Quocirca's Straight Talking: Choosing Desktop Linux

Is it 'out of the frying pan and into the fire' for the open source OS?

If you worry about the amount of time, effort and money you are spending trying to keep Windows secure or have concerns about the cost and complexity of Windows licensing, why not just ditch it and go with another option?

Many of the 8,128 respondents to an online survey Quocirca conducted recently were thinking along these lines and considering Linux as an alternative.

The appeal of Linux in this context is not surprising. With its 'free' open source status and one of its claims to fame being high security, it would appear at first sight to deal with two of the most common gripes about Windows. Yet the majority of those who have looked at the Desktop Linux option are still either at the experimental or the limited-deployment stage.

If Linux was the 'magic bullet' for making all Windows-related problems disappear, we would have expected a lot more commitment than this. Of course, in reality it is nothing of the kind. While a move to Linux might in theory tackle some of the challenges at an operating system level, it is highly likely to create a whole bunch of other problems along the way.

The feedback we have received confirms this, with the following being identified as common barriers to Desktop Linux adoption:

- Software availability and compatibility issues
- Usability, end-user acceptance and resistance to change
- The cost and challenge of end-user training and support
- The cost and challenge of porting bespoke Windows applications
- A frequently encountered dependency on Microsoft Active Directory

In short, there is a risk that a 'quick and dirty' move to Linux will lead to users being unable to run the applications they need or exchange information confidently with Windows users, which in turn is highly likely to cause resistance or even a severe backlash.

To minimise this risk, it is necessary to allocate enough resources and budget to prepare for and manage the migration in a thorough and structured manner. This includes paying adequate attention to user motivation, training and support, as well as obvious things like making sure that there are appropriate replacements for key Windows packages and that business-critical bespoke applications developed in the Windows or MS Office environment can be ported across.

Before you know it, you are into a major project with elements of software selection, application development, operational level re-configuration and testing - not to mention all of the personnel-related tasks such as the cross training of users and support staff. Business managers are also likely to put IT in the hot seat when it comes to minimising disruption to the business.

So is Linux on the desktop a complete non-starter?

Well, that depends on the nature of your user base, their application requirements and your existing IT landscape in terms of the operational and security framework.

In most cases, the cost/benefit equation is unlikely to pan out in favour of a wholesale migration to Desktop Linux in the short to medium term. Even if the return on investment calculations said that the migration costs could be recouped from the resulting operational and licence savings over a reasonable period, there is an ongoing residual consideration. Application availability and compatibility, the biggest single issue with Desktop Linux today, could in many cases continue to constrain choice and impact the ability for IT to support the business optimally.

For those interested in looking at Linux seriously as a desktop option, it is therefore necessary to be selective about its deployment. The chances are it will work best for user communities that have relatively straightforward requirements, e.g. those predominantly accessing enterprise applications such as ERP through a browser. The greater the depth and breadth of Windows functionality already in use by a group of users, the less likely it is that Linux will be a viable, cost-effective option.

Given this, it's important to do your groundwork, assessing both the level of Windows dependency and the ability to meet user needs with Linux-compatible applications on a department-by-department, workgroup-by-workgroup or role-by-role basis. Of course, there are tools and workarounds that allow some Windows applications to be run in a Linux environment or in thin-client access mode, but if you end up employing these excessively you are unlikely to be meeting your original objectives.

The potential benefits and issues associated with Desktop Linux are discussed more fully in Quocirca's report of the above-mentioned survey, which lays out an approach for assessing whether Linux is appropriate in the various parts of your organisation.

The bottom line is we would not encourage or discourage organisations from looking at Linux for desktop computing, but we would urge them to approach the prospect with their eyes fully open so they don't end up jumping out of the frying pan and into the fire.