When people look at my computer screen and don't see a Start button, they start to ask questions:
"What's this?"
"Ubuntu."
"Oh. What's Ubuntu?"
"A type of Linux."
"Oh. What's Linux?"
That's where things get tricky. Most people would say: "It's an operating system." I guess that's true, but it's also misleading. "Linux" isn't a single entity. Windows XP is always sold by Microsoft, and it is always Windows XP. Mac OS X is always sold by Apple, and it's always Mac OS X. Linux, however, is distributed in hundreds and hundreds of forms, and it is almost never identical from one "distro" to the next. Let me now explain what makes any given distro unique.
Kernel
First, a distro starts out with a kernel, which is basically a block of code that connects applications and program to hardware (Linux started out as a kernel). Most users aren't going to care about this, unless they should end up getting a distro whose kernel won't allow a certain piece of hardware to work. The only way that I know how to check hardware compatibility with a kernel is to try the kernel in question. If it doesn't work, I try a different one. Once, I booted Ubuntu with at least 6 different kernels in a short period to find which one worked best.
Desktop Environment
The next characteristic in a distro is the Desktop Environment (DE). Familiar examples of DE's include Windows Vista's "Aero" or Mac OS X's "Aqua." You might call it the "look and feel" of the operating system, but it's also much more than that. With Linux, a DE also entails the basic set of applications and libraries for the system. This is probably the most important thing for a user to consider before trying a distro. Currently, there are two major DE's in use: GNOME and KDE. Others include Xfce, Enlightenment, and Mezzo. Although most disto's usually come with a single DE, you can install an additional or replacement one. I'll write more about DE's later.
Software
One of the great aspects of Linux is the sheer amount of free software, meaning that you don't have to pay for it (more on software freedom later). There are various ways to install software on Linux, in additions to the programs that come with a particular distro. The first is to compile it yourself. This is sometimes fun, while other times it's a pain. Another, less common way is through binary installers (like with Windows). Probably the most significant way is through packages. Three examples of software packages include deb (used by Debian Linux and derivatives, such as Ubuntu), rpm (used by Red Hat Linux, and derivatives like Fedora and OpenSuSe), and ebuild (used by Gentoo-based distro's). These packages are stored on special servers, called repositories. Users can browse the repositories (or repo's) with a package manager, and select which packages they want to install. It's a pretty straightforward process, but I might give a more in-depth explanation later, anyhow.
Those are the key features that I look for in a Linux distro. I find that although there are many variations between distro's, these differences are usually analogous in some way. For instance, Debian uses the 'aptitude' command to fetch and install 'deb' packages, whereas Gentoo uses the 'emerge' command to fetch and install 'ebuild' packages.
So, if you want to call Linux an operating system, that's fine. Just keep in mind that it's also very diverse.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment