Why do you not contribute to KDE?

Good to hear! I’ll be on the lookout for any development on that!

I created an account on bugs.kde.org to suggest my idea for screen RGB balance, and I was just now checking for duplicate issues, but the search results are hard to navigate through. I haven’t found a duplicate issue, but I’m not sure if there isn’t one.

As for the development side, I mostly write software in Scala nowadays, personal software projects, mostly TUI and server backend, any quirky server idea I could think of that can be used in LAN.
I barely touched C++ and never made a project with Qt or CMake. I’m not confident to enter a whole codebase in C++ and Qt because of this.

I love to contribute to documentation, it just tickles my brain in a unique way, but I haven’t found an opportunity yet because I haven’t encountered anything in KDE that required better documentation and I know enough to write it.

I’m bilingual so I could also contribute to translation, I just haven’t checked out where to start yet, but I will soon.

1 Like

One issue I have had is the compilation time. It took me about 12 hours to compile workspace the last time I did it, and kwin and plasma-desktop each took 1hr 15m. There have been times in the past where I wanted to try fixing a bug in a KCM but gave up because it took so long to compile. Maybe it is possible to only build a single KCM and then add that to system-settings whilst running a distro-packaged plasma but I just don’t know how.

Another problem is how plasma stuff seems to be scattered across multiple repos. You never know if it’ll be in plasma-desktop, plasma-workspace, plasma-addons, or in its own repo (e.g. plasma-nm). Would be great to have a document that outlines where each piece of plasma lives. Now that I have all of workspace cloned it is fairly easy for me to use KFind to search for a string (e.g. “Middle-click to turn on Aeroplane Mode”) and jump right to the source, but this isn’t feasible for new contributors who just want to make small change to an applet.

I also once tried to update an app’s documentation but got put off having to manually edit XML.

1 Like

It’s possible to do this (and will be easier once KF6 is distributed normally) but I’m not sure if it’s well documented. It’s pretty much building that single KCM target, instaling it and then running kcmshell6. You’re on point about how much compilation stuff takes though, but that’s a pretty tough problem to solve :frowning:

1 Like

I guess the compilation time will be less relevant when Plasma 6 releases since I’ll be able to use the distro-packaged frameworks for most apps. Another example of long compilation times is NeoChat – which I assumed would be a straightforward build – which has kwin as a dependency(???).

1 Like

Yeah it’s odd, it doesn’t have a dependency on KWin specifically but one of it’s dependencies does. kdesrc-build isn’t smart enough to disambiguate, you can tell it to ignore it though with --ignore-modules kwin. It could be smarter though, especially since KWin is one of the largest projects we have :smiley:

1 Like

Wow I wish I knew about that option sooner. Definitely helpful to me for reducing compilation time!

1 Like

I can also imagine that people annoyed by old memory safety bugs now learning Rust will have a semi hard time with Qt.

There is CXX-Qt and other bridges but it seems harder than on GTK. Also I have the feeling that GNOME Builder is very developer friendly.