Each company that sells a Linux distribution of its own will be quick to tell you that its kernel is better than others. How can a company make this claim? The answer comes from the fact that each company maintains its own patch set. In order to make sure that the kernels largely stay in sync, most do adopt patches that are put into Linus’ tree (as published on www.kernel.org). The main difference is that vendors typically do not track the release of every single kernel version that is released onto www.kernel.org. Instead, they take a foundation, apply their custom patches to it, run the kernel through their quality assurance (QA) process, and then take it out to production. This helps organisations have confidence that their kernels have been sufficiently baked, thus mitigating any perceived risk of running open source–based operating systems.
The only exception to this rule revolves around security issues. If a security issue is found with a Linux kernel, vendors are quick to adopt the necessary patches to fix the problem immediately. A new release of the kernel is made within a short time (commonly less than 24 hours) so that administrators can be sure their installations are secure. Thankfully, exploits against the kernel itself are rare.
So if each vendor maintains its own patch set, what exactly is it patching? This answer varies from vendor to vendor, depending on each vendor’s target market. Red Hat, for instance, is largely focused on providing enterprise-grade reliability and solid efficiency for application servers. This may be different from the mission of the Fedora team, which is more interested in trying new technologies quickly, and even more different from the approach of a vendor that is trying to put together a desktop-oriented Linux system.
What separates one distribution from the next are the value-added tools that come with each one. Asking “Which distribution is better?” is much like asking “Which is better, Coke or Pepsi?” Almost all colas have the same basic ingredients—carbonated water, caffeine, and high-fructose corn syrup—thereby giving the similar effect of quenching thirst and bringing on a small caffeine-and-sugar buzz. In the end, it’s a question of requirements: Do you need commercial support? Did your application vendor recommend one distribution over another? Does the software (package) updating infrastructure suit your site’s administrative style better than another distribution? When you review your requirements, you’ll find that there is likely a distribution that is geared toward your exact needs.
The idea of giving away source code is a simple one: A user of the software should never be forced to deal with a developer who might or might not support that user’s intentions for the software. The user should never have to wait for bug fixes to be published. More importantly, code developed under the scrutiny of other programmers is typically of higher quality than code written behind locked doors. The greatest benefit of open source software, however, comes from the users themselves: Should they need a new feature, they can add it to the original program and then contribute it back to the source so that everyone else can benefit from it. This line of thinking sprung a desire to release a complete UNIX-like system to the public, free of license restrictions. Of course, before you can build any operating system, you need to build tools. And this is how the GNU project was born.
GNU stands for GNU’s Not UNIX—recursive acronyms are part of hacker humor. If you don’t understand why it’s funny, don’t worry. You’re still in the majority.
