We Have Mir & Wayland, But There Still Could Be X12

[Remote User]

If you honestly don't know why Wayland was invented, I suggest you do some research.

I didn't say that I didn't know that. That's just you putting words in my mouth. It must be said that the people behind Wayland have this to day about why they developed it, though; "Wayland is intended as a simpler replacement for X, easier to develop and maintain." They also are forthright in declaring that Wayland is not network transparent. Well, I'm not an X developer or maintainer, so I have no reason to switch to Wayland. I also absolutely require and depend on network transparency, so Wayland is useless to me. It not only doesn't solve any problems I have - it also destroys what works for me. X works perfectly for what I do with touchscreen GUIs at customer locations all around the world. I will never understand why some of you just can't deal with the fact that X makes a lot of things not only possible, but also easy, which would otherwise simply not be possible.

[dee.]

You mean The reasons for Wayland's existence, its need, timeliness and transition map from X have been more than exaggerated here,the freedesktop site, and phoronix.com where we make analogous comments that wayland will be some 800 times faster/better than X .

Was bo$$ mad at you for stealing his comedy routine?

[ryao]

Actually, it's quite the opposite. Quartz intended to implement lots of things, at least that's the first thing the article says: they would have needed to do a lot of extensions to the protocol. I don't know about Mir, but Wayland actually doesn't want to do a lot of new things, but to take a lot of the cruft out of the way. Then, there are some extra features, but the motivation to make it outside of X11 is not because they would need to make a lot of extensions, but because they need to remove features they find obsolete that are part of the core protocol, and removing them from X11 would mean breaking almost every legacy app, and maybe even some non-legacy, too.

The article does say that Apple would have had to do many extensions to Xorg. However, it also says that Xorg has since implemented the extensions Apple would have needed.

Anyway, "they need to remove features" is neither a valid reason for nor a benefit from changes to production software. The concept of "implementing brand-new features instead of improving old ones" is a red flag that something is wrong and is held to be one of the things wrong with Windows development:

http://blog.zorinaq.com/?e=74

Wayland and Mir are under development for Tizen and Ubuntu respectively. The fact that they each exert full control over the repositories does make them a better fit for their respective operating systems than Xorg, but only because controlling the repositories makes collaboration optional for them. In the short term, they can push through anything they need for their own interests and ignore anything that conflicts (which I have seen firsthand in certain other projects). In the long term, it is quite possible that there will be a fair amount of GUI software that will cease to be useful after these display servers have been replaced because the powers that be decided to reinvent the wheel again to fit them better.

With that said, the only explanation for Wayland that made sense so far was, to paraphrase, "We could have written Xorg extensions, but we did not feel like working with X11 anymore, so we decided to write a replacement server and protocol". I am okay with that, but I am not okay with hype over largely non-existent benefits.

[MartinN]

Then why make an analogy that someone currently using X is comparable to 56k dial up and someone using Wayland as 25 MBit/s broadband? . Sure their will be improvements but you like many others simply exaggerate said performance increases.

What the fuck has Mir got to do with my response ? It isn't a competition,has nothing to do with what we are discussing,and I'm happy for both to exist.

The analogy was to point out how backward X was compared to Wayland, as well as to quash this notion that if something's "ain't broke, don't fix it" which is what the OP was saying, as far as he was concerned, since as he stated, he didn't have a problem w/X that Wayland could solve.

What's your beef about Wayland dude, why all the hate...or maybe not hate, but dismissiveness? That's why I asked if you're a Mir lover.. It didn't really sound to me like you want both to exist, but I could be wrong.

[DDF420]

I haven't heard it will be "800x faster/better than X", but it will be better and faster than X, yes.

Are you a Mir-lover by chance?

Then why make an analogy that someone currently using X is comparable to 56k dial up and someone using Wayland as 25 MBit/s broadband? . Sure their will be improvements but you like many others simply exaggerate said performance increases.

What the fuck has Mir got to do with my response ? It isn't a competition,has nothing to do with what we are discussing,and I'm happy for both to exist.

[mrugiero]

Apple went off on its own direction with the Quartz Compositor. The result is argued to be the equivalent of what we have today with X11:

http://www.informit.com/articles/article.aspx?p=1552774&seqNum=5

The fuss over Wayland and Mir seems to be a repeat of what Apple did and I do not see the point. The argument "we are doing it because we can" is fine, but I would prefer to see it done without the hype. The only way that a different display server could improve things would be if they was already room for improvement and I do not see it with Xorg.

Actually, it's quite the opposite. Quartz intended to implement lots of things, at least that's the first thing the article says: they would have needed to do a lot of extensions to the protocol. I don't know about Mir, but Wayland actually doesn't want to do a lot of new things, but to take a lot of the cruft out of the way. Then, there are some extra features, but the motivation to make it outside of X11 is not because they would need to make a lot of extensions, but because they need to remove features they find obsolete that are part of the core protocol, and removing them from X11 would mean breaking almost every legacy app, and maybe even some non-legacy, too.

