<html>
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<p><br>
</p>
<br>
<div class="moz-cite-prefix">On 21/04/2017 10:42 μμ, Martin Gräßlin
wrote:<br>
</div>
<blockquote cite="mid:db6bde7479334c1b387d6c14ca4ceb1e@kde.org"
type="cite">Am 2017-04-21 15:05, schrieb Michail Vourlakos:
<br>
<blockquote type="cite">Following the previous discussion
concerning this topic:
<br>
<br>
<a class="moz-txt-link-freetext" href="https://www.mail-archive.com/plasma-devel@kde.org/msg62432.html">https://www.mail-archive.com/plasma-devel@kde.org/msg62432.html</a>
<br>
<br>
<a class="moz-txt-link-freetext" href="https://www.reddit.com/r/kde/comments/668yul/common_developer_forum_for_docks_wayland_case/?st=j1ru5ogg&sh=5771f221">https://www.reddit.com/r/kde/comments/668yul/common_developer_forum_for_docks_wayland_case/?st=j1ru5ogg&sh=5771f221</a>
<br>
<br>
<br>
Based on the latest information, external docks can use the same
<br>
interface plasmashell does in order to achieve this goal. But
based on
<br>
the discussions I am not persuaded that all the people involved
have
<br>
something concrete and specific for the future. Investing time
and
<br>
commitment in something that will die soon in the future is not
the
<br>
best way to proceed in my opinion.
<br>
<br>
What as a Latte developer would like to know is if this is
considered
<br>
a feasible use case for the plasma wayland branch and how the
plasma
<br>
community would like to handle it.
<br>
</blockquote>
<br>
As I have said before I don't give any guarantees on whether the
interfaces
<br>
which could be used currently, will be available to 3rd party in
the future.
<br>
<br>
Like David said we don't have a concrete plan for that and due to
that you
<br>
won't get any specific answer. I don't want to be responsible if
it breaks so
<br>
I tell you "it might break, no guarantee".
<br>
</blockquote>
<tt>this is ok Martin</tt><br>
<br>
<br>
<blockquote cite="mid:db6bde7479334c1b387d6c14ca4ceb1e@kde.org"
type="cite">
<br>
There is one thing which I can guarantee: you won't get any Latte
dock
<br>
specific changes into KWin. When you still based your work on
Plasma and Plasma
<br>
devs would have said "yeah, that makes sense" adjustments to the
handling would
<br>
have been possible. But the protocol we have is designed only for
the use cases
<br>
of Plasma. This is reflected in the naming of the interfaces
calling them
<br>
PlasmaShellSurface and PlasmaWindowManagement. Anything that is
not going to be
<br>
used by Plasma won't have a chance to be added to those protocols.
<br>
</blockquote>
<tt><br>
I dont think we need any Latte specific code in kwin. Latte is
using just two visibility modes from what kwin provides "</tt><tt><span style="color: rgb(36, 41, 46); font-family: SFMono-Regular, Consolas, "Liberation Mono", Menlo, Courier, monospace; font-size: 12px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; orphans: 2; text-align: start; text-indent: 0px; text-transform: none; white-space: pre; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; background-color: rgb(255, 255, 255); display: inline !important; float: none;">AlwaysVisible"</span>(in
order to update the struts) and "WindowsGoBelow". Almost all modes
provided by Latte are just a "WindowsGoBelow" panel window that
just updates its window mask when needed.<br>
<br>
</tt><tt>The decision for Latte to live outside plasmashell it was
very difficult for me Martin, but on the other hand it was a
technical one. It was obvious from the first patches which I tried
to get into panelview that we were trying to combine to different
design approaches, one of plasma and one of nowdock (back then). I
dont say this with hard feelings but as a simple fact. Latte
focuses in a very specific user case(simple user, fancy interface)
and is used only on desktops. Plasma panels on the other hand have
a much broader audience both on users and devices spectrum.<br>
<br>
Just a few statements from the Latte developement team...<br>
We try to use only kde frameworks and plasma libraries, and this
is a strict design decision for us as we are calling ourselves
kde. We dont want to break kwin experience in any means, if that
happens we consider it a bug and regression that should fixed. The
Latte code stays intentionally close to plasmashell in many parts
and we dont hide it, there is plasma devs blood and commitment
that we are using in many parts of our code and we are grateful
for this (this helps us to follow bug fixes of course and discover
bug fixes faster that influence both project).<br>
<br>
In the very rare case that we want something to be added in kwin,
we will communicate first the plasma devs and if they want that
feature we will implement it first for plasmashell.<br>
<br>
</tt>
<blockquote cite="mid:db6bde7479334c1b387d6c14ca4ceb1e@kde.org"
type="cite">
<br>
If there were cross-desktop specific dock protocols we could talk
about adding
<br>
support for it. But I would be very reluctant. I don't want the
cross-desktop
<br>
dock mess we have on X11. On X11 all docks which are
"cross-desktop" are broken or
<br>
are breaking the desktop environment they are running in. None
uses KWin's screen
<br>
edge auto-hide handling, none uses activities, etc. etc. In many
cases they are
<br>
affectively and knowingly breaking window managers. This is
something I don't want
<br>
to see on Wayland. I don't think that it is possible to have cross
desktop docks, be
<br>
it X11 or Wayland. Doesn't matter. The DEs are too different to
make it work.
<br>
<br>
You have the choice of having a spec which is so vague that all
desktops are covered
<br>
or one which is too limiting by using the least common
denominator. NETWM which is used
<br>
by X11 is the first case and thus it cannot work. One way for
virtual desktops? No, there
<br>
are two! And now docks start to break.
<br>
<br>
So this is really not a Wayland thing as you are putting it on
your reddit thread.
<br>
It's just that those working on Wayland now have the chance to fix
the mess the
<br>
people 25 years ago created.
<br>
<br>
Also I doubt that there will ever be a cross desktop Dock
protocol. I don't think
<br>
any DE which develops a Wayland compositor is interested in a
cross desktop
<br>
Dock protocol.
<br>
<br>
</blockquote>
<tt><br>
pretty clear... either way I dont see any interest from other dock
developers for this. we can consider it a dead <br>
end. The wayland future is pretty clear that docks will be
implemented for specific DEs. In my opinion this makes it even
more important to be able to have plasma specific docks in
wayland. (my personal opinion)<br>
</tt><br>
<tt>regards,</tt><tt><br>
</tt><tt>michail</tt><br>
</body>
</html>