Future of the Desktop
(Article taken from Management Guidelines 331 Future of the Desktop)
The desktop computing marketplace is fast-paced, with an increasing variety of options for the delivery of information both within an organisation and the wider connected community. In facilitating this information, the way in which applications are installed and accessed has become increasingly flexible, with Software-as-a-Service (SaaS) offerings a prime example.
Briefly, Saas is a form of software licensed per user, either via the vendor web server or on the user’s own hardware. The chief perceived benefits of this and other similar ‘virtualisation’ services are to reduce both the need for IT infrastructure by the enduser and capital expenditure on hardware and software.
The change in delivery of applications heralded by the advent of SaaS has had an inevitable influence on business expectations with regard to how new applications can be implemented. The challenge is to make desktop infrastructure fit for future business operations by making it capable of integrating these new applications efficiently.
This should not be exclusively focussed on SaaS, because in the majority of businesses not all applications will be delivered through this model. There will continue to be a need for IT organisations to maintain a combination of client server, web and even ‘green screen’ host based application access technologies on the desktop, alongside SaaS.
The picture is further complicated by factors including the need to protect from security threats, adapting to changing hardware specification standards, controlling access to problematic legacy or heritage applications and managing growing lists of applications that do a similar task. It is also important to manage the expectations of IT-literate employees and users who have access to the latest and greatest technology at home, but understand less the complexities of delivering IT as a service. In order to cater for so many complex yet influential factors, it is important to take a structured approach to the desktop and the business and IT environment in which it needs to operate. By balancing services and software as required, the goal is to get the best out of the available technology to support the business operation while making the IT organisation as effective and efficient as possible.
Supporting the unique requirements of each business and organisation is the key challenge. Rather than adopt a universal approach, IT strategy and architecture must be adapted and deployed in respect to individual requirements. For example, how can SaaS help to harmonise and streamline the particular desktop operations of the business in question?
When considering any new technology or a change in the way an IT service is delivered, the primary consideration should always be the benefit to the business. While many new technologies may seem to be exciting, these should be evaluated from a business perspective, otherwise how do you build the business case?
The nature of the business operation, the locations it operates from, the number of end-users it has, as well as the number and nature of the applications it uses should all be considered when forming an effective desktop IT strategy.
There are many examples, not just in the desktop or indeed the wider IT environment, where a technology has been selected without any practical consideration of how it aligns with the business operation. This can result in either the business operation having to change or the technology becoming marginalised in its use and reduced in its efficacy.
Developing a full understanding of the business operation as it stands, considered within the context of expected plans for the future, will help to provide an increased chance that the selected solution won’t be a poor investment.
Flexible technical solutions procured under flexible sourcing contracts are fundamental; however it is the way in which IT engages with the end-users that will drive the best ability for technology to fit business demands.
It is worth bearing in mind that building flexibility into contracts can be more expensive in the short term. Strategically it may be the most appropriate approach for the business to take, providing a longer-term plan is in place and that the business is ready to embrace change.
Alongside careful appraisal of internal business operations now and into the future, the wider IT landscape must also be taken into account and brought to bear on formation of a strategy.
Does the proposed plan facilitate integration with existing internal and external infrastructure? What potential changes are on the horizon that may affect the usability and compatibility of strategy and architecture?
Depending on the nature of the applications the organisation is using, the desktop normally fulfils two main roles: the provision of end-user productivity tools such as Microsoft Office; and access to business level applications and information.
A key principle with respect to the desktop device may be to keep the hardware and software components current and within manufacturer support. However, applications accessed via a web browser for example, may only be compatible with an older version of the browser. This highlights a key problem in that potentially the desktop could not be upgraded without first upgrading a core application, which may be hosted by an external organisation.
Introducing a major change, such as a shift to SaaS working, will undoubtedly have an impact on the requirements of the desktop. The desktop is made up of a combination of hardware and software components, the age and version of which will vary by organisation and probably within organisation. Understanding the mix of age, capability and version of desktop components, along with the policies for refresh and upgrade and the attendant financial considerations, is vital.
In terms of the wider IT landscape, a move away from a very distributed operation to a more centralised approach could result in the need to increase network capacity or raise availability service level requirements. Consequently, these would need to be factored in to any business case.
