Today's Linux schisms are a blessing in disguise

The Linux community is fracturing along a number of fault lines -- and that's a good thing

It's been a tumultuous year for Linux. Alan Cox quite publicly took a break. Canonical started shipping local desktop search terms to Amazon without really telling anyone, then upped the hubris level to 11 when it was discovered. The systemd wars rage on, and is under attack from a number of angles. Wayland and Mir are up in the air as successors, but that doesn't necessarily mean they won't both win in some form, depending on the distro.

We've also seen a number of Linux kernel brouhahas, with Linus Torvalds telling several individuals and companies to -- ahem -- screw themselves. And there have been relatively recent changes in the Linux kernel release scheme, leading to stable kernels in the 2.6, 3.4, and 3.8 trees -- a situation that never would have occurred back in the days when 2.6 was released. To some, all these issues might seem to indicate big problems within the Linux community, which has been closely knit since inception, raging the open source war against commercial aggressors. The tables, it might seem, have turned.

[ Prove your expertise with the free OS in InfoWorld's& Linux admin IQ test round 1 and round 2. | Track the latest trends in open source with InfoWorld's Open Sources blog and Technology: Open Source newsletter. ]

But I don't think so. I think we're seeing the growing pains of a platform destined to break into several pieces and has already done so in many ways. That may ultimately be a good thing for everyone.

The four main forms of Linux are easy to identify: mobile (Android), server (Red Hat, Debian), desktop (Ubuntu, Mint, Fedora, and so on), and embedded (usually customized Debian distros). Sure, Ubuntu Server is out there, and yes, many people run Debian or CentOS as a desktop, but these distros seem to be gravitating to one side of the line or the other, concentrating on enhancing its value as a desktop, a server, or what have you. This is where the fault lines run.

I can see a future where Linux desktops are successful, not because a general distro breaks out of the mold, but because a company like Ubuntu puts enough time and money into modernizing the desktop experience to compete with Microsoft and Apple. Given enough marketing dollars and OEM agreements, we might actually see the long-anticipated rise of the Linux desktop.

However, that Linux desktop will not necessarily carry with it the same foundation as a server-level product, which has generally been the case with desktop-centric distros in the past. Forgoing the latter for the former makes sense if your focus is on the desktop, damn the consequences. On the other hand, we're seeing a rise in the use of Red Hat (and CentOS) on the server side, along with Debian. These are clearly aimed at the needs of physical and virtual servers, providing toolsets that make those deployments easier and more robust.

Now, tailoring a distribution for a particular arena is not exactly new, but in the past, the tweaks were made to a common base. The changes were generally found in window managers, desktop environments, themes, layouts, and fiddly bits like menu bar widgets. The underpinnings were essentially the same between a desktop and a server system, as well as across different distributions altogether (generally speaking). However, we're currently seeing major changes in the fundaments, not just the presentation.

1 2 Page 1
Page 1 of 2
How to choose a low-code development platform