But that is not the point. For people who have never edited a config file, this seems quite strange. They wouldn’t know which editor to use, how to properly sav

But that is not the point. For people who have never edited a config file, this seems quite strange. They wouldn’t know which editor to use, how to properly save the file and, most importantly, how to apply the config file. Chances are, I couldn’t do it intuitively. My go to way would be to just use systemctl restart but not all Unix systems use systemd. And if you start it using the desktop GUI, systemd is not used, so I have to “ps aus | grep redshift“, copy the PID (or process name) and use kill or killall. Sometimes this doesn’t work because it gives me 10 pids because it spawned so many children and I don’t have killall installed. This is not very difficult if you have done it before, but there is steep learning curve before you can do it and quite different from “check a few boxes and click apply”. I personally prefer the config file way because it gives you way better control over your system (especially if a config file causes crashes due to a bug), it is way easier to backup and restore without guessing (program X uses 5 different paths for its gui files) and you can distribute changes easier and without the help of a specialized program. But this sums up the Linux experience quite well IMO: there are many advantages to using Linux and once you have gasped the basics, everything is less scary. But the learning curve is quite steep especially if you have never done anything like it so not many people use it and therefore there are less Linux applications. Furthermore there is also more variety which means the way I know to do might not work for your Linux distro or the old server I’m ssh’ing into.