Ever since the advent of the desktop operating system in the late eighties, enterprises have been stuck in a four to six-year cycle of major desktop refresh project work. These huge projects often involved a complicated mix of infrastructure change, hardware refresh, application repackaging/replacement, and the introduction of new management tools. As Microsoft continued down the path of semi-regular, largely incompatible major releases, the work involved in these upgrades, especially if you had thousands of machines, has been a long, complicated and ultimately expensive experience.
Now, and arguably for the first time, Microsoft is about to release an enterprise friendly major version that could change the way we look at enterprise desktop upgrade projects. In a recent blog post, Microsoft’s Jim Alkove started to talk about how the team behind Windows 10 is approaching the subject of in-place upgrading. Could we really be looking at Windows 7/8 being our last ever desktop transformation, and what does that mean for maintaining an evergreen estate?
Let’s start by looking at the information we have so far on the new release…
- Windows 10 will run on the same hardware as Windows 7 … check!
- Windows 10 will be compatible with applications running on Windows 7 and 8 … check!
- Windows 10 will support the same peripheral set as Windows 7 and 8 … check!
- Windows 10 will support in-place upgrades from Windows 7 or 8.1 … check!
So far, so good. But is all of this just a pipe dream for our enterprise desktop management team? Can we really just upgrade all of our machines to the new version and expect everything to work?
So why might we still need desktop refresh projects?
Browser versions embedded within the OS continue to change. Whilst your OS based applications may be compatible, there is no guarantee that your browser based apps will still work.
Is it really likely that all of your applications that run on Windows 7 will run seamlessly on Windows 10? I’m not a betting man, but if I were, I’d like to see some real world statistics from our friends at Citrix (AppDNA), Dell (Changebase) and Flexera as to how many of those in-house written applications will just transfer over. Even a single application failure would mean a piece of project work.
Not all software vendors will jump to support the new operating system. It is highly likely that vendors will continue to only support later versions of the OS with upgraded product sets. This forces our upgrade cycle for their applications and makes co-existence without a proper project a tough nut to crack.
Typically, organisations refresh their hardware based on an asset depreciation or a lease cycle. Whilst the principle behind sweating the asset is sound, the reality is that we don’t want to lose the budget, have to subsequently find the budget, or run on years old hardware when there are faster, power efficient machines out there that could improve our productivity.
The discussion brings into focus the question of evergreen IT – the principle that desktop, OS and application upgrades simply become part of your business as usual operations. How close we are to achieving this I can’t say, but we will certainly be exploring this concept in future blog entries.
For now, as we move forward into the brave new world of Windows 10, we will soon discover whether the reality of desktop migration is still far removed from the hype of simple, in-place, evergreen upgrades.