Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.user > #246091 > unrolled thread

Wayland vs X

Back to article view | Back to linux.debian.user

Contents

  Wayland vs X Kevin Exton <kcexton01@gmail.com> - 2022-03-08 23:20 +0100
    Re: Wayland vs X Georgi Naplatanov <gosho@oles.biz> - 2022-03-09 11:00 +0100
      Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-09 11:20 +0100
        Re: Wayland vs X didier gaumet <didier.gaumet@gmail.com> - 2022-03-09 15:00 +0100
          Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-09 16:30 +0100
            Re: Wayland vs X didier gaumet <didier.gaumet@gmail.com> - 2022-03-09 21:30 +0100
              Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-10 16:30 +0100
                Re: Wayland vs X didier gaumet <didier.gaumet@gmail.com> - 2022-03-10 18:30 +0100
                  Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-10 19:10 +0100
                    Re: Wayland vs X didier gaumet <didier.gaumet@gmail.com> - 2022-03-10 21:30 +0100
                      Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-11 00:10 +0100
                        Re: Wayland vs X Nate Bargmann <n0nb@n0nb.us> - 2022-03-11 13:10 +0100
                          Re: Wayland vs X Christian Britz <cbritz@t-online.de> - 2022-03-11 14:20 +0100
                            Re: Wayland vs X Nate Bargmann <n0nb@n0nb.us> - 2022-03-11 15:10 +0100
                              Re: Wayland vs X didier gaumet <didier.gaumet@gmail.com> - 2022-03-11 17:40 +0100
                              Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-11 21:10 +0100
                                Re: Wayland vs X Nate Bargmann <n0nb@n0nb.us> - 2022-03-11 23:10 +0100
                                  Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-12 01:10 +0100
                            Re: Wayland vs X Marco Möller <talby@debianlists.mobilxpress.net> - 2022-03-11 19:30 +0100
                              Re: Wayland vs X Tim Woodall <debianuser@woodall.me.uk> - 2022-03-11 20:30 +0100
                              Re: Wayland vs X Nicholas Geovanis <nickgeovanis@gmail.com> - 2022-03-11 22:10 +0100
                                Re: Wayland vs X Nate Bargmann <n0nb@n0nb.us> - 2022-03-11 23:20 +0100
                                Re: Wayland vs X <tomas@tuxteam.de> - 2022-03-12 07:40 +0100
                                Re: Wayland vs X Anssi Saari <as@sci.fi> - 2022-03-14 16:30 +0100
                                  Re: Wayland vs X Marco Möller <talby@debianlists.mobilxpress.net> - 2022-03-14 17:50 +0100
                                    Re: Wayland vs X Anssi Saari <as@sci.fi> - 2022-03-14 18:30 +0100
                                      Re: Wayland vs X Marco Möller <talby@debianlists.mobilxpress.net> - 2022-03-14 20:30 +0100
                                        Re: Wayland vs X Nicholas Geovanis <nickgeovanis@gmail.com> - 2022-03-14 21:10 +0100
                          Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-11 21:20 +0100
                            Re: Wayland vs X Nate Bargmann <n0nb@n0nb.us> - 2022-03-11 23:10 +0100
                          Re: Wayland vs X The Wanderer <wanderer@fastmail.fm> - 2022-03-11 21:50 +0100
                            Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-11 22:10 +0100
                              Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-11 22:30 +0100
                                Re: Wayland vs X Greg Wooledge <greg@wooledge.org> - 2022-03-11 22:40 +0100
                                  Re: Wayland vs X 황병희 <soyeomul@doraji.xyz> - 2022-03-16 07:20 +0100
                              Re: Wayland vs X Greg Wooledge <greg@wooledge.org> - 2022-03-11 22:30 +0100
                                Re: Wayland vs X conover@panix.com (John Conover) - 2022-03-11 23:00 +0100
                              Re: Wayland vs X The Wanderer <wanderer@fastmail.fm> - 2022-03-11 22:40 +0100
                                Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-11 23:00 +0100
                                  Re: Wayland vs X The Wanderer <wanderer@fastmail.fm> - 2022-03-11 23:00 +0100
                                    Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-11 23:10 +0100
                            Re: Wayland vs X <tomas@tuxteam.de> - 2022-03-12 07:30 +0100
                              Re: Wayland vs X The Wanderer <wanderer@fastmail.fm> - 2022-03-12 13:30 +0100
                                Re: Wayland vs X <tomas@tuxteam.de> - 2022-03-12 13:40 +0100
                                  Re: Wayland vs X Nate Bargmann <n0nb@n0nb.us> - 2022-03-12 15:00 +0100
                                    Re: Wayland vs X <tomas@tuxteam.de> - 2022-03-12 15:10 +0100
                                  Re: Wayland vs X The Wanderer <wanderer@fastmail.fm> - 2022-03-13 00:40 +0100
                              Re: Wayland vs X Greg Wooledge <greg@wooledge.org> - 2022-03-12 21:10 +0100
                                Re: Wayland vs X <tomas@tuxteam.de> - 2022-03-12 22:00 +0100
                Re: Wayland vs X didier gaumet <didier.gaumet@gmail.com> - 2022-03-10 18:30 +0100
      Re: Wayland vs X Pascal Obry <pascal@obry.net> - 2022-03-09 11:40 +0100
        Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-09 11:50 +0100
    Re: Wayland vs X Christian Britz <cbritz@t-online.de> - 2022-03-09 11:50 +0100
      Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-09 12:00 +0100
      Re: Wayland vs X Nicholas Geovanis <nickgeovanis@gmail.com> - 2022-03-09 18:20 +0100
        Re: Wayland vs X <tomas@tuxteam.de> - 2022-03-09 18:50 +0100
          Re: Wayland vs X Nicholas Geovanis <nickgeovanis@gmail.com> - 2022-03-09 20:00 +0100
            Re: Wayland vs X <tomas@tuxteam.de> - 2022-03-09 20:40 +0100
              Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-09 20:50 +0100
                Re: Wayland vs X <tomas@tuxteam.de> - 2022-03-09 21:30 +0100
                  Re: Wayland vs X gene heskett <gheskett@shentel.net> - 2022-03-09 21:40 +0100
                    Re: Wayland vs X <tomas@tuxteam.de> - 2022-03-09 22:00 +0100
        Re: Wayland vs X Christian Britz <cbritz@t-online.de> - 2022-03-09 21:20 +0100
    Re: Wayland vs X 황병희 <soyeomul@doraji.xyz> - 2022-03-16 07:00 +0100
      Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-16 08:00 +0100
        Re: Wayland vs X Nate Bargmann <n0nb@n0nb.us> - 2022-03-16 08:20 +0100
          Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-16 08:30 +0100
        Re: Wayland vs X "Byung-Hee HWANG" <soyeomul@doraji.xyz> - 2022-03-16 08:40 +0100
        Re: Wayland vs X Greg Wooledge <greg@wooledge.org> - 2022-03-16 12:40 +0100
          Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-16 13:10 +0100
            Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-16 13:20 +0100
              Re: Wayland vs X Emanuel Berg <moasenwood@zoho.eu> - 2022-03-16 13:40 +0100

