could do other than attempt reverse engineering with disappointing results typically. >>86942 I did not mention HDR once in that post. Again, HDR was added over a decade after Wayland development began. It isn't relevant to the core protocol. >manpages You're reading this, right? You can read. RTFM. That's all. >Examples... You clearly don't know what you're arguing about, go look up what a compositor is and you'll have the answer. >why must every compositor reimplement everything? what about compatibility? You implement the protocols that are required, then the ones you want or need and say you're compatible with those ones. A lot of compositors share a large library that does a lot of this heavy lifting. This happened on X11 (XFree86 to X.Org transition, various extensions added over time causing instability years ago) and to some degree whenever someone writes a X window manager (ICCCM, EMWH). It's just that the ecosystem settled down a long time ago so you don't notice it and people learn from preexisting code bases to make their X WM. >buy new thing, consuuume, consuuuummme, old bad, new good, keep buying If you want to use old end of life hardware, you have to accept that you're basically either in the extreme budget bracket or are doing it for the sake of it.

>>86949 and how is wayland relevant to your post? you still haven't addressed the point of wayland solving nothing.

>>86949 damn, you really wrote all that to say that wayland has no equivalent setxkbmap -device 11 ar worthless protocol, no use case

From: https://linuxiac.com/france-launches-government-linux-desktop-plan-as-windows-exit-begins Unattended remote access and accessibility for users with impairments are both huge problems with Linux desktops, Wayland essentially makes those things completely impossible to solve properly. Unattended remote access to the desktop is completely impossible in Wayland so any organization that expects to access desktops without someone clicking a dialog to allow access won’t ever be able to migrate to Linux. Even with explicit permission from the user, only certain applications that implement a particular desktop environment’s extensions will appear, other applications may appear but not allow interaction or not appear at all. For example LibreOffice may not show up on KDE and/or Gnome during remote access because it uses neither the Gnome nor KDE base libraries – in practice, hopefully it has implemented the protocol extensions for both DE’s Similar problems exist with accessibility tools like screen readers and assistive input methods, making access for people with physical impairments a minefield, and in some cases presenting legal discrimination liabilities for organizations trying to migrate users to Linux. These are showstopping issies that are now inherent to Linux after the reckless migration to a desktop graphics standard that is intentionally designed to be fundamentally incompatible with basic requirements for business adoption. It couldn’t be worse if it were intentional sabotage and couldn’t have manifested itself at a worse time. Linux desktop environments are their own worst enemies and Linux as a whole is reaping the harvest of the stupidity that they have sown.

>>63338 do you want to reinvent the wheel, or do you want to fix the one you fuarrrking got? there isn't an in between.

>use Wayland >program in rust >straight, married, having kids soon Cry about it

I mean I don't really want to use Wayland, but on Nvidia GPUs you need to use "Full Composition Pipeline" to get rid of tearing, no? Nothing I tried worked at least, not really. It adds a lot of input lag to the point I can notice it. If there's a way around this on X11 I'll gladly go back because I need a screensaver for my CRT.

>>92716 Might be solved on xLibre >>92715 Kinda unfair to your wife and kids to hide from them that you're gay

the foss world is internalizing all of the same monopolistic authoritarian myopic attitudes of corporate big tech, lots of moving fast and breaking things, lots of evangelism and hyping up products that are not finished or even started or clearly defined, lots of vendor lock-in and hierarchical bureaucracy and irrational dogmatic thinking and "my way or the highway" egomaniacal project leaders who act like steve jobs-esque "ideas guys" rather than software engineers and don't really understand or care about the technical aspects of technology. whenever i saw the famous "gnome rotting in threes" blogpost many years ago and the quote from the gnome project leader about how extensions and customization are bad because it "weakens our brand presence" i knew it was all downhill from there. and the authoritarian apple-like thinking even gets internalized by most of the userbase, they say "well, the linux emperors over at red hat have decided that things like wayland and systemd are the only way forward so we must accept it as a fact of life" and they lay down and roll over even harder than windows users do. windows at least has a very vibrant and underrated hacker community devoted entirely to defying the wishes of microsoft in every conceivable way and removing all the stupid crap microsoft shoves down users throats and bypassing restrictions and extending its functionality in all kinds of clever ways, wheras the linux community has ironically lost much of its hacker spirit and is now full of brainless zealots who say that linux is this perfect holy temple of purity that must be respected and never criticized or tampered with by the users.

Plasma 6.8 is removing X11 sessions entirely. I mostly use Wayland myself, but I'm a bit sad about this because for example some older wine-powered software doesn't work properly outside of X11, XWayland doesn't fix those issues I had with tablet support in ZBrush and Substance for example, and it's reported to work well on X11.

