Original URL: http://www.theregister.co.uk/2009/05/27/desktop_linux/

Getting real about Linux on the desktop

Selective targeting is key

By Dale Vile

Posted in Software, 27th May 2009 09:36 GMT

Reg Tech Panel Few topics in the IT industry are more contentious than the prospect of putting Linux on the corporate desktop. Opinions range from the religious view at one end, promoting a fundamentalist belief in open source as the saviour of mankind, to the reaction of corporate conservatives at the other, dismissing Linux as irrelevant to serious end user computing.

To date, the latter view has predominated, based largely on the assumption that migration of Windows user bases to Linux is generally more trouble than it’s worth. This has in turn perpetuated the significant inertia traditionally associated with the Microsoft dominated status quo.

Quite a bit has happened over the past couple of years, however, that arguably brings the desktop Linux discussion into more mainstream focus. Against the backdrop of initial goodwill from business users, Microsoft fumbled the ball quite badly with Vista, and while it soon to be releases successor Windows 7 is now being pretty well received, the whole experience has broken the Windows spell. For the first time, even the most loyal and accepting Microsoft shops have started to question the logic of simply moving automatically from one Windows version to the next.

We then have some interesting developments in the way applications are delivered to the desktop. With more and more interfaces to corporate systems and online services now being browser or RIA based, and the concept of virtualisation starting to bleed over from the server to the client side of the equation, the logic of the traditional ‘windowing fat client’ is being challenged in general.

It is ironic then that another pertinent development, namely the Mac chipping away at the edges of Windows estates, is actually a backwards step in architectural terms. The Mac desktop perpetuates the fat windowing client model of computing in an even more restrictive way by tying the operating system to the hardware and blocking any thoughts of an open virtual desktop approach. Nevertheless, the entry of Apple into the business mainstream (beyond its historical strongholds) represents another disruptive factor.

IT's the User, stupid

Lastly, we have the whole netbook phenomenon. While the jury might still be out on whether Microsoft or the Open Source camp have won the battle around smaller form factor devices, activity here has raised the visibility of client-side Linux and provided a lot of experience in how to package and roll out Linux-based offerings on a mass commercial basis. Indeed, there has been a lot more focus within the Linux community around issues such as usability and user acceptance, which is quite a departure from the traditional emphasis on perceived technical superiority.

So with all this going on and many in the Linux camp claiming a recession-friendly lowering in the total cost of ownership (TCO) of the desktop environment, has the time now come for businesses to consider a wholesale switch from Windows to Linux clients as the evangelists would advocate?

Ths short answer is

The short answer is no, but primarily because this is the wrong question to ask. The trick when considering the relevance of desktop Linux in the mainstream business context is to avoid such black and white thinking and focus on where Linux might fit as just part of the equation – i.e. consider the value of creating a mixed estate of Windows, Linux and possibly even Mac.

This view came across very strongly in a recent study conducted with input from Reg readers. Sponsored by IBM, but designed and analysed independently by Reg analyst partner Freeform Dynamics, respondents with experience identified general professional users and transaction workers as the most appropriate primary targets for Linux deployments.

The view was that the relatively light and predictable requirements of users such as these meant fewer gotchas in terms of application availability and compatibility.

Furthermore, such users often regard the PC on their desk as simply a tool for getting their job done, so don’t have strong feelings about the flavour of operating system or application software they use. These attributes mean a relatively straightforward migration, allowing the claimed TCO benefits of Linux to be unlocked without prohibitively high overhead and risk.

Unacceptable compromise

Participants in the study identified other groups,such as Windows power users, highly mobile professionals and creative workers, as being much more questionable in terms of targeting with Linux. This is because the number, type and mix of applications upon which such groups are dependent often translates to either high migration costs or an unacceptable degree of compromise in terms of end user capability or experience.

With such factors in mind, the prevailing view is therefore that one of the main keys to success when considering desktop Linux is to analyse user needs, segment users accordingly, and target deployments selectively.

Obvious, perhaps, but it was clear from some of the feedback thata common reason for desktop Linux initiatives stalling has been push-back from users who should probably never have had Linux thrust upon them in the first place, creating a generally negative political climate.

So, while looking at alternatives to the Windows desktop will not be high on the list of priorities for many IT departments, it might be worth at least considering the role Linux might play for less demanding users in particular.

For more insights into the practicalities of desktop Linux deployment in general, we hope you’ll find the report from our study useful, which can be downloaded from the Reg Library.