[mrugiero]

c) There shouldn't be too many things in the display server (that I cannot demonstrate, so feel free to disagree. but it seems to be the common opinion of most of the X server's devs)

It's not demonstrable actually, but it's considered a good practice to keep things modular. This allows for easier maintenance, and it allows for disabling obsolete features from a given environment. When you have all this things in the core, this becomes bloat for most users, since the apps they run bypass them; when they are within a different module, you only load it if you need it, and you only update it if you need it, and so on. And you can actually drop older things when you get to a better solution, one thing you can not do while maintaining legacy, if such legacy features are in the core. So, you should try to keep the core at minimum.

And, from the user point of view, applications that cannot be ported will have to use X11 anyway, because porting to the new display server or the new toolkit is the same thing. Application that can will work at least as well as before, except that they will not be limited in either the server or the client side system (if someone ports the toolkit to new display servers). So making them separate is again transparent or positive.

Also, if you follow this approach you could build a headless toolkit, that only sends the rendering commands and has no need to run the display server on the local machine. Only the one which does the rendering needs to load the display server of choice, giving a headless terminal a leaner environment.

OK, performance of Xwayland might not be that bad and actually have some positive points ...but there is still a thing that concerns me....

There was her at Phoronix in a article something about a kinda of "road-map" for Wayland and how the devs intended that Wayland evolved in coming...
...What really concerned me was that at some point is was said/shown_in_schematic(don't recall exactly) that Xwayland was a temporary thing and would be dropped...and i really really didn't liked that.

Is this still planned to be done, sooner or later ?
If so, how soon or how late ? 50 years , 20, 10, 5 ?

There was no info about that...and the most weird thing is that i started to think..."OK, so , X11 will enter in maintenance mode, if X11 enters in maintenance mode and will not evolve, Xwayland won't need to evolve either after some more time and could also enter in maintenance mode...so...WHY to REMOVE IT ?!? Sure it wouldn't take that much resources to simply maintain it !?!"

Because the thing is that there might a odd ball app that wants X and works and we still want/need to use it but if there is no more X or Xwayland, we can't.

*That* is my final true concern about all this thing with Xwayland (dunno what are intentions from Mir devs...but even if i knew , with Ubuntu, we never know what they pull of tomorrow )

If you run such a piece of software that is not moving to Wayland or even Mir even if it's said X will be dropped (I never heard such statement, neither for pure X or XWayland, but I don't know everything anyway), chances are it's unmaintained software, so I don't see why you would feel bad for running just another unmaintained piece of software.

[MartinN]

have been more than exaggerated here,the freedesktop site, and phoronix.com where we make analogous comments that wayland will be some 800 times faster/better than X .

I haven't heard it will be "800x faster/better than X", but it will be better and faster than X, yes.

Are you a Mir-lover by chance?

[DDF420]

A non-answer to your answer- what is wrong with 56k PPP connection to the internet that I should use a 25 MBit/s internet connection instead?

The reasons for Wayland's existence, its need, timeliness and transition map from X have been more than elaborated here and on the freedesktop site....

You mean The reasons for Wayland's existence, its need, timeliness and transition map from X have been more than exaggerated here,the freedesktop site, and phoronix.com where we make analogous comments that wayland will be some 800 times faster/better than X .

[MartinN]

X is not just outdated but fundamentally flawed. This has already been observed to be back in the late 80s. The biggest flaw is the lack of proper communication between the WM and the server. The result is incorrect/inconsistent rendering and bad chemistry with the toolkit elements. The second biggest issue is the insanetly high context switching and latency. The cause is the flawed IPC. Some applications (CAD, games) cope very poorly with X and there is too much chattiness and back and forth, especially with Xlib. (XCB fixes some of this). This constant blocking and chatiness has been observed and noted in the Unix Hater's Handbook.

The reason X has been around for so long is because Unix nerds are ususally content to spend their time in CLI and think that GUIs are for pansies. It was Linux's growth outside the server room and into the PC that prompted Red Hat devs to work on fixing the graphics and audio stack (AIGLX, DRI, Wayland, pulseaudio). At the end of the day a server doesn't need a GUI or an audio stack. Solaris and BSD are doing just fine without them.

couldn't have said it better... add to that this "FU, I don't care about your grandma's desktop" faux-UNIX/CLI-pride that's been prevalent in the *NIX community, which I've observed numerous times (at least a dozen on slashdot , and no s**t OS X and Windows, and SurfaceFlinger based Linux OS-es (*ahem* , are going to eat the Linux desktop alive..... Things have been definitely moving at a glacial speed on the desktop, but the pieces of the puzzle are finally in place in the kernel, the will of the GPU vendors has shifted... now it's execution time.