Page 3 of 4 — ← Prev page 1 2 [3] 4  Next page →

#246221

The Wanderer wrote:

>>> There may be multiple reasons, but one of them is that the
>>> feature set supported by Wayland (and/or the associated
>>> protocol, if any) is not a superset of the feature set
>>> supported by the X protocol.
>> 
>> They should have covered everything used anyway ... this
>> must be a blunder on their part.
>
> My understanding (which is at best secondhand, and may well
> not be fully accurate, if accurate at all) is that they
> thought they *were* doing that, since obviously that one
> feature is a historical curiosity which can't possibly be
> useful to anybody, and is only present in the spec because
> the people who wrote the spec mistakenly thought it might
> turn out to be useful.
>
> Only to discover later that people actually *did* use that
> feature, but by that point it was too late to implement that
> feature in Wayland without breaking a lot of other things
> and/or requiring a deeper redesign than would be
> deemed practical.

Sounds like poor design/modularity to begin with ...

-- 
underground experts united
https://dataswamp.org/~incal

[toc] | [prev] | [next] | [standalone]

#246231

[Multipart message — attachments visible in raw view] — view raw

On Fri, Mar 11, 2022 at 03:41:09PM -0500, The Wanderer wrote:

[...]

> The most important one for my purposes, and therefore the one that I
> remember, is the ability to have multiple desktop-like things which are
> actually all just viewports on one much-larger single area [...]

There seems to be some basis to it. And some solution. But then, you're
perhaps bound to a specific toolkit [1] [2] or perhaps compositor.

> Another limitation of XWayland as I've heard it described (by the same
> person on whose statements the previous paragraph is based, as well as
> in online discussions related to XWeston, below), as compared to a full
> X server: where you can (and, in fact, usually need to) run a window
> manager on top of an X server [...]

