Wayland's Current Release Manager Is Stepping Down, Following Samsung's Open-Source Drama

[anarki2]

Here's to the delusional peeps who still actually believe open source software is written by enthusiasts just for the sake of being enthusiastic (and also that money grows on trees).

[johanb]

Hardly. It's been stabilized on life support. People only touch it when they absolutely have to, which is the complete opposite of well-maintained. Few people understand the entire thing, and no one goes into it unless it's to fix a specific issue. It's effectively bit-rotting. All the major X developers are Wayland developers, and they'll be quite glad when it finally gets EOLed.

I agree with all of your points except for this. During the time Wayland was developed X has become very stable and a mature code-base is a good thing. Sure the code base is quirky, but for the end-user it has been very reliable the past 8 years when Wayland has been under development. I want to switch to Wayland for the features and simplified implementation, but no longer because of reliability.

[Terrablit]

I agree with all of your points except for this. During the time Wayland was developed X has become very stable and a mature code-base is a good thing. Sure the code base is quirky, but for the end-user it has been very reliable the past 8 years when Wayland has been under development. I want to switch to Wayland for the features and simplified implementation, but no longer because of reliability.

Yeah, I get that. It's definitely a mature codebase, and is relatively stable. But saying that it's stable isn't the same thing as saying that it's well-maintained. Well-maintained codebases have multiple people working on them. Now that nobody wants to work on X, we're either saying that X has peaked and there's nothing more that it could do (obviously not the case), or that everyone believes that the cost of improving it isn't worth the return.

The other problem is that, regardless of how stable something is, unless everything around it is also stable and static, it will eventually destabilize. GPUs and graphics are one of the faster-moving hardware targets. 3D applications working with X already have so many issues with screen tearing, performance and presentation in general that further innovation is just going to get more difficult. Vulkan on the other hand has an interface that allows it to integrate with windowing systems like Wayland cleanly.

Bit rot can refer to actual data rot on hardware storage, but it can also refer to software rot. Software rot is a known phenomenon. Just because we don't touch code doesn't mean it will keep working. If you've ever maintained software over the course of a few years, you'll find several cases where code starts breaking in untouched areas due to changes in surrounding code.

A lot of rules about code in the kernel (and other projects) are based around software rot and the burden of code maintenance. It's why some features get dropped or refused. It's why kernel devs require the code to meet kernel guidelines. It's why existing, unmaintained features are dropped instead of ifdef'ed. Software rot is a huge factor in why we can't just leave old code alone. People need to understand code that continues to run, and in many cases they need to refactor and rebuild it with the rest of the codebase over time. Otherwise you end up with hideous cruft and tedious build processes like they have in the Oracle database server (https://news.ycombinator.com/item?id=18442941).

[Zucca]

Honestly a bit shocked by the venom of this post.

Don't worry. debianxfce has a good record on starting sensless arguments here. Just don't feed and you'll be ok.

[microcode]

Maybe related.

https://news.ycombinator.com/item?id=19588852

[Weasel]

Firstly, pretty sure the Wayland project sparked from disgruntled Xorg devs, and I recall hearing the entire Xorg team is on the Wayland project. Isn't Xorg infamous for having a "abandon all hope ye who enter here" comment in it's source?

It doesn't matter how bad Xorg is because Wayland is just unusably crippled by design so there's no hope for that trash, except for some piece of shit mobile OS.

[Azrael5]

Honestly a bit shocked by the venom of this post.

Firstly, pretty sure the Wayland project sparked from disgruntled Xorg devs, and I recall hearing the entire Xorg team is on the Wayland project. Isn't Xorg infamous for having a "abandon all hope ye who enter here" comment in it's source?

I'm also a bit shocked at how anti-choice you sound. Don't you usually post around here saying how much you dislike Systemd? Well I don't like Xorg. Why don't you use another init system and I'll use another display protocol and we can both wish each other the best instead of insisting our respective choice of software perish? Or, if you believe in not having alternatives that strongly, you can just use Windows or macOS where everything is hard-preinstalled for you with little in the way of alternatives. Not saying it's a bad thing, just that you seem to prefer nobody change their installation from what yours is, and the previously mentioned operating systems are the best way to achieve that.

consider idiot the writer of that post so you fix your shock immediately.

[Azrael5]

The real problem with wayland concerns with the slowness which characterizes its system integration. I hope that Intel will implement it on their own Clear Linux operating system and I will be glad to return on their products to take benefit of linux.

[johanb]

Wayland, gnome3, kde are a religions, no technical reasons to use them. So it useless write technical facts for religious people. They want burn miscreants at the stake.

Technical reasons to use Wayland:
- Simplified design
- Clearly defined protocol so there can be multiple implementations
- Applications isolated by default
- Pixel-perfect by design (I even have less tearing with only wayland than with X+Freesync)

[wagaf]

Wayland is actually used and maintained by many industrials like automakers, planemakers etc. They have been the first users of Wayland and will keep using it as it's the best solution for their purposes.

Don't forget that desktop is not the primary use for GNU/Linux systems.
Most connected embedded systems are GNU/Linux-based and have been for a while now.

So I don't worry at all about the future of Wayland.

There is much less money to develop desktop use-cases for Wayland so these features are coming much slower but are mostly there now.