Is Linux on the desktop dead?
6th Jun 2011 | 10:15
People are quitting Windows, but not for open source
Is Linux on the desktop dead?
Since the turn of the century, pundits have been telling us that this will be the year of desktop Linux - and unless something really extraordinary happens, this will be the 11th year that it hasn't happened.
While the open source operating system has made significant inroads into the server and mobile markets, its desktop market share is much the same as it was a decade ago: about 1 per cent. Netbooks gave Linux a shortlived boost, but the world of PCs still largely belongs to Microsoft - with one Unix-spun exception.
Apple's Mac OS X is doing what Linux advocates long dreamed of - it's actually taking market share from Microsoft. Can Linux learn anything from Apple, or does open source need its own Steve Jobs? In short, can Linux win?
Simply the best
Linux's mission was simple: to make the world's best desktop operating system. If you didn't like Windows' system requirements, Microsoft's business tactics or the small print of the licence agreement, Linux was waiting with open electronic arms.
"There is a better way," the Linux people said. And eventually, consumers realised that the Linux people were right - they could move to a platform that offered an alternative to Windows. A platform where malware, spyware and other irritants weren't a factor. A platform designed to bring man and machine together in harmony.
There was a problem, though: those consumers bought Macs. The numbers are staggering. Apple's Unix-derived Mac OS X was shipped on 1.99 million Macs in the third quarter of 2010, according to IDC. That gave Apple an unprecedented 10.6 per cent of the US PC market.
Year on year, Apple's sales are up 24 per cent compared to just 3.8 per cent for rest of the industry, and that figure doesn't include iPads or other iOS devices. Why Macs and not Linux? In 2005, industry analyst Gartner said that desktop Linux was "up to five years away" from mainstream use.
Five years later, Linux still isn't there. Michael Silver, Research VP and Distinguished Analyst with Gartner, explains:
"No, Linux didn't become big on the desktop, and yes, Apple does appear to be causing Microsoft more problems [than Linux] these days. I think this underscores the issue that cheaper does not solve problems. Organisations are looking for solutions that provide value and solve their problems, and desktop Linux does not really do that when you have Windows and Office - real Microsoft Office - to run."
As Silver points out, there's a growing trend for people to influence the choice of computers they're given, "and while most can't officially select their platform yet, users want devices they perceive as increasing their productivity. And most would rather drive a luxury or sports car (the Mac perception) than the Yugo (their perception of Linux). Linux still has a perception of being unapproachable, and it doesn't have the marketing budget to buy the cachet that Apple does."
Marketing is certainly part of it - as the world's second biggest company, Apple can afford to splash out on the odd advert, and the firm enjoys press coverage that often borders on fawning. However, there's more to it than a massive marketing budget and Steve Jobs' famous Reality Distortion Field.
Perhaps the problem is that everybody who's interested in Linux is already using it.
OSes aren't interesting
To most people, operating systems simply aren't interesting - they would no more change their PC's operating system than they would change the engine in their car. Because of this, Linux faces an exceptionally difficult situation: to get significant market share, it needs to be pre-installed on shipping PCs.
Unfortunately, to be pre-installed on shipping PCs, it needs to have significant market share. Catch 22. "Yes, the OS is the unexamined part of the computer system for many people," Gerry Carr says. Carr is Director of Communications with Canonical, the creator of Ubuntu.
"Actually, it's more interesting than that," he continues. "Windows is so embedded, it's become invisible to users. Ask them to tell you something about Windows and they will start to talk about the word processor and the spreadsheet thing. Microsoft Office is actually the visible part."
Carr isn't convinced that users think Linux is a Yugo. He's not sure they have an opinion at all. "My read on it would be that Linux doesn't have an image at all among the vast majority of users," he says.
"There is a legacy, sure, of people in the industry who tinkered with it and didn't like it. And certainly within the IT industry, there are many who, from professional obligation or otherwise, hold and promote that opinion. But [if you] sit in a room with mainstream users, you'll find that they don't know what Linux is, never mind have negative associations with it."
So would OEM support help? With the honourable exception of Dell, which threw its considerable weight behind Ubuntu in 2007, few mainstream consumer PC firms even acknowledge Linux's existence in the consumer sector - and even Dell's Ubuntu page is looking sparse these days, with just one Studio XPS available with Ubuntu pre-installed. You'll find Red Hat Enterprise Linux on its high-end workstations, but at consumer level it's Windows all the way.
Linux just isn't something the average computer buyer is likely to encounter. The places where mainstream users now buy their PCs - Tesco Extra, Amazon, Currys Digital and so on - are largely Linux-free. If you enter 'Linux' into the search box on PC World's website, you'll be told 'Sorry, we found no results for your search.'
To add insult to injury, the site then asks if you'd like to look at an iPad.
HP and Linux
Can Linux thrive when OEMs don't offer it? Amanda McPherson is Vice President of Marketing and Developer Programmes at The Linux Foundation, a non-profit group that aims to raise awareness of and promote the use of Linux.