This is absolutely my main beef: I do chose my window manager
judiciously; after long use I know it well and perceive it as my
ally in front of applications which sometimes have (to me) strange
ideas of GUI [0].

With X, the window manager is the one implementing window decorations
(this isn't in the protocol, but it is a strong convention applications
had to follow in practice).

With Wayland, you can see the appetite of applications and toolkits
to unleash their "creativity" on the unsuspecting user.

I don't look forward to the day where the browser gives some random
javascript advertisment control over its absolute position on my
screen or over its window decorations. But this day is coming: don't
forget that our main browser provider is an ad company, and this
industry (as the closely related DRM industry, too) tends to view
your computer as /their/ "content" delivery device.

You can get a feel of that taste now with all those "GUIs" badly
implemented by the never-ending horde of javascript frameworks.
Jamie Zawinski's CADT [2] was hell, we now discovered an even
worse place.

I'll stick to X. My computer feels a bit more... mine this way.

Cheers

[0] Heck, I've even a key shortcut for xkill, afer some incarnation
    of... yes, Firefox, decided to ignore the window close button.
    They've fixed it in the meantime, it seems, but I do keep that
    shortcut around.

[1] https://www.phoronix.com/forums/forum/linux-graphics-x-org-drivers/wayland-display-server/1074550-kde-now-has-virtual-desktop-support-on-wayland
    and the links therein.

[2] https://www.jwz.org/doc/cadt.html

-- 
tomás

[toc] | [prev] | [next] | [standalone]

#246234

[Multipart message — attachments visible in raw view] — view raw

On 2022-03-12 at 01:29, tomas@tuxteam.de wrote:

> On Fri, Mar 11, 2022 at 03:41:09PM -0500, The Wanderer wrote:
> 
> [...]
> 
>> The most important one for my purposes, and therefore the one that
>> I remember, is the ability to have multiple desktop-like things
>> which are actually all just viewports on one much-larger single
>> area [...]
> 
> There seems to be some basis to it. And some solution. But then,
> you're perhaps bound to a specific toolkit [1] [2] or perhaps
> compositor.

> [1] https://www.phoronix.com/forums/forum/linux-graphics-x-org-drivers/wayland-display-server/1074550-kde-now-has-virtual-desktop-support-on-wayland
>     and the links therein.

It's more complicated than that, unfortunately.

There's a reason I didn't use the phrase "virtual desktops" in my
description of this feature; the X spec defines *two* things which are
sometimes called by that name.

One of them has a single large "desktop" with multiple viewports into
it; on that one the parts of one window that stick off the edge of one
viewport overflow into, and can be seen through, an adjacent viewport.
That's the feature I was talking about, but it is *not* the feature most
commonly called "virtual desktops", although some WMs (including, IIRC,
e16) do call it that; I don't know if it has any other dedicated name,
although the X spec does refer to it in different terminology. My
understanding is that this is the thing the Wayland developers saw as so
odd that it couldn't possibly be used/wanted by anyone and had to just
be a historical-curiosity wart on the spec.

The other defines multiple separate "desktops", which are logically
arranged into a grid for purposes of indexing and access, but which are
individually independent; anything sticking off the edge of any one of
them is not visible anywhere. That, as I understand matters, is the
feature commonly called "virtual desktops". It's my understanding that
this feature *is* possible via, and maybe even directly supported by,
Wayland.

It's difficult or impossible to tell for certain from the limited
discussion in the links provided, but it looks to me (having dug through
as far as the Phabricator discussion) as if what KDE added support for
is the latter.

(FWIW, e16 apparently supports *both* of these features, although the
major rewrite that was e17 and later dropped support for the first one;
that's one of the reasons I haven't moved forward to newer versions of
Enlightenment.)

-- 
   The Wanderer

The reasonable man adapts himself to the world; the unreasonable one
persists in trying to adapt the world to himself. Therefore all
progress depends on the unreasonable man.         -- George Bernard Shaw

[toc] | [prev] | [next] | [standalone]

