Tech Blog

PiklesOnFire's Tech Postings

Statistics & Some Complaining

First the good - Arch Linux is up to the number 7 spot on DistroWatch. making this the third consecutive week in a row we have moved up a place. Congrats to the Arch Developers, and everyone who has been contributing to this project. I'm happy to see that this distro is moving up so quickly through the ranks.

Now the bad - I have been reading up a lot on kernels lately. Don't ask me why, I've just been trying to find as much information as I could. I have known the three main categories of kernels for a while now, (Micokernel, Monolithic, and Hybrids) but I never seemed to fully understand everything about them. Some people will argue that an Operating System can be defined by just a running kernel. Others will say that an Operating System is the kernel and everything on top of it, right down to the Web Browser. Still, others will say that they don't even know what you're talking about when you ask them to define an Operating System.

The thing that has started to sort of bug me, is that of all the kernels and Operating Systems I have been looking at and trying out lately, I still seem to come back to GNU/Linux. This is only starting to bug me because to be honest, I like the way Microsoft did the NT kernel. I like how it has the display server in kernel, as well as several other components, yet has other pieces left in user space.(Haiku and SyllableOS do this as well)

Linux is coming a long way in recent years, but is it honestly coming far enough along to be a true desktop or workstation Operating System? I'm starting to have my doubts.

What I would love to see happen in the Linux community is this - We already have two branches of the kernel, stable and development, from there we should have Desktop and Server branches forked out. Desktop kernel should become the new Linux Hybrid Kernel that simply needs to be written to create a fully functional Workstation/Desktop Operating System.

Anothe serious issue with the Linux Kernel that I see is that it is bloated with a lot of cruft. In other words, it is packed full of things that no one even uses any more. We like to support a vast range of hardware, that is the Linux philosophy after all, that it will run on almost anything. Some Linux variants even run on toasters. Seriously, look that up if you haven't heard of it. The problem is though, while we support old, legacy hardware, we are adding to the size of the overall kernel.

I'd love to see a fork here as well, (read below for the other fork) we could have "Linux Legacy" which would be optimized for older systems and only support them, and also a "Linux ..." I was going to say Linux NT, but Microsoft would probably have a fit over that. The point is, is that we should have one kernel to support older technologies, and another to support newer technologies. Either that or go the way of Microsoft and stop supporting older systems, though I would hate to see that happen.

Aside from the issues that I have with the kernel, I would also like to see the two paths of QT and GTK merge. I'm tired of finding an application for KDE that does what I need, but it just won't fit into my GTK based desktop. Even with current work to make applications from one look better on the other, they still feel out of place. The Gnome and KDE teams need to bite the bullet here and create a super user interface. Something the world has never seen before.

I'd love to see Linux brought up to speed, and have a bit more standardization throughout it. I feel if the Linux community could accomplish these goals in say the next 10-20 years, they would probably have the single greatest Operating System man has ever seen.

Thank you for taking the time to read this,
PiklesOnFire

Another day, another laptop set up for powersaving...Making the move to KDE

Write a comment

New comments have been disabled for this post.