"There is a lot of pressure on OEMs to use Windows, as we have seen in the netbook market," she says. "But we see traction in different areas. HP, for instance, has announced that it's shipping Linux on desktops in greater quantities than ever before. We've also seen the use of Linux as a desktop client rise with developers. We assume those kinds of use cases will continue."
The Linux HP desktop will be shipping is its own WebOS, which it acquired when it purchased the smartphone maker Palm. In March, HP CEO Leo Apotheker told Bloomberg Businessweek that from 2012, "Every one of the PCs shipped by HP will include the ability to run WebOS in addition to Microsoft Corp's Windows."
Apotheker says that WebOS will "become a very broad and very massive platform," shipping on 100 million devices per year.
State of confusion
That could be a significant step forward for Linux on the desktop. However, it could also be hopelessly confusing. As former Apple executive Jean-Louis Gassée wrote on Monday Note, the biggest PC firm's Windows portfolio might look something like this in 2012:
"Intel-based PCs and laptops running the 'mature' Windows 7. ARM-based laptop and netbooks on Windows 8. Tablets using a version of Windows 8 with a touch interface. Some, but not all, "will include the ability to run WebOS in addition to Microsoft Corp's Windows." Simple, easy to understand. Can you imagine the sneers at Apple and Google when looking at such a picture?"
"PC manufacturers will not solve all, or any, problems by shipping Linux," Carr says. "PC manufacturers will ship what their customers are asking for. We haven't generated sufficient directed and determined demand in a broad enough mainstream market to convince them to [ship Linux] in great numbers. I don't think it is going to be a supply-led revolution. It needs to be demand led, and we'll be spending considerable time boosting that."
Canonical thinks it has the solution to mainstream users' lack of interest in desktop Linux: make something that's as attractive as, or more attractive than, Mac OS X or Windows 7 - to borrow Michael Silver's analogy, make a sports car rather than a Yugo. Canonical calls its sports car Unity.
Where Linux is winning
Unite and take over
Unity is the new interface for Ubuntu, and it's a deliberate attempt to bring good looks and user-friendliness to a platform that isn't exactly famed for its fantastic user interface design.
It's proved controversial; it was developed in-house, not by the wider community, and it's very different from earlier versions of Ubuntu and other GNOME-based distributions. By betting the farm on Unity, Canonical could alienate some of its existing user base in the hope of attracting lots of new users.
It's the Apple model: instead of asking customers what they want and then trying to meet every single request, Apple simply creates what it thinks works best and then releases it.
"You can't just ask customers what they want and try to give that to them," Steve Jobs said in 2005. "By the time you get it built, they'll want something new."
It's a big risk, but Gerry Carr for one is bullish. "We are trying to do something that has never been done before," he says. "We have taken the evolution of Linux as a mainstream OS further than anyone else; by any reasonable estimation, it's already the third biggest operating system in the world. Unity gives us the chance to arm [users] with a look, feel and experience that will enable them to convert other users to the benefits that they enjoy.
"I think it will be the first time you have someone who looks at a Linux-based OS, if you want to describe it that way, and see it on looks alone as something they want to use. That's a true first and we think it has the potential to shift the market."
Canonical isn't the only firm that thinks a more user-friendly Linux could be a big deal. Google has Chrome OS, a Linux-based operating system for netbooks, which is due for release this year.
Rise of the robots
Can Linux win? In the form of Android, it's winning already. Smartphones now outsell PCs, and as our computing becomes increasingly mobile, we're more likely to use an Android device than a Windows one.
"I agree, a traditional desktop is becoming less and less the primary computing environment that people interface with," Amanda McPherson says, adding - with some understatement - that "Linux is doing quite well."
According to Canalys, in Q4 2010, one in three smartphones ran Android. Google's platform had 32.9 per cent market share, Nokia 30.6 per cent, Apple 16 per cent and RIM (BlackBerry) 14.4 per cent. Microsoft trailed far behind with just 3.1 per cent.
Android is growing in the tablet market too. Apple's market share has gone from 95 per cent to 75 per cent, and many Android-powered rivals have yet to ship. Strategy Analytics predicts that Apple's tablet share will drop below 50 per cent by 2013, with Android doing most of the damage.
The mobile market is very different from the PC market, and Android owes its success to those differences. Windows isn't dominant on mobiles, and Android isn't a software-only proposition: Google doesn't have to persuade people who buy an HTC Desire HD to switch OSes, because the HTC ships with Android preinstalled.
In fact, you could argue that Android is mainly about hardware; you choose Android because of the device first and the operating system second.
Android looks as if it could represent the ultimate triumph of Linux over Windows, but we've been here before and it didn't end well.
For a while, netbooks seemed like the Trojan horse that would let desktop Linux sneak into the mainstream. In 2007 and 2008, netbooks ran Linux. Windows XP was being retired and Vista was too big and costly to be worthwhile on cut-price machines. By the end of 2007, Asus had shipped around a million Eee PCs, all running a customised Xandros Linux.
