I don’t want to get in another argument about what the “biggest” barrier to desktop Linux adoption is. What’s the point? Even if “big” could be quantified, who cares if one is the “biggest”? There are a lot of barriers, and they all work in tandem. These barriers to adoption rarely operate alone.
I won’t try to define what “average” means, but I will say that most people I know (friends, co-workers, relatives, acquaintances, etc.) have no idea that desktop Linux exists. Some haven’t heard the word Linux before. Some have heard the word Linux but have no idea what it is. Still others have heard of Linux but only in a server or embedded context. Good option, bad option—those debates are irrelevant to the mass populace. Linux, to them, isn’t a good or bad option; Linux is a non-option for home or business desktop use.
Among the Windows power users out there, desktop Linux is around as an idea. Some Windows power users have been itching to try desktop Linux. Some have tried it and given up on it. Others have made the migration. But almost all migrants from Windows to desktop Linux come with misconceptions about desktop Linux. I can’t tell you how many times I’ve read new users perpetuating the myths that you have to be a programmer to install or use Linux, that you have to recompile the kernel to get hardware to work, or that you have to compile most programs from source.
The perception is still out there that desktop Linux distros are still geek OSes. Yes, I know that it’s still mostly geeks that use Linux right now, but that doesn’t mean it can’t be used by non-geeks. It just means that geeks are more likely to A) know Linux exists as a desktop option B) have the inclination to download, install, and configure a new operating system and C) troubleshoot problems that come with being an early adopter.
I don’t think that we should unnecessarily puff up desktop Linux as being easier than it is, as being for all users, or for suiting all users’ needs. But we should try to present a realistic portrait of the situation. Right now, all I see are the extremes in the media—desktop Linux will solve all your computing problems! or desktop Linux will never be ready for home users… only geeks can use it.
Even if you can overcome lack of awareness and overcome poor marketing, you still have to build trust. A lot of home users are distrustful of anything free. To many, free equals crappy (or, worse yet, spyware-laden). That’s why it’s best to introduce Windows users to quality open source Windows applications before mentioning anything about desktop Linux. A user who is familiar with and trusting of Windows versions of Firefox, OpenOffice, GIMP, FileZilla, Audacity, and Thunderbird is far more likely to be a successful migrant to desktop Linux. After all, most computer users don’t know what an operating system is. To them, a computer is the applications you run, not the system that allows you to run applications.
Some people can’t get past downloading application installers from the internet and then clicking through wizards. Even though centralized package management makes logical sense, you have to have an open mind to appreciate its benefits. Choice can seem overwhelming at first, but it can be an advantage instead of a hindrance.
If you’ve been reading my recent posts about my parents-in-law, you know that vendor lock-in can be a huge barrier to any migration (even to “user-friendly” Mac OS X). If a site requires Internet Explorer 7 to work properly, you need to use that site, and User Agent Switcher and IEs4Linux aren’t cutting it, then you need Windows. If you have a ton of iTunes store-bought DRM-constrained songs and you don’t want to break the law, you may be stuck using OS X or Windows. If you make your living with InDesign, Photoshop, Illustrator, and Flash; you may find it extremely difficult to do the same with Scribus, GIMP, Inkscape, and F4L.
Yes, desktop Linux has flaws. Yes, there are GUI frontends that haven’t yet been developed or fully implemented. The good news is that developers are working on fixing these problems. Every six months, you see vast improvements—more GUI configuration options, better artwork, more polish.
Again, as you may know from the saga of my parents-in-law’s switch to Mac, the problem is not unique to desktop Linux. Mac, also having a single-digit marketshare, sometimes cannot plug-and-play with certain devices. Having trouble with ATI? Having trouble with Lexmark? Can’t get your Windows-only program working in Wine or Crossover Office? Well, that’s life when you’re using a relatively obscure and not widely used OS.
Need for facilitation
Dell is now shipping preinstalled Ubuntu laptops and desktops… only in the US. You just bought a Windows computer and don’t want to buy a new computer. You don’t want to bother with installing and configuring a new operating system. You don’t want to sift through compatibility lists to find good hardware to buy for a Linux distribution. You don’t know anyone who can fix your Linux problems, and you are not good at doing searches on Google. You don’t have a proper support infrastructure. Well, if any or all of the above apply, then you have no easy route to desktop Linux adoption.
In the Windows world we live in, stores worldwide (online and in-person) sell Windows computers. You probably have a Windows computer already. You probably have Windows-compatible hardware. You probably didn’t have to install or configure Windows. And you probably know a bunch of tech-savvy friends or relatives who can troubleshoot your Windows problems.
In the end…
No one of these barrier types is “the biggest.” They are all big and work together. A small market share goes hand in hand with unfamiliarity, lack of awareness, lack of third-party support, and software insufficiencies. All of those, in turn, help to keep the OS’s market share low, as does vendor lock-in. For those of us who do care about increasing desktop Linux adoption, the best we can hope for is many small strides made on all fronts.