The concept of technological convergence goes back a long way, but consumer electronics manufacturers first started to talk seriously about product convergence during the late 80s and early 90s. In 1992 Apple organised a low-profile gathering of over one hundred senior executives from the entertainment and IT industries in Hakone, Japan to discuss common goals and standards. Now, over twenty years later, we have smartphones and smart TVs.
More recently, Canonical’s founder Mark Shuttleworth focussed attention specifically on operating system convergence as it relates to hardware formats, with the launch of the crowdfunded Edge ‘superphone’ project last year.
Canonical and Ubuntu
In an October 2011 blog post, Shuttleworth discussed Canonical’s aim to make Ubuntu an operating system that would unify a broad spectrum of devices. He explained that the Unity shell, introduced in April 2011 with Ubuntu 11.04, was named and developed with this goal in mind.
It wasn’t until 22 July 2013, with Canonical’s announcement of the crowfunding campaign for the Edge superphone, that Shuttleworth first used the term ‘convergence’, with particular emphasis on the Ubuntu operating system. Although many have counted the Edge as a fail (it raised $12,814,286 on Indiegogo – just 40 percent of the $32,000,000 target), it was undoubtedly successful in focussing attention on OS convergence.
Shuttleworth’s promotion of convergence, where all operating system development is based on a common code base, using the same set of development tools to produce an OS and associated applications that run on a wide range of hardware formats, with essentially the same user interface, has prompted other operating system vendors to comment on their own plans in this direction. Shuttleworth’s prediction that Canonical would reach convergence with Ubuntu before Microsoft would with Windows was particularly provocative — and may even be counted as another fail given that Microsoft’s Windows Phone 8.1 and Windows 8.1 now have very similar start screens and increasingly unified application development tooling.
The challenges of OS convergence
The majority of desktop operating systems over the last few years have been written and compiled to run on x86 processors. The problem with moving to a common code base for desktop and mobile is that most mobile phones and many tablets use System-on-a-Chip processors (SoCs) that obey a different instruction set — ARMv7.
In an ideal world, all operating system coding would take place in one high-level language, using a common set of libraries. The code would then be compiled to suit the required instruction set and hardware format. The design of the user interface (UI) would ideally need to be flexible enough to adapt to different display sizes and input methods without compromising usability. Perhaps most significantly, application developers would have an SDK that allowed them to produce applications that would run across all hardware formats.
It may seem sensible for OS vendors to give end users a sense of cross-device familiarity via a common UI design. However, the desktop environment is very different from that of the smartphone or even the tablet. The danger in a converged OS is that the desktop experience may be compromised by veering too far towards the simplified, small-screen, touch-oriented world of the smartphone and the tablet.
On the application front, today’s desktop applications are often criticised for being over complicated and feature heavy, but it’s hard to imagine how a collection of minimalist mobile-centric apps can replace a fully-fledged productivity suite like Microsoft Office or LibreOffice.
Microsoft: Windows and Windows Phone
Microsoft’s dominance in the desktop operating system market (reportedly still accounting for over 90 percent), and its sheer size as a company, almost inevitably results in a certain inertia. Whatever the precise reason, Microsoft has been slow to react to the diminishing importance of the desktop market and the growth in mobile computing. Indeed, it would seem that the company’s board encouraged Steve Ballmer to retire in 2013, after 13 years as CEO, because of his failure to respond quickly enough to these changes. Under new CEO Satya Nadella, Microsoft is firmly focused on a “mobile first, cloud first” strategy.
Starting with Windows Vista in 2006 the recent history of Windows has not been a smooth run. Windows Vista was not well received as a replacement for Windows XP and many users didn’t see the advantage in updating from the 2001-era XP until its April 2014 end-of-support date became too pressing to ignore.
With Windows 8, Microsoft replaced the traditional desktop’s Start button and menu with a tiled Start screen that closely resembles that of Windows Phone 8. Relatively simple ‘modern’ apps were available from the Windows 8 start screen that opened full-screen, with a similar look and feel to their Windows Phone equivalents. Clicking on the Start screen’s Desktop tile opened a Windows desktop (in Windows 8.1, Microsoft has added the option to boot directly to the desktop, which now has a Windows button in the bottom left-hand corner that accesses the Start screen).
Like so many of the phone- and tablet-led UI designs, the problem with Windows 8, and even the improved 8.1, is that the sense of a logical structure, with clear navigation, has been lost in the pursuit of simplification. Some things, such as system settings, are now quite hard to find. Windows 8.x is awkward because it tries to bring together ‘modern’ phone-style full-screen apps (accessed from the Start screen) with more traditional legacy desktop applications.
