Given all the dismal market-share statistics so lovingly reproduced by Microsoft and like-minded partners, it's not entirely surprising to see observers declare that the dream of the Linux desktop is dead.
After all, we've all seen the reports, month after month, suggesting that Linux's market share is stuck around 1%. For September, in fact, Net Applications says it was even lower, with a figure of .85%. Taken as evidence of Linux's success or failure on the desktop, such news appears grim indeed.
The problem with such statistics, however, is that they are not an appropriate way to measure the uptake of a free piece of software. In general, such market-share data is based on sales figures: "X copies of Windows sold per month translates into Y market share for the operating system," for example.
But how is Linux obtained? Unless it's purchased with paid support--surely a minority of cases--it's typically a free download, with no sales transaction involved.
Certainly there are sites, such as DistroWatch, which keep tabs on page hits for each Linux distribution, thereby providing a snapshot of market interest, if nothing else. Canonical, meanwhile, recently started tracking original equipment manufacturer (OEM) installations of its Ubuntu Linux.
For the most part, however, it's almost impossible to know how many copies of Linux are in use on desktops around the world. Even tracking downloads won't give an accurate picture because of the ease--not to mention the complete legality--of sharing copies of the software. This is clearly a data-collection challenge that needs to be addressed.
But is Linux really dead on the desktop? I don't think so. Here's why.
1. It's in More than a Third of Businesses
The most obvious news from the Linux Foundation's study last week was that Linux is blowing Windows away on servers, and will continue to do so in the coming years. Servers and embedded uses, indeed, are where Linux currently does best, so that wasn't entirely surprising.
What received much less fanfare--and yet in many ways was much more notable--was the proportion of large businesses that are using it on the desktop. The figure cited, to be specific, is 36% of large enterprises currently using Linux on the desktop, while another 12% are evaluating or planning deployments over the next year.
To me, that just doesn't sound like the business world's response to something that is dead.
It's also important to realize that Net Applications' measly .85% figure is by no means the only one out there--it's just the best publicized.
W3Counter, for one, gives a 1.5% share for Linux in September--a marginal improvement over the figure from Net Applications, but still a little better. Wikimedia's Visitor Log Analysis Report for this summer, meanwhile, puts Linux at 1.9%, while an O'Reilly Media estimate puts it as high as 10%.
Again, though, the bottom line is that there currently isn't any way to know exactly how many desktop users Linux has. Drawing conclusions from these figures is like conducting a census by counting only the people who happen to be out on the street at any given time.
3. The Ubuntu Factor
Then, of course, there's Ubuntu, which has long been the most popular and consumer-friendly Linux distribution out there. I have to say that the excitement over the recently released version 10.10--Maverick Meerkat--surpassed anything I have seen before.
Not that that's surprising, mind you -- Ubuntu is the biggest "brand name" in the Linux world, and it's far and away the most popular distribution on DistroWatch. This latest release, too, is clearly the most consumer-friendly Ubuntu so far, and multitouch capabilities are now a part of its netbook edition. Canonical's R&D efforts for Ubuntu also suggest that there's great promise for the future.
With Ubuntu, so many of the issues that have dogged Linux on the desktop in the past--such as drivers--are going away. I believe Ubuntu is Linux's best tool for extending further into the mainstream, and recent data shows that's already happening. Just today, for instance, Australia's Sydney Linux Users Group said that the operating system is visibly reaching past the Linux community and into wider mainstream usage than ever before, with schools and government agencies among the country's leading users.
"I'm noticing Ubuntu gradually gaining acceptance as a desktop environment with more and more enterprise-sized companies, government departments and non-profit agencies adopting it as an acknowledged part of their networks," the group's secretary, Melissa Draper, said.
None of this is to say that Linux doesn't still face challenges, of course. Inertia, first and foremost, is what maintains Microsoft's monopoly, I believe, along with Redmond's active efforts to preserve it. Then, too, there's Linux's fragmentation, lack of a single corporate sponsor and lack of marketing--the all-important tool that keeps Apple in the game, in particular.
There's also the question of whether the desktop will even matter in a few more years, as cloud and mobile computing increasingly take hold.
But for now, to say that Linux is dead on the desktop is a great exaggeration. It's in more than one-third of large enterprises, and it's on more home computers than we'll ever know, thanks at least in part to the emerging Ubuntu brand. It also boasts more advantages for businesses and individuals than either of its main competitors.
Marketing may be helping Windows dominate in the short term, but ads can only take Microsoft so far--particularly given the financial and security costs of its products to users. It won't be much longer before Linux is generally viewed in the mainstream as a compelling and attractive alternative.
This story, "Reports of Linux’s Desktop Death Are Premature" was originally published by PCWorld.