Virtualization Technology News and Information
The Turing Test for VDI

What do virtualization executives think about 2009?  A Series Exclusive.

Contributed By Gordon Riley, co-founder, Tranxition

The Turing Test for VDI

2009 will see IT struggling to achieve at least some of the cost benefits of VDI without imposing major disruption and lost productivity on the user base. The elephant in the VDI planning room is how to get from today’s highly personalized fat client environment to at least a partially VDI-based infrastructure. The proponents of VDI have done a credible job of delineating the benefits of VDI and the technical issues of deploying a VDI-based environment are becoming well understood but so far no one is talking about the “soft” issues around moving to VDI. As organizations begin desktop virtualization pilots or begin implementation in select areas of their organizations they will be forced to confront this issue head on.

For users to accept VDI as a reasonable facsimile of their traditional fat client desktop, VDI must pass the “Turing Test” of virtualization.  The Turing test was first proposed by the famous computer scientist Alan Turing in his 1950 paper “Computing Machinery and Intelligence”. The essence of the test is that if a human cannot tell whether they are conversing with a computer or a fellow human, then the computer is said to have passed the test of appearing intelligent.

In virtualization terms, if IT users cannot tell whether their environment is based on a traditional fat client or on some form of virtualized desktop, then virtualization has succeeded in being an acceptable replacement for the traditional fat client environment. While there will be some visual telltales that prevent a complete masking of the virtual environment, the essence of the test is whether the user will be as productive in the virtual desktop scenario as they have been in their traditional environment. This acceptance will be based on many factors such as launch (boot) time, performance and overall responsiveness, but will also be heavily dependent on the virtual environment being familiar and comfortable. If the user is forced to move from a highly customized, rich personal environment to a generic “greenfield” (a term from the default wallpaper installed with Windows XP) desktop devoid of their personal customization, virtualization will fail the user acceptance “Turing Test”. The  loss of the familiar may cause VDI pilot projects to “fail” over soft issues that do not necessarily reflect the technological underpinnings of virtualization.

Pilot projects must therefore plan to bring the user’s rich environment along into the virtualized world rather than simply rolling out baseline generic desktop images. As Microsoft’s experience with Vista has shown, users will not welcome change that leaves them with a net negative (whether it be performance, application incompatibility or simply the unfamiliarity of a new environment) in spite of any overall benefits to the organization.  2009 will therefore see virtualization platform vendors - VMWare, Citrix and Microsoft - addressing how they can provide the benefits of desktop virtualization (such as baseline standardization and security) while not forcing users into unwanted generic environments or increasing storage costs to the point that they outweigh the  benefits of virtualizing the desktop. 

While they will have to grapple with this as an ongoing issue, they will first have to respond to the problem of bringing the familiar to the new environment as part of the initial deployment. This is a specialized area of expertise that none of these vendors have much history with. Citrix historically is the antithesis of rich user experience and Microsoft has consistently stumbled with their ability to understand the importance of this issue with each new generation of operating system. Look for the big players to find and incorporate this expertise directly into their platforms. The journey to desktop virtualization will take many steps but if the new desktop cannot pass the Virtualization Turing Test it will stumble on the first step out of the gate.

About Gordon Riley

Gordon is the co-founder of Tranxition Corporation and has over 25 years experience in the IT industry in a variety of both technical and executive leadership roles.  He has worked with both enterprise and consumer focused software companies in North America, Europe and Asia.  Gordon has also served on industry standards committees such as the Desktop Management Task Force (now the Distributed Management Task Force) reflecting his ongoing interest in the evolution of the management and economics of desktop computing.

Published Wednesday, December 10, 2008 5:54 AM by David Marshall
Filed under:
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
<December 2008>