GNOME 3.29.2 Released As The Second Step Towards GNOME 3.30

[tildearrow]

Wayland... ...Xorg

I want to ask. Why does everyone compare Wayland to X.Org? X.Org isn't the only X11 server on Earth...

Wayland doesn't need hacks to ensure correct display (no tearing, lag's and flickering) which Xorg needs.

Wayland compositors with tearing can exist.

[dragon321]

Wayland is going to be a freaking mess for the next decade at least, while providing no tangible benefit to end users.

Let me introduce something to You. It's called "client isolation". That's right, X isn't capable of it, Wayland is. It's very important thing in security, and X can't do it. Well, I say more - Windows and macOS can do it as well. What about different DPI for each screen? Another thing that Xorg cannot handle properly. Xorg isn't suitable for 21st century, that's true.

"NIH" when Wayland is made by a lot of Xorg peoples? Lol. Wayland exist not because "NIH", but because of Xorg limitations which you can't solve without rewriting from scratch (which Wayland does). Wayland doesn't need hacks to ensure correct display (no tearing, lag's and flickering) which Xorg needs.

[kaprikawn]

Doesn't matter what it is considered to be. Wayland is going to be a freaking mess for the next decade at least, while providing no tangible benefit to end users. It is a disaster. They might as well just put the effort into redesigning X. The major issues Wayland was supposed to solve were solved by DRI3, Wayland has no reason to exist now other than the NIH syndrome. Any additional issues with X could be solved procedurally over the years instead of just abandoning for a retarded solution that still isn't production ready after a decade and will probably never be.

What a tool.

"They might as well just put the effort into redesigning X"

Wayland is a redesign of X, moron

"Wayland has no reason to exist now other than the NIH"

Wayland was started by X devs and is administered under the same umbrella as X i.e. Wayland and X are from the same place and people, so how can it be 'not invented here' syndrome?

[cybertraveler]

I didn't claim Wayland was finished and practical for every day use right now. It's not.

Regarding how long Wayland and the software ecosystem has been in development and whether it has been too long. What is too long? Wayland is a massive change to the GNU/Linux desktop ecosystem. How long should we expect to wait for this change to be user-ready? I don't know. What I do know is that the Wayland effort has come a very long way since a decade ago. I can install Ubuntu and enable the Wayland compositor and it's a nice experience. It's not ready yet, but it's in good shape. There are huge numbers of people that want the improvements that Wayland is designed to bring and there are huge numbers of people working on writing the Wayland code to make those improvements a reality.

Wayland will come. When it's ready I'm confident I wont ever want to go back to Xorg. I want a faster, less clunky and more secure desktop and Wayland is the absolute best and most likely effort to bring me that.



And regarding your comment Templar: nice argument. Quality. /s

[TemplarGR]

Wayland has been hugely successful. There are huge numbers of programs, libraries, toolkits that have adopted it. There are multiple Wayland server implementations and the work on making Wayland viable has never once ceased: the pace of improvement to Wayland is only increasing.

There are huge advantages to using Wayland. Some of them being:

- It does not have to be backwards compatible with decades worth of X apps. This means they can create a clean, simple design (not a mess as you described it). This makes implementing Wayland easier. This means you are less likely to get bugs and less likely to get security issues than if you were to simply forever build ontop of X.

- Wayland provider far better support for isolation of clients. This has practical security benefits including: your browser app can't read the keystrokes going to your text editor app or your password manager app. This has not been practically implemented on X in a way that most X users can benefit from. It might be because practically implementing it on X could require a large redesign of X. What is Wayland, if not: a large redesign of X. There are Xorg developers working on Wayland to make it everything they wish Xorg could be.

- Wayland is designed to work well with real, modern hardware. It's not focused on supporting outdated hardware designs or hypothetical future designs which may never be built because they are impractical. This again has the benefit of allowing Wayland to be a more simple design that is very much fit for purpose.

- Practically speaking, X is failing in many areas which should not be a problem in 2018. One example is how X badly handles full screen applications, lock screens and screensavers. Wayland has been specifically designed to make these 3 things just-work and without any hacks or corner-case issues. There's actually a phoronix article on this: https://www.phoronix.com/scan.php?pa...item&px=OTI5MQ

Wayland isn't finished yet, but we are far enough along that all the popular GUI toolkits can natively speak it, lots of the popular desktop environments can serve as Wayland compositors and special efforts have been made or are being made on loads of the most popular programs to make them work well on Wayland (eg Firefox).

I for one am looking forward to running a lighter, more graceful, more secure and potentially faster desktop environment in the future that is built with the Wayland protocols. Wayland wont be perfect but it will be better than X and will make a GNU/Linux desktop more competitive with a Mac OS X and Windows desktop.

