Those who know me know that I’m a huge proponent of free/open-source software (FOSS).  I have four computers, and all run Fedora Linux on them.  Only one has a dual-boot configuration with Windows XP and one has Windows XP on it in a VirtualBox installation. And for what I use, it’s fantastic. I have the best of both worlds. I have the freedom offered to me by Linux and the Windows utilities that I use. I can easily switch from one to the other or use both simultaneously. And, with Windows, I can use its proprietary software I need for my job (e.g. Microsoft Outlook) in addition to open-source equivalents like OpenOffice.org, Thunderbird, Pidgin, and others.

Blogs have long debated if we’ll have a “Year of the Linux Desktop.” In summary, it’s when Linux takes over significantly more market share than the 1% it currently enjoys. For some, it’s yet to come; for others, it’s already passed.

We all know that on servers, Linux is a force with which to be reckoned. For desktops, we’ve seen bits and pieces with the early versions of the Ubuntu operating system. Netbooks were released with a custom Linux distribution that popularized other distributions when users realized that the stock distribution was sub-par. But we just haven’t overcome that threshold of making Linux a truly viable desktop operating system.

But unless the FOSS community comes to its senses and overcomes a few macro-level “big obstacles,” I don’t think that we’ll ever have a “Year of the Linux Desktop.”

Big Obstacle #1: Just because it’s proprietary/closed-source software doesn’t mean it’s bad.
This is something that the FOSS community can’t seem to accept. And instead of seeing what the closed-source software is doing right and trying to adopt some of these technologies to FOSS development, we’re cursing them out. Really, Free Software Foundation? There are a plethora of other reasons why Windows is so popular that have nothing to do with illegal activities.

Look at some of the big pieces of software that we use on a daily basis: Flash. MP3/MP4 codecs. Java. Google Chrome. These are closed-source (but zero-cost) pieces of software that make computers work. To install them on Linux systems takes a good arm twisting and a sense of trying to outwit the computer. Look, the world isn’t going to switch to open-source/open-standards formats. MP3 and MP4 will be more popular than OGG/Theora because more consumer devices support MP3/MP4 instead of OGG. Can one play OGG-formatted files on one’s iPod? No.

Another big issue is power management. Laptops and netbooks running Windows have native power managers that control devices and limit power consumption. Natively, there is nothing for Linux. There are a couple power managers that are developed by individuals in the community but are for one distribution or certain hardware.

Big Obstacle #2: Sometimes, bleeding-edge isn’t the best thing.
I think that as a marketing campaign to try to attract users to the various Linux distributions, it’s effective. But bringing people to Linux is different than keeping them on Linux. Let’s not have a thousand poorly-developed features; instead, let’s have a stable operating system that works. My first adventure in Linux was on Ubuntu operating system, starting in May 2008 with Ubuntu 8.04. It was a great operating system because it just worked. But the developers decided that with each new release, the edge had to be pushed and bleeding-edge became the norm…usually at the expense of core features.

With Ubuntu 10.04, designated as a long-term support release, I was hopeful that a lot of the smaller bugs that definitely pestered me (and I’m sure the community at large, as well) would have been resolved. Take inserting and removing flash drives. I save almost everything to a flash drive. Each time I inserted a flash drive, my computer screen locked up for a minute. The same thing happened when I tried to open a folder from either the desktop or the “Places” menu.

This issue wasn’t resolved. And, as blogger and Linux guru fewt (Andrew Wyatt) explains, there were many more serious issues that weren’t resolved. Instead, the most new notable feature was the shifting of the title bar buttons from the right to the left side, much to the consternation of many. It’s as though the eye-candy took precedence over making a stable, usable operating system. For a long-term support release, it just doesn’t seem to make sense.

Big Obstacle #3: Simplify, simplify, simplify!
Whenever I have to install a fresh operating system on any computer, it’s always a pain to install proprietary drivers, closed-source software, and various other utilities.  The problem with Linux is that unless you’re well-versed in the command line, you’re lost.

Take installing Flash, for instance. Instead of just double-clicking an executable file to install it like I would on Windows, I had to download one file, add something into the software repositories, install a whole host of files from the command line, create a couple symbolic links, and only then was I good to go.  Or to install MP3/MP4 codecs, I really had to convince my operating system that I was okay with installing non-free software despite Linux’s philosophical diatribes on why closed-source/proprietary software is bad.

If Linux is going to have any shot of having a chance of working for consumer desktop operating systems, there should be a way to install drivers/modules for hardware that isn’t natively supported in the Linux kernel. The most promising thing I saw was Ubuntu’s “restricted hardware drivers” program. On my laptop with a nVidia graphics chipset, and Broadcom wireless card, it recognized them both, installed the default drivers, and then just worked. It was all automated and required no input from me. When I switched to Fedora, installing the nVidia drivers required manually adding them into DKMS, installing them on the kernel, modifying GRUB to disable the standard graphics drivers, and restarting the computer. The first time, it didn’t work and I had to reinstall the operating system.

I’m sorry, but computer users are fickle. If that’s what it takes to install something, they’ll ditch Linux and move back to Windows or Mac.  Why do users love Windows or Macintosh? They’re simple. They’re easy to understand. And often times, they just work. Perhaps a way to fix this would be to redouble the Linux community’s efforts onto one or two distributions, package managers, and desktop environments. By having scores of distributions and desktop environments, everyone is making baby steps on their own systems whereas if the community collaborated and worked together, think of the progress that would be made. I’ll start by making a couple suggestions: Let’s stick with Debian, use RPM packages, and GNOME. Why? These have been around for a while, have stood the test of time, and have the greatest potential to “just work.”

Big Obstacle #4: The Linux “community” must be more open to criticism.
If there’s one thing that Linux has that no other operating system has, it’s a great community-level support structure.  Each of the major Linux distributions has a fantastic community system that I can’t find within or without computing.

But that community has to realize that there are people with varying degrees of experience to Linux. It can’t denounce people that complain about Linux as “paid Microsoft shills.” If Desktop Linux is to stand a chance, and especially if Desktop Linux should be taken seriously, the community has to realize that Linux is now on the same playing field as Windows or Macintosh. There are people who are going to try a Linux distribution, say that it doesn’t work, and go back to Windows because something on Linux just didn’t work or because their favorite software title wasn’t available to Linux or whatever. To be cursed at by Linux zealots will only reinforce the notion that Linux isn’t for them. Rather than denounce that person, that criticism should be taken to heart and work on an improvement.

This leads me to the biggest of the big obstacles:

Big Obstacle #5: Desktop Linux won’t have one big saviour. It will have to come from community collaboration.
I think the thought within the Linux communities is that it will be one distribution that will save Desktop Linux, which would come at the expense of all the progress that’s been made on other distributions. I don’t see it that way. The saviour of Linux will be its community but only if it realizes that it must collaborate with and not put down other Linux communities. This goes back to my point #3: by consolidating Linux development into one or two core distributions, significant progress can be made to make Linux a truly viable desktop operating system that can compete with Windows and Macintosh.

Finally, don’t say that “I don’t get Linux.” Maybe I do, maybe I don’t. I think that the great thing about Linux is that there are so many different flavors, versions, and features. But if we’re going to have this long-awaited “Year of the Linux Desktop,” the current models haven’t gotten us anywhere. I’m prepared to argue that they’ve taken us back a few steps. Some change is seriously needed.

So…I’m wondering what your thoughts are on this. There are a lot of blog posts that have been written before this one, and these are just my two cents. Is there anything I missed?

Cheers-
E