KDE Plasma as the default desktop environment

Linux's moves to shun other OSes like FreeBSD will end up turning against them, because people fed up with Linux will know for certain that FreeBSD is not "a distro," but its own thing, and they'll be more prone to try it as a clear alternative; or not. I'm very bad at predicting things.
 
hedwards said:
Inside of Linux it seems like it was mostly a tool to get Poettering a job at MS as that's where he works now. I haven't heard about anything that SystemD does that's both helpful and not available in any of the other systems that could slot into the old system's spot.
Click to expand...
It looks like you have no direct experience on how things work in a professional environment with thousands of Linux powered VMs. I can assure that SystemD when used like it was intended is an amazing tool for both me as a sysadmin and my for my colleagues who develop the apps we use in our production environment.

With that said, as a 57 YO "grey beard" I prefer the simplicity of FreeBSD and I use it as my daily OS on my laptop (14.3 as 15.0 doesn't work reliably on it unfortunately).
 
fmc000 said:
It looks like you have no direct experience on how things work in a professional environment with thousands of Linux powered VMs. I can assure that SystemD when used like it was intended is an amazing tool for both me as a sysadmin and my for my colleagues who develop the apps we use in our production environment.

With that said, as a 57 YO "grey beard" I prefer the simplicity of FreeBSD and I use it as my daily OS on my laptop (14.3 as 15.0 doesn't work reliably on it unfortunately).
Click to expand...
I'd prefer to avoid this degenerating into a flame war, I've been following the development of systemd for a good chunk of its existence and it does a bunch of stuff that's just not appropriate to an startup and service management system. For example, there isn't any reason why I should have to reload services when I edit fstab in order for mount -a to pick up changes. The software knows that the file has changed in order to demand that I reload the related service. But, it doesn't just ask for permission to reload it before proceeding.

If it didn't overstep it's boundaries I wouldn't particularly care one way or the other. It also wouldn't create pressure on projects like Gnome and KDE make a choice about whether to go systemd only or not. There's also the issue that Poettering has at times glommed things into the software that he had to admit he didn't understand.

As far as the mass deployments go, there's a ton of tools that can get the job done depending upon the specifics of the deployment. Mass deployments were happening well before systemd was started and they'll continue even if the project somehow disappears. Nothing that it does justifies it overstepping its boundaries into things that already have effective tools are and unrelated to the process of starting and managing systesm.
 
I do not use KDE anymore, but I hope the project is successful with KDE. If I were "running" things I would have a small team of maybe 2-3 people building a display server, an application framework, and a desktop environment from scratch using locally hosted AI models to keep control of the learned data. Whether we like it or not that is what will be writing things in 20 years in assembly we won't even be able to read. Might as well embrace it now and put it good use. 🤣
 
Anyway, making KDE one of the candidates to be (easily) installable DE is not a bad thing. But to avoid mis-leading, multiple candidates should be needed.
 
Systemd was primarily designed to enable linux be deployed in large-scale cloud services, particularly using containers like docker on VM's, and to integrate with other middleware, where it has been very successful. It can be of some use on a single user desktop, at times, but the real motivation was to enable linux in the cloud. Which is why redhat (read: IBM) spent a lot of money developing it. Poor LP always gets 'blamed' for it, but if he hadn't led the development, someone else would have done, since developing systemd was a business imperative. From a commercial standpoint it was also a way of putting a lot of new IP into linux and making linux (well; you could say the 'mainstream' branch of linux) into something that IBM now 'owns', to a first approximation. Ignoring the desktop, which is of little importance in the cloud. Systemd is essentially a server space technology. The same goes for supporting features in the linux kernel, such as cgroups.

All of which is not to say that some of the criticisms of systemd are not valid, of course. Nothing is perfect...
 
You must log in or register to reply here.
Back
Top