SDL3 Begins Dumping A Lot Of Old Code: GLES1, OS/2, DirectFB, WinRT, NaCl & More

[s_j_newbury]

with Wine the numbering didnt really make any difference. Windows is also a moving target so this year they may have great support for apps, but next year if Microsoft release a new API, the cycle starts again and "wine is terrible" again until they add the necessary API and fix any bugs that are found.

Previously there was a time aroudn 2005 where it seemed like wine had actually arrived and only a few bugs remained for the major software targets. but things changed a lot since then.

Also, around 1995, Windows1-3 support was pretty good, but then things changed a lot! :-)

[TemplarGR]

They can not do this. The software that is build on top of SDL might still have X11 dependencies and will need to go the xwayland way because of that. This can also be true in 10 years from now on. So X11 is still very important to have in SDL. Qt and GTK need to do the first step by deprecating X11 support, so it's known that it will be removed withing the next major version.

For pulseaudio, I'm with you. I dont see pulseaudio to be around in 2 years anymore.

I am not familiar with SDL, but i am not sure this makes sense. SDL3 brakes backwards compatibility anyway. Current software, or evolutions of it, will still probably be using SDL2. Any new software in the future will be using SDL3 from scratch. Sure, even then, there will be software that has some old codepaths etc, but it is going to take some effort to convert to using SDL3 anyway, might as well just remove X11 and pulseaudio remnants as well.

[TemplarGR]

PulseAudio maybe, X11 is not an option.

X11 is actually more of an option than Pulseaudio...

I don't understand why people confuse certain things in the software world. SDL3, is a new library. Games/software made for it, won't run on SDL2, and SDL2 games won't run on SDL3. It breaks backwards compatibility. Thus, we need to have it released, before software gets made for it. By the time SDL3 and software that needs it are out and stable, X11 will be long gone from the Linux desktop. It already is for most people. Aside from a few chronic complainers who refuse to adopt to change, the rest of the Linux world just uses whatever is default in their distro, and most distros default to Wayland these days...

In other words, X11 is essentially EOL. It is on its last legs, getting minor fixes and no new development, for years... It makes no sense to start coding a new library for future software, breaking compatibility, and maintain support for a legacy display manager. No sense at all. SDL1/2 won't be going anywhere for those who want to keep using the stuff they are using now with X11.

[brad0]

X11 is actually more of an option than Pulseaudio...

I don't understand why people confuse certain things in the software world. SDL3, is a new library. Games/software made for it, won't run on SDL2, and SDL2 games won't run on SDL3. It breaks backwards compatibility. Thus, we need to have it released, before software gets made for it. By the time SDL3 and software that needs it are out and stable, X11 will be long gone from the Linux desktop. It already is for most people. Aside from a few chronic complainers who refuse to adopt to change, the rest of the Linux world just uses whatever is default in their distro, and most distros default to Wayland these days...

In other words, X11 is essentially EOL. It is on its last legs, getting minor fixes and no new development, for years... It makes no sense to start coding a new library for future software, breaking compatibility, and maintain support for a legacy display manager. No sense at all. SDL1/2 won't be going anywhere for those who want to keep using the stuff they are using now with X11.

So clueless.

[timofonic]

So clueless.

I agree. He made a few good points accidentally, but that's all.

It's true XOrg/X11 is dying. Wayland is the future, but still not the present in all scenarios.

Wayland is made as a series of protocols and easily extensible, that's very good. But it's taking too much time to make certain protocols and I disagree how extremely slow they are at moving them to Wayland Core. They still have an XOrg mindset instead a more Vulkan-like one, IMHO.

So, I expect 4-7 years until massive Wayland adoption. It needs to have not only feature parity with XOrg, but even more features usually done by tons of hacks and dirty userland taking advantage of XOrg extremely patched up design.

I also consider and extremely big fail the lack of an official project similar to wlroots, but with x1000 more manpower to ease Wayland adoption. Despite all that nonsense about Wayland being a protocol and blablabla, I think those are lame excuses done by stubborn developers and zealots.

And despite being extremely experimental, I also consider Arcan metaproject to be conceptually quite superior to both XOrg and Wayland in many aspects. Unfortunately, I think it will never leave to be a design concept in code form.

To me, Wayland is a good idea but failed execution. If the project doesn't make a considerable management change and improve many Core stuff, it will be another case similar to PulseAudio and PipeWire.

I'm not the best developer of the world, but I consider project management part sometimes may be even more important than the coding part to avoid stupid design decisions in a software project.

[Quackdoc]

If they really want to remove jack support they should remove pulseaudio support first: pipewire is in a better shape to replace pulseaudio than jack yet.

IMO you cant really remove pulse without the jack as it isn't a very useful daily audio server, so people with PC's who do both roles and need both is not uncommon, and they would be primairly impacted. and pipewire isn't always a good replacement for jack yet.

[Quackdoc]

Last I heard PipeWire can replace JACK (but not sure if as a drop-in solution) so I wonder why it was reverted.

it is a drop in solution, "for some users" some users are experiencing breaking bugs with pipewire that break the whole workflow, so many users are stuck on jack and/or pulse

[abott]

That is of course the goal but there has been a lot of bugs and issues in the SDL Wayland backend to work out to get it on par with X11. This has required changes in other places also, like the creation of libdecor and some Wayland protocol additions (afaik), and so on.

It shouldn't. There's not nearly as much cruft as X11. X11 is the hack. Wayland should be a fraction of the complexity and also provide MORE features.

Some of these concerns are hilarious and just so insane that it shows not a ton of people actually understand what they're commenting here, rofl.

[david-nk]

Reading this thread is unreal. Not just one misguided person, but a whole bunch of people acting as if the last distro using X11 was released sometime 15 years ago... except no, X11 is still universally used everywhere. The mere idea of dropping X11 support anywhere in the next decade is ridiculous.

[uxmkt]

1.0 doesn't matter; it's just a version number. Remember how long it took Wine to reach 1.0, despite being in good shape?

Time for some jesting! https://0ver.org/