Don’t let Microsoft timescales dictate your Windows migration
Move at a pace that is comfortable for you
There is a lot of talk at the moment about desktop migration schedules. With the majority of enterprises still managing XP estates, the big question is whether to make a short term move to Vista, hang on and wait for Windows 7, or dig in and not think about it until you really have to.
A minority - typically represented by small tech-savvy professional services firms in the IT, engineering and media sectors, in particular - will take the uncertainty as a cue to investigate alternatives such as Linux and Mac OS X. This is a perfectly legitimate thing to do, but outside the scope of this article.
For most, the key question remains about when and how to move forward in terms of Windows releases. Having looked at the options, there isn’t a single ‘right’ answer to this that applies to every organisation. So what’s the best way to think about the issue and come up an appropriate plan?
A couple of things are worth bearing in mind as a foundation for the discussion.
First: the physical deployment of the operating system itself, i.e. getting the new version of Windows onto target machines, can involve significant operational effort, but the real cost and pain of most previous Windows migrations has been associated with testing and remediating applications, utilities, drivers and so on. We know from those who have already made the move from XP to Vista that this is not a trivial task in that particular case.
Real compatibility issues are typically encountered with both commercial and in-house developed software that need to be identified, understood and fixed methodically. tools exist that can help with the process, but there is no getting away from the need to treat the exercise as a properly planned, resourced and funded project if you have anything more than a handful of PCs.
Crafting a grafting
Second: from what we have seen and heard so far with the beta version, Windows 7 is largely an optimisation of Vista rather than a generational leap forward. I am sure that Microsoft and others might debate this throwaway comment from various different angles, but the pertinent point tis that software designed to run on Vista will very likely run on Windows 7 with no compatibility issues.
Microsoft’s claims here seem to be corroborated by feedback from beta testers. Indeed, it seems every man and his dog has now given it a whirl and is reporting positive experiences in developer communities and social media circles.
The upshot is that the bulk of the effort put into the ‘hard bit’ of an XP to Vista migration i.e. the remedial compatibility work, will not need to be repeated during a subsequent move from Vista to Windows 7. Of course it’s unrealistic to expect perfection, and you would be mad to assume compatibility without going through a full re-test of everything. But the chances are that most things will pass and need no further action, making the whole thing much less costly, time-consuming and risky.
So how does this help when considering migration strategy and timing? Let’s not pretend the situation is ideal, but at least it puts control of the decision back in your hands. With early experiences and Microsoft promises indicating a relatively small difference in effort between a two step (via Vista) and one step (skipping it) migration, the consequences of choosing one route or the other are not that great.
So it’s up to you, really. If it suits you to move sooner rather than later, then there is no need to hold back from implementing Vista for fear of having to duplicate a huge amount of effort. Yes, there will be additional work involved in the two-step journey you will ultimately be taking, but the second refresh will be much less daunting.
If you plan to wait for Windows 7 before rolling out your next major upgrade, you can begin testing and remedial work in advance using Vista. Having said this, the stability of the current beta bodes well for the future, so working Windows 7 directly into your test and preparation activities may be possible sooner than we might expect, based on past experiences with new Windows releases.
The last question that remains is how to decide if and when to move forward. Critics maintain that there is simply not a business case to migrate from Windows XP, so why bother at all?
I would not dispute the difficulty in justifying a Vista rollout at from an end user value perspective, although larger organisations that have taken Vista fully on board report significant payback from operational improvements. I personally haven’t formed a strong view on whether Windows 7 will change the cost/value equation very much, though my own experimentation with the beta (admittedly just a single data point) has revealed a noticeably better user experience that could arguably make a difference to productivity - it’s too early to make a call on this, though.
But the cost/benefit calculation associated with the operating system per se is not the main issue. After all, the only reason Windows is there is to run applications. And with the relentless upgrade spiral among software vendors, practicality says you can’t fall too far behind without running into support and compatibility problems at the other end of the conveyor belt.
There’ll come a time when the latest releases of solutions that are important for your business will no longer run on XP, in the same way that vendors have gradually dropped support for pre-XP versions of Windows. So unless you are one of those looking to defect to Linux or OS X, the move from XP is inevitable at some point down the line. And it’s always better to act proactively than reactively.
So, whether you are an enthusiastic pioneer or a grudging pragmatist, the good thing is that you can largely set your own pace in terms of migration timing, which is exactly as it should be.
Dale Vile is research director at industry analyst firm Freeform Dynamics.
Sponsored: Benefits from the lessons learned in HPC