I almost exclusively used my KDE/Plasma setup with Wayland, I decided as an experiment to try use X11 for a while and to my big surprise, it's all either the same or slightly better with X11. Steam, for example, seems slightly more responsive, and apps that were running through XWayland before, now start significantly faster. I don't use multimonitor setup or HDR, so there's that.

idk niggers wayland just werk (You)'re just a bunch of le heckin contrarians

>>63336 >toolkits which dont support Wayland dont work on Wayland >DEs which dont support Wayland dont work on Wayland >WMs which dont support Wayland dont work on Wayland >"No BSD support" https://git.sr.ht/~shrub900/neuswc your complaints are bizarre. its completely OK if you dont want to use Wayland, you can just stick to the 1m+ SLOC 40yo unmaintaned slopfest that is X. personally i prefer Wayland for all its faults (none of which you identified)

>>93460 x11 doesn't support sway, what a piece of junk.

lainanons, what are your opinions on x11 vs wayland for daily driving? is x11 going to be outdated tech soon or is wayland just reinventing the wheel?

from what I've used and seen, wayland is full of very annoying, borderline unbearable people, so I won't be using it any time soon. niri was nice to play around tho

"hey anons please start a flame war in the replies" To me, having a meaningful choice of display server is more important than the (mostly overblown and meaningless) benefits of choosing one over the other. I want the current situation to continue forever. I want 3, 4, 10 competing fuarrrking display servers. Billions must CHOOSE.

>>93733 strewth. I like how Wayland is kind of intended to encourage multiple implementations of the core protocol with different features. X technically could be this way but hasn't been for a very long time. Long story short unless you have a very particular set of requirements, there isn't really a reason to pick one over the other in $CURRENT_YEAR. There are reasons to pick particular desktop environments. I avoided GNOME back when it was X11-based and I avoid it now.

I soykaf on wayland for a long time, and then I tried it again recently, and all the annoying little issues I had before were gone. I don't want to like it... but it seems to get rid of so many tiny little issues I had with x11 that most people wouldn't have noticed. One issue, though. Screensavers. There seems no simple way of getting one working for my CRT on wayland.

>>93719 I mean I use X11 because I like dwm+dmenu and my wife is using Wayland because she likes KDE Plasma. They are fine for their use cases. And, by the by, xlibre is doing just fine. Question: What is the point of this thread?

we don't need 2 useless 'debate' threads that are not going to go anywhere on the same topic

we dont need another thread on this. the entire topic is a waste of time and anyone who actually uses linux instead of just talks about it would know that

Next big KDE version releasing in fall (iirc), will have x11 fully completely utterly deprecated and unsupported. My only personal interest in x11 is running old proprietary windows software under wine, maybe I'll install xfce as a sidekick for those rare usecases.

This thread is pretty much fully out of date now. Virtually everything fundamental works in Wayland now, there's only a handful of experimental protocols that have to make it to staging to fill up the last niche functionality gaps. The only part that's really important is session management, the rest is small stuff or stuff like trying to implement support for that old horrible UI multi window layout that a very very small number of programs still force (this was last popular in the early 2000s and it was losing ground like a sinkhole even then). There's some other minor improvements in the pipes but honestly a lot of users will struggle to know they are not there. Using Wayland over the network is supported by various projects with different approaches. XDG portals and Pipewire should be standard otherwise your distro is broken crap and should be ignored. The griping about accessibility is valid but it is out of scope for Wayland, it was highly irresponsible to not break it out of X before into an independent project but there's simply no way to force it into existence in a bazaar style software ecosystem. X even in its diminished form has way too big of a scope for a software project, and despite its huge bloated scope it can't support new types of window management or multi monitor setups correctly, and has other odd limitations. I simply have NO idea why Unix-like minimalists like it, it is the stereotypical bloated runaway train project. However there are several ways to use X11 inside Wayland now, both rootful and non-rootful, and it's likely that independent X11 use will gradually die as it no longer confers advantages to 99% of users.

>>94213 Oh dear... Why does any Wayland advertisement have to be that rude? You know that crapping on others do not give your argument any points, right? If you truly believe in something, than point out what's good about it. On the topic of "functionality gaps", I'm a bit ootl: are there any good solutions for global hotkeys? Ideally without having to rewrite the whole application with some specific Wayland-only library.

>>94216 >global hotkeys Depends on your compositor. In sway those are right in the confiig file.

>>94217 Could you please elaborate? So I can register some hotkey to be transferred to an app or what?

>>94218 You can bind arbitrary hotkeys to arbitrary program invocations using the config file. You can also change the keymap programmatically using swaymsg. Apparently sway doesn't yet have native support for sending key presses specifically, but other compositors like hyprland do.

>>94216 >Why does any Wayland advertisement have to be that rude? It's the big lie technique. Mere persistence and intensity wears people down and eventually "history" marches forward and whatever is desired becomes the "new normal". It's very easy to make people believe anything, all that is needed is time and consistent effort.