#246235

[Multipart message — attachments visible in raw view] — view raw

On Sat, Mar 12, 2022 at 07:22:11AM -0500, The Wanderer wrote:
> On 2022-03-12 at 01:29, tomas@tuxteam.de wrote:
> 
> > On Fri, Mar 11, 2022 at 03:41:09PM -0500, The Wanderer wrote:
> > 
> > [...]
> > 
> >> The most important one for my purposes, and therefore the one that
> >> I remember, is the ability to have multiple desktop-like things
> >> which are actually all just viewports on one much-larger single
> >> area [...]
> > 
> > There seems to be some basis to it. And some solution. But then,
> > you're perhaps bound to a specific toolkit [1] [2] or perhaps
> > compositor.
> 
> > [1] https://www.phoronix.com/forums/forum/linux-graphics-x-org-drivers/wayland-display-server/1074550-kde-now-has-virtual-desktop-support-on-wayland
> >     and the links therein.
> 
> It's more complicated than that, unfortunately.
> 
> There's a reason I didn't use the phrase "virtual desktops" in my
> description of this feature; the X spec defines *two* things which are
> sometimes called by that name.
> 
> One of them has a single large "desktop" with multiple viewports into
> it; on that one the parts of one window that stick off the edge of one
> viewport overflow into, and can be seen through, an adjacent viewport.

Ah, I see. I dimly remember that one (I'm that old ;-)

> That's the feature I was talking about, but it is *not* the feature most
> commonly called "virtual desktops", although some WMs (including, IIRC,
> e16) do call it that; I don't know if it has any other dedicated name,
> although the X spec does refer to it in different terminology. My
> understanding is that this is the thing the Wayland developers saw as so
> odd that it couldn't possibly be used/wanted by anyone and had to just
> be a historical-curiosity wart on the spec.
> 
> The other defines multiple separate "desktops", which are logically
> arranged into a grid for purposes of indexing and access, but which are
> individually independent; anything sticking off the edge of any one of
> them is not visible anywhere. That, as I understand matters, is the
> feature commonly called "virtual desktops". It's my understanding that
> this feature *is* possible via, and maybe even directly supported by,
> Wayland.

And then, there are window managers (Fvwm) which offer "big" desktops
(where the visible screen is a window into, which can be moved around
seamlessly) and then several of that "virtual desktops".

Best of two worlds :-)

Whether it uses that X functionality is an implementation detail I
don't know, alas.

> It's difficult or impossible to tell for certain from the limited
> discussion in the links provided, but it looks to me (having dug through
> as far as the Phabricator discussion) as if what KDE added support for
> is the latter.
> 
> (FWIW, e16 apparently supports *both* of these features, although the
> major rewrite that was e17 and later dropped support for the first one;
> that's one of the reasons I haven't moved forward to newer versions of
> Enlightenment.)

I have the hunch Fvwm is for you. I'm using it with one virtual desktop
which is 3x3 the size of my screen ("pages" in Fvwm parlance). Of course
it's segmented as nine pages, but windows sticking out of my current
page end up sticking into the corresponding neighbour. And I can get
my screen (aka viewport) to straddle page boundaries (which I don't do
usually, but hey).

Cheers
-- 
t

[toc] | [prev] | [next] | [standalone]

#246236

[Multipart message — attachments visible in raw view] — view raw

* On 2022 12 Mar 06:38 -0600, tomas@tuxteam.de wrote:
> On Sat, Mar 12, 2022 at 07:22:11AM -0500, The Wanderer wrote:
> > 
> > The other defines multiple separate "desktops", which are logically
> > arranged into a grid for purposes of indexing and access, but which are
> > individually independent; anything sticking off the edge of any one of
> > them is not visible anywhere. That, as I understand matters, is the
> > feature commonly called "virtual desktops". It's my understanding that
> > this feature *is* possible via, and maybe even directly supported by,
> > Wayland.
> 
> And then, there are window managers (Fvwm) which offer "big" desktops
> (where the visible screen is a window into, which can be moved around
> seamlessly) and then several of that "virtual desktops".

That is what I recall from a bit over 25 years ago when I bought a 1.2
GB hard drive to have enough space to install the X disk sets in
Slackware 96.  The default WM was Fvwm95 and I used it with a large
virtual desktop for several years.  Then I chose to try Afterstep, then
IceWM for some time before moving into the desktop world alternating
between KDE and Xfce and now Gnome for the most part and my virtual
desktop equals the screen size.

Gnome calls them "workspaces" and I typically use four per screen.  It's
default is to create them dynamically but I use a fixed number.  This
way I set up my work flow the same as on systems where I used Xfce which
defaults to four desktops.