Careful appraisal of the current landscape is critical to achieving an effective IT strategy. For most organisations, the desktop device is a key component of many IT systems. As such, a change in the desktop provision could have an impact on other systems, so any changes to the desktop computing domain should be made while considering the IT landscape holistically. While there will always be a temptation to focus on the short to medium term, a desktop roadmap should also consider what the likely future developments are and what the current trends are with respect to this strategy. In doing so, heavy investment in technology with a limited lifespan can be avoided.
SaaS could be used to enrich the end user experience by providing them with access to a greater suite of applications and licensing costs could be reduced by moving towards a concurrent usage model, however regular users will continue to use locally stored applications until the supporting infrastructure is capable of delivering more services from central locations.
Desktop virtualisation services provide the ability to quickly provision a standard desktop and are a good fit for certain user profiles. However this is unlikely to be a full replacement for ‘thick client’ builds until the supporting infrastructure is improved. Examples of where this could be adopted in the short to medium term are for 3rd parties, contractors or partners where historically they may be provided with a corporate PC, but alternatively could be provided with access to virtualisation, thereby reducing the amount of physical hardware required.
The main advantage of this technology is that the organisation will not need to maintain client builds in diverse locations since the client builds will reside upon central servers. This provides cost saving opportunities, together with being able to more readily enforce the security policy, for example distribution of software patches. Further, with an increasing focus on business continuity virtualisation could be used in conjunction with a secure remote access system to provide the user with the corporate desktop experience from any PC. From a security perspective this is much more preferable to providing users with the ability to perform a remote take-over of their physical desktop since tighter control can be implemented on the inbound connections to an organisation.
However this solution relies upon a low latency transport network and since the client machines can no longer operate in an isolated manner, the network must also be reliable. Some functionality of the desktop may also be lost, for example two-way audio transmission, which is a key requirement for the deployment of softphones. Virtual desktop may also be better suited to certain user profiles, for example, making a ‘standard desktop’ available to partners or 3rd parties without having to give them a physical machine. Finally, caution must be taken before a wider adoption is considered since the service the transport network will be required to provide will fundamentally change.
Standard desktop builds or ‘gold builds’ have been around for many years, with this approach being largely adopted by large organisations when Windows NT 4 was the de facto standard operating system. Whilst these provide end-users with a standard environment that is generally reliable, frustration is often caused by the ‘one size fits all’ approach and also the lack of flexibility that limits the end-user to install or update any applications. Furthermore, applications are often included within the standard build that are not used by the entire user-base; this leads to poor value in terms of application licensing costs. Where applications are not included within the standard build, remote deployment can be time consuming and have an impact on other parts of the infrastructure, above all the network. Poor asset and software licence management can also result in users not being given back the required set of applications in the event of the end-user device being rebuilt; this can impact the productivity of the end-user.
Improved desktop management tools can address some of the issues that have historically plagued the use of standard builds. Many end-user device suppliers will install the build prior to delivery; this negates the need for costly staging areas, and also enables the equipment to be delivered directly to the end-user, thereby reducing lead times. There has been much debate, applicable to both the standard desktop model and also when using virtualisation, about the desktop operating system. While Windows XP has had security related issues in the past, it is now regarded as being stable and meets most organisations’ business requirements. There has been a limited uptake of Windows Vista, partly because of reduced support for legacy applications and also because it is difficult to build a business case around the migration from Windows XP. A common view is to wait until Windows 7 with Service Pack 1 is available before considering an upgrade from Windows XP. This is still unlikely to provide a noticeably improved experience to the end-user, however with the imminent withdrawal of support for XP by Microsoft, the benefit of change will largely be driven by mitigating risk to the business, by maintaining vendor support.
Changes within other technology domains, for example the introduction and adoption of next generation wide area networks will provide a more stable foundation for centralising resources, which is likely to lead to a greater uptake of technologies such as SaaS or virtualisation. Until this point it is unlikely that many large organisations will be able to build a sound business case around centralising the desktop resources, although we expect growth will continue for certain user groups.
Whilst Microsoft operating systems and office suite applications may remain the de facto choice for most organisations, notable developments of open source applications has led to low-cost viable alternatives that should be considered. This may not be appropriate for all types of user groups, however this choice would meet the functional requirement for many end-users.
You can purchase Guidelines 331: Future of the Desktop (pdf)
Qualifying Members can download it for free. Find out more about Membership.