>>94213 >using x11 inside wayland GEG... waylandcux what is this? heralded as a glorious replacement yet still reliant on x11 building this functionality into wayland is beyond retarded, beyond inefficient. >wayland is le heckin' better because it's so heckin' minim-AAAACK x11 is fundamentally broken, but wayland is even worse need a new display protocol where the server is the literal machine you are interacting with, and the client is where you are interacting from, not the illogic of treating windows as clients have the server specify to the client content with parameters to render in some way, e.g. draw this n dimensional vector or this shader using the vector's area as a mask, text, etc., then have the client render it with fallbacks. can be described better but general idea is present

>>94223 Thanks for the explanation. The invocation is not what I was hoping for, but hyprland logic should be fine. >>94226 I guess it makes sense. Though from this point of view these messages look even more repulsive.

>>94218 >>94216 >Oh dear... Why does any Wayland advertisement have to be that rude? You know that crapping on others do not give your argument any points, right? If you truly believe in something, than point out what's good about it. You're pretty much stuck with it regardless of how you feel, technological change is rude and upsetting by default, there's no legal contract with you on the X to Wayland transition that says the occurrence won't offend. I'm sure standalone X will exist for some time longer but the stakeholders are thinning out, once the last of the genuine legacy users are gone there isn't really going to be any user base that couldn't just run on Wayland. Any incompatible graphics hardware is over a decade old at this point, it's usually completely proprietary driver wise, and the company who made them is either uncooperative or no longer exists. Rootful XWayland and non-rootful XWayland fills the gap for most legacy users who don't have hardware issues, so any sane organization will just move their commodity hardware over to something compliant anyway when it's time to dump the old hardware, and move all X-only programs to XWayland. Individuals could be fooled by a Wayland kiosk program renamed "X" that runs rootful XWayland with their favorite window manager at this point, mostly the same pack of idiots who want to run a forked X.Org could be fooled if they were not allowed to check what processes were running. Valve has moved the gaming user base to Gamescope which either uses XWayland or Wayland directly, Red Hat has concluded all X.Org work, KDE and GNOME will discontinue X11 support completely, all major distros are working on moving away completely at some point as major package maintenance costs time and money, Wine already has Wayland support. No sane organization wants to prop up X.Org for long and nobody has come up with a realistic Wayland replacement either (whoever wants to has had like 10 or 15 years to do it, whoever these hypothetical legendary programmers are they sure take their time). This isn't an advertisement, it's more of an obituary or epitaph for X.Org as the primary display server for Unix-likes. >On the topic of "functionality gaps", I'm a bit ootl: are there any good solutions for global hotkeys? Ideally without having to rewrite the whole application with some specific Wayland-only library. The XDG Global Shortcuts Portal is probably what will be supported in the end. There are some compositors that force global shortcuts themselves like Hyprland. >>94228 >yet still reliant on x11 It's more like X11 would be relying on Wayland to operate in this case. You don't need X11 to run Wayland and can either not install or remove XWayland completely if you want. >building this functionality into wayland is beyond retarded, beyond inefficient. It's not really built into Wayland, it's basically a X server without the X drivers that translates X stuff to Wayland. These days it can be controlling the root window usually in a minimal Wayland kiosk server or more commonly a nonrootful server that speaks to X clients if you have any. >need a new display protocol where the server is the literal machine you are interacting with, and the client is where you are interacting from, not the illogic of treating windows as clients >have the server specify to the client content with parameters to render in some way, e.g. draw this n dimensional vector or this shader using the vector's area as a mask, text, etc., then have the client render it with fallbacks. can be described better but general idea is present What do you actually hope to get from this design? I don't see the point.

Wayland enjoyers have been listening to retarded X sperging for years. Get over it.

I'm hoping Wayland putting so much of the burden on compositors, will funnel more resources into KDE. Would rather have 1 good option than 20 mediocre ones.

>>94246 What? All current most popular tiling WMs are wayland-based and use something like wlroots underneath.

>>94242 >You're pretty much stuck with it It's all in the use case. A lot of stuff could be done with a plain framebuffer, and you're burying good ol' X11. Also, Linux is not the only system. >regardless of how you feel To be honest, I don't care much. Tried Wayland once, did not work for me. I might try again later. >technological change is rude and upsetting by default I've seen a lot of technological changes, that were praised and cherished. Whole industries changing course and a lot of happy people, both consumers and workers, that stuff. So I do not really go along with you on that. >there's no legal contract with you on the X to Wayland transition that says the occurrence won't offend. Quite discourteous of you to change the topic of my question. I've lamented on the tone of these posts, not on being offended by the technological changes. It's just tiring to see people throwing feces. The fact of perceived inevitability does not give one the right to behave this way. Share what you fancy, not crap on the others. All in all, thank you for the thorough response.

>>63336 i am keep using xfce! its perfect expect the dpi scaling

>>94280 I am using windows 2000.