- Nate

-- 
"The optimist proclaims that we live in the best of all
possible worlds.  The pessimist fears this is true."
Web: https://www.n0nb.us
Projects: https://github.com/N0NB
GPG fingerprint: 82D6 4F6B 0E67 CD41 F689 BBA6 FB2C 5130 D55A 8819

[toc] | [prev] | [next] | [standalone]

#246238

[Multipart message — attachments visible in raw view] — view raw

On Sat, Mar 12, 2022 at 07:50:33AM -0600, Nate Bargmann wrote:
> * On 2022 12 Mar 06:38 -0600, tomas@tuxteam.de wrote:
> > On Sat, Mar 12, 2022 at 07:22:11AM -0500, The Wanderer wrote:
> > > 
> > > The other defines multiple separate "desktops", which are logically
> > > arranged into a grid for purposes of indexing and access, but which are
> > > individually independent; anything sticking off the edge of any one of
> > > them is not visible anywhere. That, as I understand matters, is the
> > > feature commonly called "virtual desktops". It's my understanding that
> > > this feature *is* possible via, and maybe even directly supported by,
> > > Wayland.
> > 
> > And then, there are window managers (Fvwm) which offer "big" desktops
> > (where the visible screen is a window into, which can be moved around
> > seamlessly) and then several of that "virtual desktops".
> 
> That is what I recall from a bit over 25 years ago when I bought a 1.2
> GB hard drive to have enough space to install the X disk sets in
> Slackware 96 [...]

Ah, memories. I'm older: my first one was Twm ;-P

After an excursion which took me all the way to Gnome (I liked those
around 2-ish, actually), then to Xfce I'm back with Fvwm. Phew.

Fvwm95 I never understood: it wants to look & feel like Windows95. I
hated Windows since 3.1 :-)

> Gnome calls them "workspaces" and I typically use four per screen.

To come back on topic: can Gnome (is their WM still called Metacity?)
straddle workspaces with the viewport?

Otherwise we'll stick with our more advanced WMs ;-)

Cheers
-- 
t

[toc] | [prev] | [next] | [standalone]

#246260

[Multipart message — attachments visible in raw view] — view raw

On 2022-03-12 at 07:38, tomas@tuxteam.de wrote:

> On Sat, Mar 12, 2022 at 07:22:11AM -0500, The Wanderer wrote:
>
>> On 2022-03-12 at 01:29, tomas@tuxteam.de wrote:

>>> There seems to be some basis to it. And some solution. But then, 
>>> you're perhaps bound to a specific toolkit [1] [2] or perhaps 
>>> compositor.
>> 
>>> [1] https://www.phoronix.com/forums/forum/linux-graphics-x-org-drivers/wayland-display-server/1074550-kde-now-has-virtual-desktop-support-on-wayland
>>>     and the links therein.
>> 
>> It's more complicated than that, unfortunately.
>> 
>> There's a reason I didn't use the phrase "virtual desktops" in my 
>> description of this feature; the X spec defines *two* things which
>> are sometimes called by that name.
>> 
>> One of them has a single large "desktop" with multiple viewports
>> into it; on that one the parts of one window that stick off the
>> edge of one viewport overflow into, and can be seen through, an
>> adjacent viewport.
> 
> Ah, I see. I dimly remember that one (I'm that old ;-)

>> The other defines multiple separate "desktops", which are
>> logically arranged into a grid for purposes of indexing and access,
>> but which are individually independent; anything sticking off the
>> edge of any one of them is not visible anywhere. That, as I
>> understand matters, is the feature commonly called "virtual
>> desktops". It's my understanding that this feature *is* possible
>> via, and maybe even directly supported by, Wayland.
> 
> And then, there are window managers (Fvwm) which offer "big"
> desktops (where the visible screen is a window into, which can be
> moved around seamlessly) and then several of that "virtual
> desktops".
> 
> Best of two worlds :-)

I believe e16 also does this, and if I understand matters correctly,
that's just the intersection of these two features.

If you cut the number of viewports per desktop down to 1 and the size of
the desktop to the size of the viewport, you get traditional virtual
desktops.

If you cut the number of virtual desktops down to 1, and increase the
number of viewports per desktop and (correspondingly) the size of each
desktop, you get that other virtual-desktop-like feature whose proper
other name I don't recall (but which I use routinely).

(Now that I think about it, one of these two might be properly called
"multiple desktops", and the other "virtual desktops" - but I'm no
longer positive which one is which.)

I haven't tried, because I don't want to spend the screen real estate on
the necessary pager for switching among virtual desktops, or expend
potential keybindings to be able to switch among virtual desktops
without a pager - but I'm fairly sure I could trivially configure e16 to
have a 2x2 or 3x3 or 4x4 grid of virtual desktops, each of which
consists of a 4x8 grid of viewports, each of which provides access to an
area the size of my screen resolution.

