The death watch for the X Window System has probably started • Hacker News Board

Re: The death watch for the X Window System has probably started

#291

Post by traverseda » Sun, Jul 07, 2019, 9:32 PM UTC

acdha wrote: post #190

traverseda wrote: post #118

Earlier quoted context omitted.

>PRs and public ownership of future project direction. https://en.wikipedia.org/wiki/The_Cathedral_and_the_Bazaar That's kind of the point of open-source, public ownership. Sure, you need someone to be in charge and reject outright bad ideas, but when you're refusing to add options that support other people's personal preferences that's when it's starting to become a problem. It's all these authoritarians that think…

Do you have examples of working pull-requests which were rejected for what you consider bad reasons? I’m sympathetic to maintainers choosing not to take on additional technical debt so I’d especially be looking for things like willingness to shoulder long-term support costs and other constraints imposed by accepting a PR.

One thing that comes to mind is gnome's refusal to implement type-ahead in nautilus and the gtk3 file picker. It's a very popular and visible issue that I won't get into here, but suffice to say that there are a lot of people who would really like a more traditional type-ahead.

There aren't any specific pull requests I'm aware of although there are forked versions available.

When asked about it the developers involved just say "no", and any attempt to work on a compromise or discuss what would be necessary to include a flag somewhere deep in gcnonf to revert the behavior is met with a resounding "we're not interested in that feature". Which is fine, but don't keep other people from implementing a feature that they like just because it doesn't fit in with your vision.

That attitude seems pretty endemic to open-source projects involving a lot of (ex) redhat engineers. It's not that the PR is being refused, it's that it's impossible to even have the discussion.

Re: The death watch for the X Window System has probably started

#292

Post by ekc » Sun, Jul 07, 2019, 9:33 PM UTC

llukas wrote: post #278

boudin wrote: post #246

Earlier quoted context omitted.

Intel and AMD cares about it. If the Linux desktop market would grow, Nvidia would quickly change their mind. Now, this market growing is another question... But for me, nvidia gpu are just not a possibility because they lack good drivers on Linux.

So how do you want to grow Linux market share if you say FU to a significant part of the potential market? Steam hardware survey is pretty clear about AMD:Nvidia GPU ratio.

(For gamers.)

Re: The death watch for the X Window System has probably started

#293

Post by ubercow13 » Sun, Jul 07, 2019, 9:33 PM UTC

logicprog wrote: post #288

ubercow13 wrote: post #265

Earlier quoted context omitted.

With pulse, wasn't the main complaint that it was buggy for too long after being pushed on users? Seems similar to Wayland (the ecosystem) in that regard

I haven't heard of Wayland being buggy?

For me, at least with sway, it is still too buggy, I just switched back

Re: The death watch for the X Window System has probably started

#294

Post by ekc » Sun, Jul 07, 2019, 9:33 PM UTC

Keyframe wrote: post #256

imtringued wrote: post #252

Earlier quoted context omitted.

It is very condescending to label Intel, AMD and pretty much every other GPU vendor other than Nvidia as nothing.

Discrete GPU market share is 80%+ Nvidia, last time I checked.

Discrete GPUs are a write-off of a minority of the total market, last I checked.

Re: The death watch for the X Window System has probably started

#295

Post by csande17 » Sun, Jul 07, 2019, 9:37 PM UTC

ddevault wrote: post #273

calcifer wrote: post #266

Earlier quoted context omitted.

> Their alternative has genuine technical problems that would render large parts of wlroots broken. If this is indeed true (and I have no reason to doubt you) then I wish you and the rest of the Sway folks would focus on those issues, instead of the political arguments because from the outside it all looks very petty when you write stuff like "Nvidia doesn't support Sway". Nvidia doesn't even know about you.

I've personally spoken at length with Nvidia graphics driver developers. They know who we are and they know the technical problems. If you have no reason to doubt me, then perhaps you shouldn't.

Are the technical problems documented anywhere where non-NVIDIA-driver-developers can learn about them?

Most of the discussion I've read online argues that to support NVIDIA's solution, everyone would need to add if(nvidia) checks to their code and maintain them forever. But this argument only really holds in Wayland's bizarro-world where all programs that arrange windows must also bake in code that interfaces directly with hardware to draw them on the screen.

Absent any other technical problems, it's hard not to conclude that this is just an attempt to use politics to cover up one of Wayland's poor design choices.

Re: The death watch for the X Window System has probably started

#296

Post by neilv » Sun, Jul 07, 2019, 9:38 PM UTC

segmondy wrote: post #220

neilv wrote: post #95

Earlier quoted context omitted.

