[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    KDE backward compatibility issue
From:       Vasily Khoruzhick <anarsoul () gmail ! com>
Date:       2015-01-30 10:41:42
Message-ID: CA+E=qVfu0MJGx74E=o5UF9+BDsrb-yZwFqtngUgDBja_zArKyA () mail ! gmail ! com
[Download RAW message or body]

Hi,

I'd like to rise KDE backward compatibility issue once again.

I've read [1] and [2] and as far as I understand, dropping xembed
support was more political decision than technical difficulty. Please
kindly understand that breaking applications for users is a no-go.
I'd like to remind a situation when KDE was broken by a change in
kernel API (IIRC it happened several times). But since kernel policy
is "we never break userspace", these changes were reverted.

As a software developer and a KDE user I'd like to see similar
approach in KDE. Nowadays KDE is a software platform, not only a DE,
it provides some APIs to other applications. So KDE's situation is
somewhat similar to kernel's - you should not break backward
compatibility for older applications.

There're some proprietary applications using xembed tray and Qt5 <
5.4, for example - dropbox and viber. There's no way to fix them. As
for GTK apps - it's required to add libappindicator support to the
application, and it's not what app devs are willing to do (correct me
if I'm wrong). So proposed workaround requires some work for app devs.
And I suspect that they don't understand why THEY need to fix the
thing YOU broke.

Looking at all changes happened to KDE for several past years I've an
impression that some KDE devs suffer from NIH syndrom. Broken system
tray (and using instead specification which wasn't adopted by all
other DEs), nepomuk -> baloo transition, no KDE PIM (once again... I
remember times when I lost all local mail archive during transition
from kmail 1 to kmail 2), missing support for autostarting scripts,
one more rework of solid (it isn't dropped yet?) - battery indicator
always shows that battery is charging... Etc, etc.

I understand that some of these issues we'll be addressed within
several KDE releases, but it is somewhat like 1 year. Guys, do you
undestand that you propose to users non-complete and somewhat broken
software for at least one year?

We (KDE users) have already seen issues like that back in KDE 4.0-4.1 times.

Guys, could anyone tell me what prevents you from making transition to
KDE 5 smoother? What's the reason of dropping legacy support? What's
the reason of breaking non-KDE applications?

[1] http://blog.martin-graesslin.com/blog/2014/03/system-tray-in-plasma-next/
[2] http://blog.martin-graesslin.com/blog/2014/06/where-are-my-systray-icons/

>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<
[prev in list] [next in list] [prev in thread] [next in thread]