Although the Windows and Windows Phone operating systems are some way away from full convergence, Microsoft has recently made it easier for developers to create platform-spanning applications. At the recent BUILD 2014 conference, David Treadwell, corporate vice-president of the operating systems group, announced Universal Windows Apps, based on a converged Windows runtime (WinRT), that allow developers to use a common code base to create Windows Store apps that are optimised for smartphones, tablets and PCs. Microsoft is also working to streamline the app store experience — for example, allowing a customer to buy an app once and have it available on their smartphone and their PC, rather than having to visit both the Windows Phone Store and the Windows Store.
Apple: Mac OS X and iOS
Some analysts and industry commentators think that Apple is committed to the convergence of Mac OS X and iOS, and there have been some recent changes in OS X that are plainly influenced by iOS. However, top Apple executives have consistently poured cold water on the idea of a ‘one size fits all’ operating system – sometimes dubbed ‘iAnywhere’.
In January, Macworld published an interview with Apple senior vice-president of software engineering Craig Federighi and Apple’s senior vice-president of worldwide marketing Philip Schiller. In the article, Schiller said: “We don’t waste time thinking, ‘But it should be one [interface]. How do you make these merge together?’ What a waste of energy that would be.” Federighi was equally clear: “To say [OS X and iOS] should be the same, independent of their purpose? Let’s just converge, for the sake of convergence? [It’s] absolutely a non-goal.”
At a MacBook Air refresh event in 2010, Steve Jobs commented on the use of touchscreens for laptop or desktop use: “We’ve done tons of user testing on this and it turns out it doesn’t work. Touch surfaces don’t want to be vertical. It gives great demo. But after a short period of time you start to fatigue, and after an extended period of time your arm wants to fall off. It doesn’t work. It’s ergonomically terrible. Touch surfaces want to be horizontal. Hence, pads.”
Jobs’s view certainly still seems to be gospel at Apple, because Craig Federighi echoed it in the aforementioned Macworld interview: “It’s obvious and easy enough to slap a touchscreen on a piece of hardware, but is that a good experience? We believe, no.”
The best evidence of Apple’s thinking, of course, comes from its actions. At the recent WWDC, the forthcoming OS X 10.10 ‘Yosemite’ was revealed to have tighter integration both with its new mobile counterpart, iOS 8, and with iCloud via the new iCloud Drive. Key to the new OS X/iOS integration is a technology called Handoff, which gives devices that are signed into the same iCloud account and within Bluetooth range of one another the ability to detect what apps are being used, allowing you to seamlessly pick up tasks on another device. It may not be OS convergence as such, but it is OS communication (another new feature was AirDrop support between OS X and iOS devices) and it achieves a similar end: smoother workflows for users of different device form factors.
Google: Chrome OS and Android
Google curates the development of two Linux-based operating systems: Android and Chrome OS. Android is a mobile OS, used on smartphones and tablets; its source code is open source (Apache License 2.0), but the binaries, usually mixed with some proprietary code, are preinstalled on phones and tablets, with updates pushed out at the device manufacturers’ discretion. Chrome OS, a browser-centric operating system that runs web apps, is a proprietary system that, in turn, is built on the open-source Chromium OS. Chrome OS is essentially a web thin client and is primarily used on netbook-like Chromebooks, as well as the set-top-box-like Chromebox and LG’s all-in-one Chromebase. Chrome OS is updated on a rolling release model and supports both x86 and ARM platforms, and so in that sense is already convergent.
Industry observers have criticised Google for this dual-OS approach. For example, back in 2009, Microsoft’s then-CEO Steve Ballmer, speaking at a Microsoft Worldwide Partner Conference, said of Google: “I don’t know if they can’t make up their mind or what the problem is over there, but the last time I checked, you don’t need two client operating systems. It’s good to have one”. This was a little disingenuous coming from Ballmer, because at the time Microsoft was running two completely incompatible client OSs: NT-based Windows XP/Vista/7 and CE-based Windows Mobile.
Google cofounder Sergey Brin has said that Android and Chrome OS will likely converge over time. A year ago, Andy Rubin, Android’s founder and head of Android development at Google, stepped down and his role was taken by Sundar Pichai, previously head of Chrome OS. This naturally led to speculation about the possibility that the two operating systems would merge. However, Google’s executive chairman Eric Schmidt confirmed shortly afterwards that, for the time being, Chrome OS and Android would remain separate – something that new OS chief Sundar Pichai reiterated recently.