>> It's difficult or impossible to tell for certain from the limited 
>> discussion in the links provided, but it looks to me (having dug
>> through as far as the Phabricator discussion) as if what KDE added
>> support for is the latter.
>> 
>> (FWIW, e16 apparently supports *both* of these features, although
>> the major rewrite that was e17 and later dropped support for the
>> first one; that's one of the reasons I haven't moved forward to
>> newer versions of Enlightenment.)
> 
> I have the hunch Fvwm is for you. I'm using it with one virtual
> desktop which is 3x3 the size of my screen ("pages" in Fvwm
> parlance). Of course it's segmented as nine pages, but windows
> sticking out of my current page end up sticking into the
> corresponding neighbour. And I can get my screen (aka viewport) to
> straddle page boundaries (which I don't do usually, but hey).

There are other features of e16 which I like (and I can try to run
through the list in detail if desired, although it might take me a while
to dredge everything up, since I only think about most of them when
setting up a new machine or when having to live without them for a
while), but I've heard positive-sounding things about fvwm before,
certainly.

-- 
   The Wanderer

The reasonable man adapts himself to the world; the unreasonable one
persists in trying to adapt the world to himself. Therefore all
progress depends on the unreasonable man.         -- George Bernard Shaw

[toc] | [prev] | [next] | [standalone]

#246246

On Sat, Mar 12, 2022 at 07:29:19AM +0100, tomas@tuxteam.de wrote:
> With X, the window manager is the one implementing window decorations
> (this isn't in the protocol, but it is a strong convention applications
> had to follow in practice).

Except for when Google Chrome decided to implement its own window
decorations, which were of course completely foreign compared to
the rest of the toplevel windows on my display.

Fortunately, they decided to revert that decision, and today you can
configure Google Chrome to look and act like a normal window.

> I don't look forward to the day where the browser gives some random
> javascript advertisment control over its absolute position on my
> screen [...]

Are you kidding?  Javascript could do this *ages* ago.

I wrote a series of little pages that showed how insidiously evil
Javascript actually is (or was):

https://wooledge.org/~greg/jsabuse/

Some of them still work.  Some do not, as current browser versions have
tightened up their settings and no longer give Javascript quite as much
freedom as they originally did.

> or over its window decorations.

I wouldn't know anything about that in particular.

[toc] | [prev] | [next] | [standalone]

#246247

[Multipart message — attachments visible in raw view] — view raw

On Sat, Mar 12, 2022 at 03:05:03PM -0500, Greg Wooledge wrote:
> On Sat, Mar 12, 2022 at 07:29:19AM +0100, tomas@tuxteam.de wrote:
> > With X, the window manager is the one implementing window decorations
> > (this isn't in the protocol, but it is a strong convention applications
> > had to follow in practice).
> 
> Except for when Google Chrome decided to implement its own window
> decorations, which were of course completely foreign compared to
> the rest of the toplevel windows on my display.

I haven't ever tried Chrome or Chromium (and I'll try to avoid them
for as long as I can): you reinforce me on that :)

> Fortunately, they decided to revert that decision, and today you can
> configure Google Chrome to look and act like a normal window.

Phew!

> > I don't look forward to the day where the browser gives some random
> > javascript advertisment control over its absolute position on my
> > screen [...]
> 
> Are you kidding?  Javascript could do this *ages* ago.

In the classical setting it can only beg the window manager (through
ICCCM). The latter has the last word on it.

> I wrote a series of little pages that showed how insidiously evil
> Javascript actually is (or was):
> 
> https://wooledge.org/~greg/jsabuse/

Nice :-)

(FWIW: I can click at 3.html and 4.html: I guess Mozilla begs
the WM to move/resize, but the WM says "nope".

> Some of them still work.  Some do not, as current browser versions have
> tightened up their settings and no longer give Javascript quite as much
> freedom as they originally did.

I sure hope. But I'm sceptical: the browser maker's perspective
is the ad industry's perspective (I'm not assuming some evil
conspiracy, just plain boring cultural immersion).

> > or over its window decorations.
> 
> I wouldn't know anything about that in particular.

I don't even want to ;-)

Cheers & thanks for the little javascript snippets. You definitely
picked your background colours with fury :-D

-- 
tomás

[toc] | [prev] | [next] | [standalone]

#246166

Le jeudi 10 mars 2022 à 16:27 +0100, Emanuel Berg a écrit :
[...]
> What are the pure X applications? This is the software I use
> almost every day
> 
>   https://dataswamp.org/~incal/SOFTWARE
> 
> The ones that start with an "x" are xterm and xpdf ...