Other manufacturers followed: Acer with Linpus Lite, Dell with Ubuntu, HP with SUSE and so on. It wasn't long before Microsoft noticed, and with Windows 7 some way off, it hastily resurrected Windows XP and offered PC firms big discounts.
By December 2008, nine out of ten netbooks on sale were running Windows; today, most shipping devices run Windows 7 Starter Edition.
What happened? Lenovo's Matt Kohut told Tech.Blorge, "There were a lot of netbooks loaded with Linux, which saves $50 or $100 or whatever it happens to be… [and] there were a lot of returns because people didn't know what to do with it. Linux, even if you've got a great distribution, still requires a lot more hands-on than Windows. So we've seen people wanting to stay with Windows, because it just makes more sense - you take it out of the box and it's ready to go."
There are several reasons why that bit of history seems unlikely to repeat itself, though. Firstly, and perhaps most importantly, Microsoft doesn't have a familiar OS it can resurrect to fend off the Android horde: Windows Mobile never had the traction Windows had on the desktop, and by the time Microsoft canned it in 2010, it was looking hopelessly outclassed.
The new Windows Phone is a vast improvement, but OEMs aren't embracing it the way they did Windows on netbooks - and Windows Phone's market share is falling, not rising.
According to ComScore, Microsoft's share of the US smartphone market dropped from 9.7 per cent to 8 per cent between October 2010 and January 2011. Microsoft claims to have shipped - not sold - 2 million Windows Phone devices during that period, but so far at least it doesn't appear to be stopping the decline in Microsoft's mobile market share.
The much-hyped deal with Nokia to make Windows Phone the Finnish firm's platform of choice may address that, but not in the short term: the first Nokias running Windows Phone may not arrive until 2013. By then, Android could well be dominant.
If things look bad for Microsoft in the mobile market, tablets must be making it downright miserable: as with netbooks, the rise of the tablet appears to have taken Microsoft completely by surprise, and there isn't an OS to put on them.
If you take the smartphone and tablet markets together, you'll see that traditional roles have been reversed. Here Microsoft is the minority player, and most of the mobile market is split between two Unix-derived operating systems: Apple's iOS and Google's Android.
That's unlikely to change in the short term, and it could have long term effects too. "I think these changes have made people aware of a couple of things," Carr says. "First, that the choice of OS matters - this awareness is on mobile devices, not necessarily PCs - and secondly, that the choice need not be provided by Microsoft. That's critical to opening up the world to some genuine competition based on quality of the computing experience, not ability to dominate the supply chain."
That doesn't mean Android's success will necessarily translate to desktop computers, though. "I've been surprised by how little bleed there has been from the mobile devices to the fuller PC experience," Carr says.
"Android's success on smartphones doesn't pre-suppose success on netbooks, for instance. The main driver of that will be marketing and delivery […] the success or failure of WebOS as an instant-on product, Chrome as a cloud netbook OS or Android as a phone OS will have little effect on the success of Linux or any other OS in the core PC market […] we aren't going to get a free ride onto people's primary machines because of the choices they have made on their tablets and phones and other devices. We still have to work hard to convince users and the industry of the value of an alternative."
The success of Android and iOS could help desktop Linux indirectly, though. Smartphones and tablets are helping usher in what Steve Jobs calls "the post-PC era", where always-on mobile devices connect to the cloud.
The growing capability of HTML5 and runtimes like Adobe AIR mean the underlying OS is becoming less important. "Android, Chrome OS and WebOS tablets running Linux are poised to become an important part of the 'desktop' equation," McPherson explains. "It just may not look like what we today consider to be the desktop."
HTML5 introduces important features that let browser-based applications act like stand-alone software, and browsers are evolving alongside it: the latest versions of IE, Chrome and Firefox include hardware acceleration, which uses your graphics card to deliver better in-browser graphics and video.
Over time, HTML5 should enable developers to provide in-browser applications as good as, or better than, traditional desktop software.
Microsoft's vision of 'three screens and the cloud' is nearing reality, although those might not all be Microsoft ones: provided your browser can handle HTML5, it doesn't matter what the underlying OS is.
"I think there will be a range of low cost, small screen PCs that will be popular, emerging on a range of architectures," Gerry Carr says. "I think there will be a critical mass of Linux use as a technology that will bridge mobile devices, PCs and cloud computing."
That won't be a single flavour of Linux - it will be a cocktail: Ubuntu on a desktop, Android on a tablet, a sprinkling of WebOS and a dash of Chrome OS too. "The platform environment today isn't a zero sum game,"
Amanda McPherson says. "Since Chrome uses Linux, if Chrome does well it benefits the Linux platform - just like Android or WebOS. Since Linux is used in so many ways by so many companies and projects, it doesn't make a lot of sense to focus on one application."
The future of Linux is on tablets, smartphones, PCs and perhaps devices still deep in development. "This is desktop Linux," Carr says. "It's just that our desktops will work and act differently."
Liked this? Then check out Linux needs rebranding
Sign up for TechRadar's free Week in Tech newsletter
Get the best tech stories of the week, plus the most popular news and reviews delivered straight to your inbox. Sign up at http://www.techradar.com/register