SDL3 Will Keep Wayland Default At Least For The Time Being

[spicfoo]

It is dumb conspiracy, I don't believe in that at all. But when these people have so much power in a critical project, and they use it to stall or block important protocols, for sometimes half a decade, it becomes a question if these people, very knowledgeable or not, are beneficial to the community as a whole.

We don't need the perfect protocols, we need something that works right now, it has been 15 fucking years already!

People have power in open source projects only because they are knowledgeable and willing to work on defining the solution and writing the code. If you want to contribute, you are welcome to. Desktop Linux severely lags in funding and things are going to take time as a result. If you want to fix that, give people working on it more money or contribute time to test or write code or provide feedback. Do something more concrete instead of dumb conspiracies or firing people. That's madness.

[scottishduck]

Worth remembering that Xorg server will be fully removed from RHEL10 in 2025 so there’s a fairly strong push from a number of corporate sides to get these wayland protocols sorted quickly.

[Volta]

Didn't SDL developer know in advance that Wayland lacked the things he was talking about? If he knew, did he report it? If so, did anyone care?

[ssokolow]

People have power in open source projects only because they are knowledgeable and willing to work on defining the solution and writing the code. If you want to contribute, you are welcome to. Desktop Linux severely lags in funding and things are going to take time as a result. If you want to fix that, give people working on it more money or contribute time to test or write code or provide feedback. Do something more concrete instead of dumb conspiracies or firing people. That's madness.

I think they're referring to the "Protocols in the "xdg" and "wp" namespace are ineligible for inclusion if NACKed by any member." clause in the Wayland repo governance document.

[TemplarGR]

I expect good points to be made from both sides in a logical and calm manner without any name calling or gas lighting whatsoever. The dialog here will be productive, and the underlying problems are going to be solved with a nice round of civil discussions on the issues and possible solutions.


...Right?

Of course. That's why i came here to comment to all the X11 fanbois that they are n00bs and clueless, X11 sucks, it is garbage and no one should use it, and if you are not using Wayland your underwear smell funny! Love civilized discussions!

[CommunityMember]

Worth remembering that Xorg server will be fully removed from RHEL10 in 2025 so there’s a fairly strong push from a number of corporate sides to get these wayland protocols sorted quickly.

Enterprise Linux distributions focus their resources on where the money is, and that is server(s), not desktops.

All the Enterprise distributions are looking to remove packages which cause them ongoing maintenance issues/costs which have limited value to their paying customer base, and that includes native Xorg support (XWayland is apparently considered "good enough" for those paying enterprise customers who run graphical apps on their servers).

[V1tol]

I think people here just don't understand what it means to work on a large company. I do work so I will explain. Bureaucracy.

Imagine you have a problem in your software, like a bug. In 15 minutes after it was discovered you know how to fix it, fix will take 30 minutes + 30 minutes on green\blue or canary deployment. But the whole operation takes 10 days. Why? Go create a ticket and ask a management to include it in current sprint changing current capacity. If that was successful and you actually did your job - now you need someone from QA to run full regression test suite because their leadership insist on that. They are busy so it takes 3 days for a QA person to start working on it. Now you need to ask Release Engineers to schedule a time for deployment (I am not joking, we have this crap in our company). Today is Wednesday and they don't have time, so that is scheduled on Monday. And all that is accompanied by numerous Jira tickets, tons of Slack messaging and calls consuming precious time.

So I don't think the "they stall protocol adoption" theory to be a conspiracy. But I think they definitely don't do it intentionally - that's just how the system works in their case. HDR, Wayland, VSync, Tearing, low-latency or whatever fancy thing is really not a priority for commercial systems. And as a humble developer you can't just kick open a door to your boss and scream in his face "lets do f**king deployment now" when you are on a salary.

[mrg666]

The way the commit starts "myriad of unresolved problems" was the indicator of this outcome. I love open source that they shut this commit down. Next time they submit the merge request with fixes instead of reversions. The way is forward not backward.

[dlq84]

It is dumb conspiracy, I don't believe in that at all. But when these people have so much power in a critical project, and they use it to stall or block important protocols, for sometimes half a decade, it becomes a question if these people, very knowledgeable or not, are beneficial to the community as a whole.

We don't need the perfect protocols, we need something that works right now, it has been 15 fucking years already!

If don't think you understand open source if you think they somehow hold Wayland hostage... If the community didn't like the way it's handled it would have been forked ages ago.

[tesfabpel]

It seems with this new comment they're now aiming to check at runtime whether the protocols fifo-v1 and commit-timing-v1 are present to automatically use Wayland as the backend.

https://github.com/libsdl-org/SDL/pull/9345#issuecomment-2023413737

Since the performance issues motivating this PR are resolved by the combination of fifo-v1 and commit-timing-v1, let's check for those protocols before using Wayland by default.

@Joshua-Ashton, I'll go ahead and close this PR. Can you open a new one implementing that?

For those following along, this change guarantees that high performance Vulkan applications like games have the best experience by default. Hopefully these protocols will land soon, but either way the user has final control over the backend being used, via the SDL_VIDEO_DRIVER environment variable.