Yes, a lot of open source goodness is due to Red Hat, and a lot of open source programmers' livelihoods are and have been due to Red Hat. Red Hat earned a lot of goodwill among early Linux people, and fortunately they were able to sustain their work with enterprise business revenue flow, of actual dollars (not just goodwill). It's also true that a lot of people have been implicitly delegating a lot of foundation on w…

There's no Red Hat. They got by IBM for $34 billion, I keep seeing a lot of reference to Red Hat in this post. Please call it, what it is. IBM. It's all about money. Much can be said about this, but money has really corrupted and polluted OSS.

Of course the possibility of Red Hat being ruined has been on everyone's minds since the acquisition.

But things akin to the Wayland push happened before that, so I wouldn't attribute that to post-acquisition.

I won't call it IBM, so long as Red Hat retains some autonomy to exercise their know-how, and distinct identity (not kill some of the golden goose for which they paid serious gold), and so long as most of the the most skilled Red Hat people are happy and stay. (If they tried to commoditize everyone as worker drones, that would be unfortunate.)

Also, FWIW, recall that IBM have for many years been a strong supporter of various open source. (Maybe too bad they didn't buy Java/Sun in the end.) For money reasons, of course, but there seemed to be a lot of goal alignment among open source people, and overlapping with libre people.

Re: The death watch for the X Window System has probably started

#297

Post by ufo » Sun, Jul 07, 2019, 9:42 PM UTC

themacguffinman wrote: post #270

shmerl wrote: post #242

Earlier quoted context omitted.

I recommend you simply to switch to AMD if you are using Linux and want a high end GPU. That's the only way for you to get a usable desktop with modern features. Or wait until Intel will release a new GPU next year, it should work well with upstream too. Nvidia has no interest in making their cards play well with Linux, so just avoid them completely to keep your sanity and don't waste your time trying to work around…

> I recommend you simply to switch to AMD if you are using Linux and want a high end GPU. That's the only way for you to get a usable desktop with modern features. This is just not true. Gnome with Wayland is one of the most popular and modern DEs and it supports Nvidia's proprietary drivers today.

Nvidia still doesn't work as well as other gpus under wayland, which is why GbOME still uses X by default for NVIDIA

Re: The death watch for the X Window System has probably started

#298

Post by traverseda » Sun, Jul 07, 2019, 9:44 PM UTC

0815test wrote: post #194

traverseda wrote: post #118

Earlier quoted context omitted.

>PRs and public ownership of future project direction. https://en.wikipedia.org/wiki/The_Cathedral_and_the_Bazaar That's kind of the point of open-source, public ownership. Sure, you need someone to be in charge and reject outright bad ideas, but when you're refusing to add options that support other people's personal preferences that's when it's starting to become a problem. It's all these authoritarians that think…

If you don't like the current maintainers' policy on "future project direction", you can always fork the project. Yes, this involves an increased maintenance burden since you need to actively review what the upstream is doing and figure out how to cleanly pull their changes - but guess what, that's the very same burden that dealing with your proposed changes would have placed on the upstream themselves! "Authoritaria…

There are ways to architect software so that it can be cooperatively developed, so that you can fork and maintain an individual component instead of an entire stack.

Re: The death watch for the X Window System has probably started

#299

Post by noobermin » Sun, Jul 07, 2019, 9:44 PM UTC

csande17 wrote: post #79

I would use Wayland, but the most actively-developed tiling window manager, sway, doesn't work with the NVIDIA graphics card I own. And come to think of it, isn't that a really weird problem to have? On X, the compositor (responsible for actually drawing all the windows on the screen) and the window manager (responsible for deciding how to arrange the windows and what their title bars/borders should look like) can be…

I made a comment on this below, X already exists, why can't people live with X/fix X and thus utilize the already existing code vs recreate all the interfaces that X already has?

Re: The death watch for the X Window System has probably started

#300

Post by quotemstr » Sun, Jul 07, 2019, 9:45 PM UTC

robert_foss wrote: post #201

noobermin wrote: post #146

I seek to be enlightened, not to flame: what exactly is so wrong with X that it can be retrofit to work better? Wayland is taking a while, which is expected sure, but for now, X works fine although with issues.

The design assumptions that went into making X were made before the modern computer or Linux existed. Modern X is a Frankensteinian collection fixes upon fixes and unmaintained code.

You didn't actually answer the OP's question. There is nothing Wayland does that could not have been done via the X11 extension mechanism. X has already been reinvented several times --- few things use the core protocol nowadays, after all! For example, XRender for text is now universal.

The thing that's different about Wayland is that for reasons that nobody ever explains clearly at a technical level, this time, a few influential people created a new and (IMHO, inferior) protocol instead of continuing to use the X extension mechanism.

Security? Isolate the X session data to individual client connections. Buffer management? Do it like GLX does. Want server-side decorations for some reason? What's stopping you? Sure, ICCCM isn't trivial, but it exists, and it's flexible.