Google acquired Android in 2005 by buying original developer Android Inc, and launched Chrome OS in 2011. Arguably Google has more invested in Chrome OS because, unlike Android, Chrome OS is an entirely internal development based on the Chrome browser. Given that Google is a web/browser-oriented company, it still seems possible that, in the long term, Android may fade away and Chrome OS will take over.
Google’s 2014 I/O developer conference is just around the corner (25-26 June), and we’ll be keeping a close eye on the announcements for any further convergence clues.
A flexible concept
OS convergence, at least in the short term, is probably going to be a flexible concept. Developers may find it easier to create user interfaces for different devices that have superficial similarities and to add software interfaces that allow mobile-style apps to run on the desktop, rather than to completely rewrite and unify their code bases.
The end game
With the crowdfunded Edge project, Mark Shuttleworth and Canonical proposed a superphone that was powerful enough to be used as your main computer. When docked with a large desktop monitor and a full-size keyboard, this device would replace your desktop computer and deskphone. Rather than local hard disks, it would use cloud services for bulk storage. Away from the desk, users could access all of their applications and data via the superphone’s much smaller touchscreen, using a smartphone-optimised interface.
Canonical is aware that, rather than force exactly the same interface across all devices, the user interface should adapt to the available input and output form factor, with Unity 8 reconfiguring its layout appropriately. However, no-one outside the company has a clear idea of exactly how this will look on the desktop. The earlier Ubuntu for Android seems promising because, when docked to a desktop display and keyboard, it looks and behaves almost exactly like the current Ubuntu desktop, with the added benefit of integrating the phone functions. The current demo of Unity 8 shipping with Ubuntu 14.04 LTS is very stilted and only shows a sketch of the interface for phone and tablet formats. It would be unfortunate if Canonical chose to move Unity 8 on the desktop too far in the direction of mobile/tablet-style paged icons or tiles.
Eventually, if Shuttleworth’s superphone vision is followed to its logical conclusion, Ubuntu will cease to be a FOSS desktop Linux distribution. This is because the phone, along with its service subscription revenue model, will have replaced desktop computers. Canonical will be in the business of repackaging Ubuntu as essentially a smartphone operating system, which would not be available as binaries for ‘free’ download and installation. End users would only receive a ‘free’ operating system as part of their mobile contract, and would only receive updates and upgrades when their mobile service decided to push them out.
If the superphone were to come to dominate the general computing environment, then operating systems from Microsoft, Apple and others would clearly have to adjust in a similar fashion.
However, Canonical has made it plain that its concepts for Ubuntu running on anything other than the desktop are entirely contingent on it attracting mobile device manufacturers as business partners. The development of Ubuntu for Android is currently on hold, and further work will only take place if a phone manufacturer decides to market a dual-OS phone. Canonical has announced that two phones running Ubuntu for Phones will be released later this year, but these are single-OS products that won’t support a docked desktop mode. A docked desktop mode will be available for implementation once the Ubuntu convergence work has been completed, but this will still depend on phone manufacturers deciding to offer it as a feature.
For a superphone to maximise its utility as a mobile computing platform, docking it to a standard QWERTY keyboard and large desktop display is really only a way point. What’s needed is a way of adding small, portable input and display hardware that offers the ease of use of a large high-resolution display and a desktop keyboard.
For the display, perhaps Google Glass points the way forward. Glass is currently specified as offering “the equivalent of a 25 inch high definition screen from eight feet away” (the actual resolution is 640 by 360). If the apparent distance could be brought down to around three feet with higher resolution, it could perhaps replace a desktop display.
Start-up Atheer Labs has demonstrated the Atheer One binocular glasses, due to ship in spring/summer 2015, that can display 3D images at 1,024 by 768 pixels, with a phone connection and support for hand movement as an input. Or there is the much-feted gaming-oriented Oculus Rift , currently at 1,080 by 960 resolution in the Dev 2 version. Both of these headsets could provide not just a virtual screen, but also a fully immersive 3D experience.
The conventional QWERTY keyboard might perhaps be replaced by a combination of tracked hand gestures and a voice-driven interaction with a virtual assistant such as Apple’s Siri or Microsoft’s Cortana.
Soon, perhaps, not only will people talk to themselves in public, as they currently do with their smartphones, but they will also make strange gestures, and apparently see and react to things that aren’t there. Welcome to the (possible) future of computing.