If I understand correctly, a GUI application is (or not) compatible with Wayland, depending on the graphical toolkit with which it has been developed.
https://wiki.debian.org/Wayland#Toolkits
So I think that the apps developed with an incompatible toolkit or no toolkit at all are not wayland compatible and need to run on a nested X server (Xwayland) inside Wayland.

[toc] | [prev] | [next] | [standalone]

#246112

Hello,

One drawback at the moment is the color management of Wayland when
using the NVidia proprietary driver (not sure about Wayland on
nouveau). This is just not supported and if you need proper color
management like me using for example darktable then Wayland is not an
option yet.

If I understand correctly there is work on this part, so at some point
Wayland will be the best choice for the already mentioned big
enhancements it brings. 

Regards,

-- 
  Pascal Obry /  Magny Les Hameaux (78)

  The best way to travel is by means of imagination

  http://www.obry.net

  gpg --keyserver keys.gnupg.net --recv-key F949BD3B

[toc] | [prev] | [next] | [standalone]

#246113

Pascal Obry wrote:

> One drawback at the moment is the color management of
> Wayland when using the NVidia proprietary driver (not sure
> about Wayland on nouveau). This is just not supported and if
> you need proper color management like me using for example
> darktable then Wayland is not an option yet.

Hm, those sound familiar, do you have a command to find out
which one is used?

-- 
underground experts united
https://dataswamp.org/~incal

[toc] | [prev] | [next] | [standalone]

#246114

Hi Kevin,

On 2022-03-08 23:01 UTC+0100, Kevin Exton wrote:
> I tried Wayland some years ago now (might have been when they first
> trialled it in Ubuntu) but decided not to stick with it.
> 
> Since more desktop environments are beginning to choose Wayland as the
> default display protocol, I was wondering if others think there's
> significant user benefits to making the change?

I had it running for a while recently with KDE Plasma, because I wanted
to try out waydroid, which only supports Wayland. My impression was,
that it has improved a lot and is in a somewhat usable state, but I
quickly changed back to X.

I felt that programs designed for X and programs designed for Wayland
don't interact well (copy/paste and so on) and Plasma support for
Wayland is not as mature as for X. The worst experience was the
proprietary Citrix client, which I need to use on a daily basis for my
work. They say they don't support Wayland at all. Well it did somehow
work but it was absolutely not satisfying (again copy/paste, alt+tab
behaviour and so on).

I came to the conclusion that I don't really need or want Wayland at
this point of time. It might be technically better, but X does all I
need these days. And I like the X forwarding feature of SSH.

My biggest concern is that we will see over the next years a big
division of the GUI application eco system. Some programs will stay
optimised for X, some will even require Wayland, like the mentioned
waydroid. Hopefully a full transparency can somehow be achieved.

Regards,
Christian

-- 
http://www.cb-fraggle.de

[toc] | [prev] | [next] | [standalone]

#246116

Christian Britz wrote:

> My biggest concern is that we will see over the next years
> a big division of the GUI application eco system.
> Some programs will stay optimised for X, some will even
> require Wayland, like the mentioned waydroid.
> Hopefully a full transparency can somehow be achieved.

That so, I was told the applications would remain the same,
only the protocols were to be changed.

I have tons of X stuff in shell functions, that means that
won't work with Wayland?

-- 
underground experts united
https://dataswamp.org/~incal

[toc] | [prev] | [next] | [standalone]

#246128

[Multipart message — attachments visible in raw view] — view raw

On Wed, Mar 9, 2022 at 4:40 AM Christian Britz <cbritz@t-online.de> wrote:

> Hi Kevin,
>
> On 2022-03-08 23:01 UTC+0100, Kevin Exton wrote:
> > I tried Wayland some years ago now (might have been when they first
> > trialled it in Ubuntu) but decided not to stick with it.
> >
> > Since more desktop environments are beginning to choose Wayland as the
> > default display protocol, I was wondering if others think there's
> > significant user benefits to making the change?
>
> I had it running for a while recently with KDE Plasma, because I wanted
> to try out waydroid, which only supports Wayland. My impression was,
> that it has improved a lot and is in a somewhat usable state, but I
> quickly changed back to X.
>

Here's something about the X ecosystem that I don't understand anymore.
I've read the statement
that "X isn't developed anymore, only XFree86". In other words, X source
was re-based to Xfree86
a long time ago. So it must be that in the XFree86-based X Windows, only
the X server is under active
development, nothing more than that. Is that accurate?


