The new Windows 8 Metro desktop, the latest incarnation of Mac OSX, Android, Ubuntu's Unity desktop and GNOME 3, love them or hate them, all came about because of the success of Linux on the desktop. No, you didn't misread that and, no, I'm not talking about market share.
Linux has been ahead in terms of desktop innovation and the others have been copying since the turn of the century. There are lots of reasons why Windows is and will likely remain dominant on conventional desktop systems but lack of a better alternative has never been one of them. The quality of the Linux desktop has been fine for more than a decade.
With a new major Windows release just around the corner we have been treated to the usual slew of articles proclaiming the Linux desktop dead and the Linux market share as vanishingly small. I even saw one article claiming Linux desktop usage declined from 2.5% of the market to under 1%. Of course, no source was given for this ludicrous claim and no survey methodology was provided. As I pointed out in my 2010 article, numbers like this come either from deliberately parsing data in a way that is unfavorable to Linux or from wildly unreliable web counters. For example, if you want to claim Linux is unimportant you don't use numbers from Forrester Research, who count both tablets and netbooks as part of the desktop market.
The reality really hasn't changed in the last two years. Linux use is growing very slowly on the conventional desktop but expanding rapidly on mobile devices, tablets and low end netbooks with ARM CPUs, mainly in the form of Google's Android distribution. Microsoft is still the dominant player on the conventional desktop and is likely to remain so because the market, as it is structured today, makes it impossible for Linux to compete.
There are only two ways Linux could make headway on the consumer desktop: the first is if Linux is available preloaded in stores the way it was on netbooks for a couple of years. That is happening with low end Android netbooks at K-Mart here in the United States and similar low end stores but it's a tiny sliver of the market. There is no incentive for retailers to carry Linux even if Microsoft doesn't bring pressure to bear. If you sell a Windows system or a Mac you get to sell all sorts of software, from anti-virus to Office. With a Linux system that's all free. The markup on the systems themselves is small. There is no money to be made selling Linux retail the way the market is currently structured.
The other avenue to desktop success is the corporate marketplace. The big corporate players in the server room, Red Hat and SUSE, have never devoted the marketing resources to the corporate desktop. The server is simply more profitable.
A number of recent articles have claimed that the Linux desktop is perpetually playing catch-up or is somehow inferior. Others claim that consumers have rejected Linux. To put it bluntly these claims are arrant nonsense and represent revisionist history. Sales of Linux on netbooks remained robust for two years after the introduction of netbooks with Windows preloaded. Dell claimed that one third of their netbook sales were Linux even when finding the Linux offerings on their website required considerable detective work. Both Dell and ASUS also refuted claims that Linux return rates were higher than Windows.
Another point these articles all conveniently ignore is the percentage of the overall computing market that belongs to the desktop is declining significantly rendering Microsoft's dominance less and less important. There is an x86 desktop version of Android and I expect it's only a matter of time before Google begins marketing it seriously. After all, the new desktop paradigm, regardless of your choice of operating system, bears a strong resemblance to Android and the other Linux distributions that preceded it.
Regardless of which operating system you choose, one fact remains clear: if you're using the new desktop paradigm you are using a system based on Linux desktop design.