Software > Open Source

Paula Graham's Open Source Update - November 2011

By Paula Graham, Fossbox

Computanews' regular open source correspondent Paula Graham of Fossbox wonders if the desktop has long to live, how the cloud is being dealt with by the FLOSS community and the differing views on the way Linux distros are going.

The Desktop – Dead or Alive?

Every now and then, there's a flurry of abuse between supporters and detractors about the Linux desktop. Predictions of the always-elusive year of the Linux desktop from the Linux community alternate with much cackling and crowing  from the proprietary community that the end is nigh. In the second year of the Linux Desktop Conference things are looking good in many ways – the Linux desktop is now mature and is overtaking Mac in the desktop market but the community has also seen an unusual level of storm and strife. There are many factors involved but, by and large, they all track back to the 'cloud'.

The chaotic and fast-moving desktop anxiety-fest is not limited to Linux, of course. It's not at all clear what consumers will make of the interaction between personal devices and software-as-a-service across the board. Some predict that mobile and tablet devices will kill of the PC in short order as data moves entirely onto the cloud. Others (myself included) are pretty adamant that most of our data will remain on personal PCs and laptops with ancillary use of cloud storage and services. Linux is seen as more vulnerable because it has not yet got a grip on the tablet market. Will 'cloud' kill the Linux desktop – and even the desktop in general?

Convergence or Hybrid?

Many predict that consumer computing will move entirely onto the cloud but this is unlikely to happen immediately. Some argue that the ARM processor-based systems are now as powerful as desktop computers were a couple of years ago – this is true, but current chip technology is reaching its limits so until quantum or molecular computing arrives, the ARM chip is probably not going to satisfy power-users. I love my Maemo smartphone and my Android tablet – but I'm certainly not planning to sit down to do a day's work on them. Not am I planning to alienate all my data to remote services with dodgy reps for abusing customer data.

The Free Software community is strongly committed to a hybrid solution where software-as-a-service (SaaS) distributes data held by users with subscription models guaranteeing user privacy and ownership of data. This is in stark contrast to the approach of most of the mainstream players who barter wholesale surrender to Big Brother for free consumer services. This is, of course, a deal that is broadly accepted by consumers but far less so in the corporate sector. Some of the cloud service providers also leave a lot to be desired in terms of customer service. On top of that, cloud services are prime targets of cracker attacks and, recently, have also experienced a fair amount of down-time.

Service and hardware providers are nevertheless starting to merge as Google scarfs up Motorola's patents, Nokia ousts Linux Meego in favour of Windows 8 (if they can ever drag people kicking and screaming from XP) and Google is about to have another go at foisting Chrome OS on to the populace. Virtualisation and cloud services are changing rapidly but, again, in the corporate world, virtualised hybrid still seems to be holding as the strategy of choice. So what does all this mean for the Linux desktop? Cloud-ready Apple products cost an arm and a leg so there may well be a niche for Ubuntu here among those who value control.

Ubuntu's (dis)Unity

Ubuntu has been undisputed champ on distrowatch for years – and it's just dropped to #2 behind Linux Mint (based on Ubuntu). This is undoubtedly fallout from the Unity desktop saga but, let's be clear, the Ubuntu community is still growing rapidly and still represents the likely future of Linux on the desktop. Mark Shuttleworth (of Canonical, the company behind Ubuntu) sets out this future:

"By (release) 14.04 LTS Ubuntu will power tablets, phones, TVs and smart screens from the car to the office kitchen, and it will connect those devices cleanly and seamlessly to the desktop, the server and the cloud."

The Unity desktop will be a kind of 'HQ' for your mobile world but, unlike Google OS and Apple's approach, your world will be synced from your devices to the cloud and back rather than from the cloud to your devices one-way. Ubuntu is already offering the 'Ubuntu One' range of cloud services, including storage, which come pre-integrated into Unity along with a social networking 'deck'. This is compatible with the wider Free and Open Source hybrid approach and Canonical hopes it will bring new users to the highly convergent Ubuntu mobile/desktop.

A losing battle?

Many prognosticate a losing battle for Ubuntu (and the handful of other convergent Linux-based mobile OSs) up against Google and Apple. Nevertheless, Google's first effort to move us entirely into the cloud bombed, so it's possibly a question of whether market clout will finally determine the issue or whether users actually will cling to controlling their own data and thus be moved towards the Open Software Service Definition (OSSD) hybrid model. All the indications are that corporate users will care much more about this latter than consumers. It may also boil down to a question of how fast Ubuntu can bring a Unity-based tablet/mobile device to market. Unity already works on a touchscreen but you still need to attach a keyboard to get full functionallity – Canonical estimates a year to bring a truly mobile OS to market – and that may be too long.

Whatever the outcome in the global marketplace, the first moves towards a convergent Ubuntu desktop has resulted in an epic row. Canonical's user testing indicates that the new desktop is very popular with non-technical users and this is borne out by feedback from users at Fossbox workshops. Non-techies love the new desktop. The problem is with the established Ubuntu community where there is significant disaffection in the ranks.  This is partly due to aesthetic luddism and partly due to the magnitude of the change resulting in buggy implementation in the current version of Ubuntu.

Lack of customisation

However, much of it is due to the lack of customisability in Unity. 'Traditional' Linux users love Linux because we can tinker, personalise, customise, and generally feel like we're in control. The traditional Windows/Mac user, however, wants to be presented with something ready for them in a polished state. This is the core of the dilemma. In order to make Linux more appealing to non-techies, it has to be less appealing to techies. Hence the sudden boost to various Ubuntu-based desktop OSs such as Mint and Bodhi which don't come with the sternly un-customisable Unity shell - and you still can lash the cloud services on.

A cloudy Future for the VCS?

So, will VCS users be happy relocating all their data and software to 'cloud' services and using a 'dumb' hand-held device? Unreflective adoption of free cloud OSs and services in the VCS could lead to spectacular levels of lock-in as well as poor DP practices and loss of data. At the same time, FLOSS is producing some really good SaaS on the OSSD model such as Feng Office and the various Drupal derivatives including CiviCRM and Open Atrium. Civi, in particular, has found substantial acceptance in the VCS. There are also a number of integrated cloud applications for surveys, mailouts etc.

Between OSSD collaborative tools and the convergent Linux desktop, the VCS could follow the corporate sector's example and choose to move towards hybrid cloud integration whilst retaining control over data and privacy. Given its history of lock-in and the current squeeze on resources, it's perhaps more likely that the VCS will have far less control over its data and its ICT in the future. VCS providers and users are already moving towards Google or Microsoft cloud services – Google for those without money for tech support and Microsoft cloud from VCS providers on the subscription model. This may be a serious mistake, but it will probably be a couple of years before anyone can tell for sure. The hybrid Linux/cloud desktop approach may yet be the way to go and is at least worth evaluating – it's free!