> I felt that programs designed for X and programs designed for Wayland
> don't interact well (copy/paste and so on) and Plasma support for
> Wayland is not as mature as for X.

........

> My biggest concern is that we will see over the next years a big
> division of the GUI application eco system. Some programs will stay
> optimised for X, some will even require Wayland, like the mentioned
> waydroid. Hopefully a full transparency can somehow be achieved.
>

And isn't that what happened once already? When XFree86 re-based X Windows?


> Regards,
> Christian
>
> --
> http://www.cb-fraggle.de
>
>

[toc] | [prev] | [next] | [standalone]

#246129

[Multipart message — attachments visible in raw view] — view raw

On Wed, Mar 09, 2022 at 11:15:54AM -0600, Nicholas Geovanis wrote:

[...]

> Here's something about the X ecosystem that I don't understand anymore.
> I've read the statement
> that "X isn't developed anymore, only XFree86". In other words, X source
> was re-based to Xfree86
> a long time ago. So it must be that in the XFree86-based X Windows, only
> the X server is under active
> development, nothing more than that. Is that accurate?

I think the authoritative place is:

  https://x.org/wiki/

(you can donate via SPI, if you wish ;-)

Cheers
-- 
t

[toc] | [prev] | [next] | [standalone]

#246131

[Multipart message — attachments visible in raw view] — view raw

On Wed, Mar 9, 2022, 11:49 AM <tomas@tuxteam.de> wrote:

> On Wed, Mar 09, 2022 at 11:15:54AM -0600, Nicholas Geovanis wrote:
>
> [...]
>
> > Here's something about the X ecosystem that I don't understand anymore.
> > I've read the statement
> > that "X isn't developed anymore, only XFree86". In other words, X source
> > was re-based to Xfree86
> > a long time ago. So it must be that in the XFree86-based X Windows, only
> > the X server is under active
> > development, nothing more than that. Is that accurate?
>
> I think the authoritative place is:
>
>   https://x.org/wiki/
>
> (you can donate via SPI, if you wish ;-)
>

See I can't tell if that supports my claim or yours :-)
Here's why: I went to that site and among others, looked at the page for
distribution-specific X mail lists. Out of Redhat, Suse, Oracle OpenSolaris
and OpenBSD, not one of those X-specific lists was active. Most not in
years.

But Debian's X list is thriving. Does that imply that only Debian carries a
torch for truly platform- and server-agnostic X Windows?
I didn't check Canonical/ubuntu.

Cheers
> --
> t
>

[toc] | [prev] | [next] | [standalone]

#246132

[Multipart message — attachments visible in raw view] — view raw

On Wed, Mar 09, 2022 at 12:51:24PM -0600, Nicholas Geovanis wrote:

[...]

> See I can't tell if that supports my claim or yours :-)

When in doubt, it supports your claim. I'm not aware of having made
a significant one ;-)

> Here's why: I went to that site and among others, looked at the page for
> distribution-specific X mail lists. Out of Redhat, Suse, Oracle OpenSolaris
> and OpenBSD, not one of those X-specific lists was active. Most not in
> years.
> 
> But Debian's X list is thriving. Does that imply that only Debian carries a
> torch for truly platform- and server-agnostic X Windows?
> I didn't check Canonical/ubuntu.

I think Wayland has taken all the hype for a while. But I think the
people involved are smart enough to try to factor out common work

I hope X isn't going anywhere the next 10..15 years. After that, I'll
be gone myself, so... ;-)

Cheers
-- 
t

[toc] | [prev] | [next] | [standalone]

#246133

tomas wrote:

> I hope X isn't going anywhere the next 10..15 years.
> After that, I'll be gone myself, so... ;-)

:( *sob*

-- 
underground experts united
https://dataswamp.org/~incal

[toc] | [prev] | [next] | [standalone]

#246136

[Multipart message — attachments visible in raw view] — view raw

On Wed, Mar 09, 2022 at 08:42:31PM +0100, Emanuel Berg wrote:
> tomas wrote:
> 
> > I hope X isn't going anywhere the next 10..15 years.
> > After that, I'll be gone myself, so... ;-)
> 
> :( *sob*

Don't worry. Life is like that :)

Cheers
-- 
t

[toc] | [prev] | [next] | [standalone]

Page 3 of 4 — ← Prev page 1 2 [3] 4  Next page →

Back to top | Article view | linux.debian.user

csiph-web