I don't know where you are living currently but it is not planet earth.

[Brisse]

In practice, the mentioned fields by you work fine on Xorg, while they usually don't with the existing Wayland compositors. I don't see any groundbreaking progress with Plasma or Gnome on Wayland over the past year, it's still astronomically far from meeting my expectations.

I have to agree with this from a users perspective. I think cybertraveler did well to point out why Wayland is needed, but the current implementations are, despite a decade of development, very immature and every time I receive an update related to desktop compositing and Wayland I more often see things being more and more broken rather than fixed.

[aufkrawall]

Practically speaking, X is failing in many areas which should not be a problem in 2018. One example is how X badly handles full screen applications, lock screens and screensavers. Wayland has been specifically designed to make these 3 things just-work and without any hacks or corner-case issues. There's actually a phoronix article on this: https://www.phoronix.com/scan.php?pa...item&px=OTI5MQ

In practice, the mentioned fields by you work fine on Xorg, while they usually don't with the existing Wayland compositors. I don't see any groundbreaking progress with Plasma or Gnome on Wayland over the past year, it's still astronomically far from meeting my expectations.

[cybertraveler]

Doesn't matter what it is considered to be. Wayland is going to be a freaking mess for the next decade at least, while providing no tangible benefit to end users. It is a disaster. They might as well just put the effort into redesigning X. The major issues Wayland was supposed to solve were solved by DRI3, Wayland has no reason to exist now other than the NIH syndrome. Any additional issues with X could be solved procedurally over the years instead of just abandoning for a retarded solution that still isn't production ready after a decade and will probably never be.

Wayland has been hugely successful. There are huge numbers of programs, libraries, toolkits that have adopted it. There are multiple Wayland server implementations and the work on making Wayland viable has never once ceased: the pace of improvement to Wayland is only increasing.

There are huge advantages to using Wayland. Some of them being:

- It does not have to be backwards compatible with decades worth of X apps. This means they can create a clean, simple design (not a mess as you described it). This makes implementing Wayland easier. This means you are less likely to get bugs and less likely to get security issues than if you were to simply forever build ontop of X.

- Wayland provider far better support for isolation of clients. This has practical security benefits including: your browser app can't read the keystrokes going to your text editor app or your password manager app. This has not been practically implemented on X in a way that most X users can benefit from. It might be because practically implementing it on X could require a large redesign of X. What is Wayland, if not: a large redesign of X. There are Xorg developers working on Wayland to make it everything they wish Xorg could be.

- Wayland is designed to work well with real, modern hardware. It's not focused on supporting outdated hardware designs or hypothetical future designs which may never be built because they are impractical. This again has the benefit of allowing Wayland to be a more simple design that is very much fit for purpose.

- Practically speaking, X is failing in many areas which should not be a problem in 2018. One example is how X badly handles full screen applications, lock screens and screensavers. Wayland has been specifically designed to make these 3 things just-work and without any hacks or corner-case issues. There's actually a phoronix article on this: https://www.phoronix.com/scan.php?pa...item&px=OTI5MQ

Wayland isn't finished yet, but we are far enough along that all the popular GUI toolkits can natively speak it, lots of the popular desktop environments can serve as Wayland compositors and special efforts have been made or are being made on loads of the most popular programs to make them work well on Wayland (eg Firefox).

I for one am looking forward to running a lighter, more graceful, more secure and potentially faster desktop environment in the future that is built with the Wayland protocols. Wayland wont be perfect but it will be better than X and will make a GNU/Linux desktop more competitive with a Mac OS X and Windows desktop.

[TemplarGR]

Still no automatically adjusted fps/refreshrate with Mutter on Wayland?
Is a Wayland session still considered to be a technical preview, or what would be the explanation?

Doesn't matter what it is considered to be. Wayland is going to be a freaking mess for the next decade at least, while providing no tangible benefit to end users. It is a disaster. They might as well just put the effort into redesigning X. The major issues Wayland was supposed to solve were solved by DRI3, Wayland has no reason to exist now other than the NIH syndrome. Any additional issues with X could be solved procedurally over the years instead of just abandoning for a retarded solution that still isn't production ready after a decade and will probably never be.

[Brisse]

Wayland session still breaks quite often on my end. (In fact it's currently unusable for some reason, seemingly since I updated Mesa at one point…)

Yea, it broke for me recently on Debian Sid when I received GNOME 3.28.1 but then I added amdgpu.dc=1 to my kernel boot parameters and it started working again. It's completely broken without that boot parameter though.