XP the faithful servant
XP has been a faithful servant of the corporate world. Most corporations have built a locked down image of XP to run on their desktops, and this is still widely regarded as the best way to manage a large estate without explosive support and testing costs. This has come at the price of speed and flexibility for the business and of varying suitability and usability for users, but this is a trade-off that the CIO (and his colleagues) have been prepared to make.
Now that XP is two generations old, MS can stop selling “downgrades” to XP, and there is a hard stop in 2014 when support will cease. No matter how well Windows7 may compare to XP, upgrading is a huge commitment of time and money with payback largely reliant on the development of new applications that it enables. So, many Corporates are caught in a trap, facing a forced global desktop upgrade to Windows7, which could require substantial investment in new more powerful hardware, but does little to reduce the support costs or show any upside to the business. But there is a better way…
The Real Price of Lock-down
Desktop strategy has been focused on achieving lock-down for years. We all know and understand the reasons why – control; security; manageability; simplified testing and reduced support costs.
The current Locked-Down Desktop paradigm aims for consistency across all desktops. To maintain it in the face of the forced move to Win7 the options are not attractive:
- Do nothing and risk running unsupported XP
- Pay Microsoft to support XP as a “special”
- Re-implement the estate to Windows7 with a 50% hardware refresh and no significant addressing of the users’ current requirements
The first wave of escapees
Since 2003 some companies have segmented their users and allowed those that require little or no access to transactional systems to self-manage their own machines, and to use the public networks. The benefits of this can be substantial, both from cheaper support to operations costs, and in terms of flexibility. However a parallel – although much cheaper – infrastructure has had to be built.
This hard segmentation, with a locked-down corporate infrastructure running next to one based on self-configured consumer technologies utilising public infrastructures, is an interesting step in the right direction. It demonstrated that there was a real value proposition but also the difficulty of defining the hard segmentation for individuals needing to occasionally dip into the corporate systems. The challenge remained of how to mix and match the benefits of cheaper public infrastructures and the benefits of self support and configuration while maintaining the controls of the lock-down.
The current Smart Moves
Technologies have continued to improve and communication speeds and coverage have expanded to the point where a new paradigm can be glimpsed. For example, a combination of virtualisation of the desktop, remote management and selective upgrades and desktop optimisation to avoid wholesale replacement could enable:
- Device independence, so an organisation is not tied to specific platforms;
- Smart ways of delivering services to new or upgraded end-devices; indeed, this may obviate the need to buy more powerful machines, or to upgrade altogether;
- Remote monitoring and management of the delivery of the service and of the devices despite their proliferation;
- Managed access to third party devices, providing flexibility in supporting contractors and temps without giving them a corporate PC.
The benefits for the IT organisation will be from being able to manage a single, centralised virtualised desktop (which is what the device lock-down was supposed to achieve). The benefits for users would be computing capability matched to role, potential freedom to select / configure devices and to self-support if preferred.
The way out of the XP trap
The real question is how to get there from here without getting stuck with the costs of each model and without unacceptable spend on the way.
A sensible plan of how to get there should cover these steps:
- First segment the users (casual, mobile workers, temps, contractors, core data entry etc) s0 that those who will benefit the most can be addressed first;
- Understand the needs of each segment and how the technologies can support them, recognising that this is not just a question of choosing the “next PC”, but potentially of reconfiguring work so as to take advantage of more innovative and attractive solutions;
- Review the infrastructure components and capacity – servers for virtualisation; networks to support virtual desktops and data volumes for Citrix; old-fashioned but important, as nothing will sink the plan more quickly than poor performance if a component is miss-sized;
- Understand what needs to be put in place to support each segment and plan on the upgrades;
- Have a user migration plan that dovetails with hardware end-of-life and prioritises the high value hits (e.g. contractors).