As mobile devices such as tablets start to erode sales of traditional PCs, there is a wealth of innovation in form factor, as manufacturers and their chip suppliers jockey to control the dominant design. The other battle, of course, is over the operating system for the ‘post-PC’ world. The current situation is well-known ‘ Windows has utterly dominated the PC, with Linux and Mac in minor positions; but Microsoft has failed to repeat that feat in smartphones or, even more importantly, in post-PC categories like tablets.
Here, Android and iOS have scored convincingly, but the sector remains highly immature. The handsets may be tied up by these two platforms, but there are many opportunities for new operating systems to enter the game when it comes to emerging form factors such as cloudbooks (thin client notebooks and tablets relying on the cloud for most of their storage and content); touchscreen Ultrabooks and other tablet/PC hybrids; next generation tablets and so on. One thing is clear ‘ Linux will have a far greater role than it did in the PC and Microsoft will struggle, as the early pains of Windows RT have illustrated, to establish significant share.
The point was reinforced by Microsoft’s former best friend, Intel, whose CTO Dirk Hohndel ‘ also a Linux kernel developer ‘ told last week’s developer conference that, thanks to Android and the rise of Chromebooks, Linux was now the leading end user OS.
Windows will continue to be important to some users because of the legacy base of applications and customer habit ‘ the same goes for Mac OS, but with its integrated hardware/software approach, Apple has to ensure both its form factors and its operating systems remain relevant, a tough feat as users move towards the cloud. The iPad has been a big step to keeping iOS as a major force in post-PC devices, and future evolutions of the MacBook Air (Apple’s Ultrabook equivalent), plus possible convergence of the two OSs, will be significant. However, some of the Apple magic has faded, and whether it can regain it will depend in large part on how it fills the hole in its device/OS offering where a full cloud device should be.
Microsoft has the same issue, and that gradually widening gap is where many new OSs will seek success. Mozilla, Baidu and Canonical are important contenders, as are some effective Android forks, like Amazon and the newest kid on the block, Cyanogen. Of course, Google also has Chrome in this space, and the option to converge that with Android in future if that would help solidify its position.
The element all these contenders have in common is Linux, which is the winner whichever organization takes the prize. At this week’s LinuxCon event in New Orleans, there was intense emphasis on the mobile and embedded versions of the open source kernel. In a panel debate including creator Linus Torvalds, there was agreement that the proliferation of Linux-based mobile devices has promoted the embedded branch of the movement to a far higher priority. Torvalds said: ‘If you look at just the last merge window, most of the actual code was on the embedded side. That’s maybe because the embedded side has all these wild and wacky devices, and most of the kernel code these days is device drivers.’
He added: ‘The reason Linux runs really well on cellphones is that cellphones grew up. They’re already thousands of times more powerful than the original machine that Linux came to be on.’
The sentiments were echoed by Jim Zemlin, executive director of the Linux Foundation, which hosts some mobile Linux implementations such as Tizen (the combination of Intel’s end of its MeeGo project with Nokia, and another OS, LiMO, with Samsung as its main backer). He said: “You see this tremendous acceleration of code happening due to an incredibly broad device enablement. Nobody is making anything these days without Linux, unless your name is Microsoft, Apple or BlackBerry.’ Indeed, of the eight contributors to the kernel in the past year, four were primarily coming from the mobile and embedded area (the list was Red Hat, Intel, Texas Instruments, Linaro, Suse, IBM, Samsung and Google).
For now, of course, Google’s Linux-based platforms are the most important in defining the early post-PC landscape ‘ however far Android, in particular, may drift from the ‘pure’ Linux kernel. It is proving hard for new operating systems to work around Google. Jolla, the Finnish start-up which is developing its Sailfish Linux platform around Nokia’s discarded MeeGo project, announced this week that it would support Android apps, a move already made by BlackBerry and others. Of course that addresses the problem of a limited supply of applications that any new OS faces ‘ however cloud-based they may be, they have to accept that many users will still be clinging to downloadable apps for some time to come. And it allows a customer to migrate to a new platform while still holding onto their favorite Android software. All the same, it also begs the question ‘ why bother, when apps and web services are the key element of the OS for most consumers, and Google plows more development dollars into its Linux variants than any other contender?
Some are taking a different approach, looking to fill Android’s gaps and improve the experience without departing from the platform altogether ‘ though in the process they can effectively provide an alternative Android UI and revenue base to Google’s own. Amazon is the best example, but Samsung and HTC have also worked on distinctive user experiences that drive their own content sales.
A different approach to exploiting Google’s weaknesses comes from new start-up Cyanogen. The organization has been known in the Android community as an open source project, CyanogenMod, but has now raised $7m from Benchmark Capital and turned into a fully fledged company with a team of 17 developers and led by CEO Kirt McMaster. It initially aims to focus on aspects of Android that Google has neglected, in its opinion, but could evolve into a separate platform over time. Co-founder Steve Kondik said: “There’s almost this entire developer ecosystem around it. We’re going to do a lot of work around the user experience. You’re going to see us building secure services. To get more people to use it, we have all these little touches planned.”
The initial step seems modest ‘ releasing an installer app which will create an easier path to update the handset’s ROM, and will restore factory settings in the case of problems, as well as resume installation during power loss. But McMaster has a bigger vision. He said: “Benchmark and others are excited that this has a shot at being the third mobile phone OS. It’s not going to be Windows Phone, it’s not going to be BlackBerry. People don’t realize the scale we have.” There are about 8m people who have installed CyanogenMod itself, or are running custom Android ROMs based on it, such as MiUL, which is backed by Chinese handset maker Xiomi.
More realistically perhaps, another route is for Cyanogen to trade on being the only major Android supplier without its own hardware, and develop software for OEMs which want to differentiate their experience from the vanilla Google one.
Others, like Canonical, are braver, avoiding Android altogether; and relying on converting a powerful base of users to web services and HTML5 rather than walled app stores (in the case of its converged Ubuntu platform, with a heavy focus on business). Most importantly, it is not trying to outdo Android at its own mobile game, but to move the debate forward to a truly post-PC platform, delivering a common, web-based user experience to all kinds of devices including TVs and cars.