Posted Oct 24, 2012 16:42 UTC (Wed) by mjg59 (subscriber, #23239)
[Link]
Wayland is a local display server. It's like asking whether the Linux kernel supports ssh. Sure, if someone writes it, but it's not in the scope of kernel development.
Wayland and Weston 1.0 released
Posted Oct 24, 2012 17:03 UTC (Wed) by wmf (guest, #33791)
[Link]
Writing a pedantically-correct FAQ that encourages misunderstandings is a good way to FUD your own project. It should say something like "Remoting is out of scope for the Wayland protocol or Weston compositor, but this will be solved with a proxy that will be bundled with your distro and will Just Workâ˘".
Wayland and Weston 1.0 released
Posted Oct 24, 2012 17:07 UTC (Wed) by wmf (guest, #33791)
[Link]
I hate to respond to myself, but I looked at the actual Wayland FAQ and it's not as bad as the quote above implies, but it's still pretty confusing.
FAQ could be more helpful
Posted Oct 25, 2012 4:28 UTC (Thu) by amtota (guest, #4012)
[Link]
I agree, although I am biased as I maintain xpra.org, I would have expected the Wayland devs to get in touch with us or at least provide a link for those who do want to keep network transparency. Oh well.
Wayland and Weston 1.0 released
Posted Oct 26, 2012 0:17 UTC (Fri) by jschrod (subscriber, #1646)
[Link]
Ah, the systemd argument.
We're replacing "XYZ" that does "ABC", calling it "xyzzy". Believe us, we're doing best work ever, you won't notice anything, and we'll all live happily ever after.
But, "XYZ" does "DEF" as well. How shall we do it in the future?
Well, "DEF" is not the scope of our replacement. It doesn't belong to "xyzzy".
But... "XYZ" did it...
You don't understand: We define the scope. And we're in the position to tell you that while "ABC" might be of interest to you, all the rest of the world doesn't care. So we define that "ABC" wasn't part of "XYZ" and we don't have to replace it.
But... "XYZ" did "ABC"...
As we said, we don't care. We define, you don't need "ABC".
But then... "XYZ" is not a replacement if it doesn't deliver the same functionality.
You're wrong. You're not in the position to define that. We have the power to define what is a replacement and what is not. And we tell you: "xyzzy" is a full replacement for "XYZ", for all relevant functionality. A maze of twisty little passages, all alike.
The story of the Linux desktop,
Joachim
PS: Chosing the name "xyzzy" as illustration is clearly not correct, as it doesn't support remoting to "Y2", even though it should. But, maybe, sometime in the future we'll be able to kill the dragon (aka we'll be able to really replace X) with our bare hands.
Wayland and Weston 1.0 released
Posted Oct 26, 2012 8:17 UTC (Fri) by dgm (subscriber, #49227)
[Link]
> We're replacing "XYZ" that does "ABC", calling it "xyzzy".
Strawman. Wayland does not replace X11. X11 can be run on top of Wayland, just like it does on Windows or OSX. The rest of the argument does not make sense.
Wayland and Weston 1.0 released
Posted Oct 26, 2012 8:25 UTC (Fri) by jschrod (subscriber, #1646)
[Link]
> Wayland does not replace X11.
When Qt und GTK focus on Wayland, these DEs will have native Wayland applications, and X will not be a first class citizen any more. It will only be supported as an legacy interface. As long as we don't want to be stuck using current software versions, that means that de-facto Wayland *will* replace X11 in the medium-to-long term.
Besides, the experience of running X11 on Windows or OS X is no good advertisement, either. I've done both, and good integrated support is something different.
Wayland and Weston 1.0 released
Posted Oct 26, 2012 9:32 UTC (Fri) by paulj (subscriber, #341)
[Link]
This is rubbish. Qt and GTK are not DEs, but toolkits. Further, these toolkits *both* have long supported rendering outputs *other* than X11 for a long time (from Windows GDI to direct framebuffers, even HTML). So there's just no basis to say that the addition of another rendering context, Wayland, is going to change the quality of X11 support.
The only thing that could change that is if X11 becomes so little used that no one cares to support it anymore.
Wayland and Weston 1.0 released
Posted Oct 24, 2012 17:41 UTC (Wed) by daniels (subscriber, #16193)
[Link]
Yeah, it does read quite badly. If you keep on reading, you see it arrive at the real answer of 'yes, it's coming, we just don't think it belongs in core protocol', but unfortunately it starts from a pretty literal interpretation of the question.
Sorry about that. :)
Wayland and Weston 1.0 released
Posted Oct 24, 2012 18:02 UTC (Wed) by apoelstra (subscriber, #75205)
[Link]
> Yeah, it does read quite badly. If you keep on reading, you see it arrive at the real answer of 'yes, it's coming, we just don't think it belongs in core protocol', but unfortunately it starts from a pretty literal interpretation of the question.
> Sorry about that. :)
Thanks for your polite replies in this thread. Myself, and I suspect many others, don't take the time to read up on this stuff, and only encounter Wayland discussions on LWN. And as you say, there are many armchair software developers slinging crap around, and the communication from the Wayland devs thus far leaves something to be desired.
So, it's nice to see people rising above the fray and clearing up misconceptions, rather than getting caught up in flamewars (or completely ignoring the thread).
Wayland and Weston 1.0 released
Posted Oct 24, 2012 22:31 UTC (Wed) by rahvin (subscriber, #16953)
[Link]
As one of the armchair idiots let me just say I was right along with everyone else's negativity until I read the excellent LWN articles and subsequent wayland developer comments that were in replay to the article. Wayland appears to me that it's going to solve some real problems, remove some serious legacy cruft (almost 90% of X isn't even used) and bring the Linux desktop back to the Unix approach which is a simple tool for each task and X is not a simple tool.
The worst part is the misunderstanding about what Wayland is and what it's trying to do as can be seen in the slashdot article and unfortunately in the posts here as well. (as noted, it doesn't help that the FAQ is badly worded in some cases).
Wayland and Weston 1.0 released
Posted Oct 25, 2012 0:14 UTC (Thu) by nix (subscriber, #2304)
[Link]
Agreed. I'll only be unhappy with Wayland if major toolkits don't bother to implement remoting -- but if they do (which seems likely, or Wayland will never take off), it's probably going to become as capable as X already is, with more scope for expansion.
Wayland and Weston 1.0 released
Posted Oct 25, 2012 15:08 UTC (Thu) by renox (subscriber, #23785)
[Link]
> Agreed. I'll only be unhappy with Wayland if major toolkits don't bother to implement remoting
I think that "a proxy pushing giant bitmap" remoting will be eventually available whatever the toolkit used, of course in some case (text) it's quite inefficient compared to XRender glyph cache's remoting..
Wayland and Weston 1.0 released
Posted Oct 25, 2012 22:52 UTC (Thu) by rahvin (subscriber, #16953)
[Link]
The only thing I can do is point to the comments in the previous articles that even if you are running Wayland you can still run X on top even if someone doesn't do a connector or other tool that only runs the parts of X that are needed.
Wayland and Weston 1.0 released
Posted Oct 27, 2012 8:22 UTC (Sat) by paulj (subscriber, #341)
[Link]
The major toolkits already implement remoting that works with Wayland: X11. Hopefully there'll be better remoting protocols in time.
Wayland and Weston 1.0 released
Posted Oct 28, 2012 16:01 UTC (Sun) by nix (subscriber, #2304)
[Link]
True! So we *already* have nothing worse than before, as long as someone keeps the X11->Wayland stuff maintained (and I suspect, given the installed base of X applications, that it's not going to fall unmaintained for a long, long time).