For the first time since computers became a normal part of office life, end users won't be able to predict what their "PC" will look like in 10 years. That's partially due to the expansion of IT-as-a-service technologies that are making it possible to give users secure, reliable access to data and applications no matter where they are or what device they're using.
But it's mostly due to increases in the power, connectivity, ease of use and stylishness of a whole range of nontraditional computing devices -- primarily smartphones and tablets -- and to heavily networked applications such as social networking, software as a service and cloud computing services that are easy to get, easy to use and often free to consumers.
Clouds, virtual servers and SaaS are the kinds of infrastructure technologies that would normally be invisible to end users.
Five years ago, few commercial applications were available by subscription; now it's surprising when one isn't.
End users have been trying to consumerize IT since the 1980s. They started by sneaking in PCs, then they surreptitiously set up LANs and later brought in laptops, PDAs, cell phones, BlackBerries, tablets and other gadgets that IT either couldn't or wouldn't support.
When users can buy sophisticated data services to support not only gadgets, but also applications, the economics of the IT vendor's business change -- as do the role and goals of IT.
Rather than being "just say no" organizations intent on standardizing and cost-cutting, IT shops have had to start collaborating with end users, who want to choose the devices they use at work, says Dave Bucholz, principal IT engineer in charge of evaluating new end-user technologies at Intel.
Intel -- like Kraft, CarFax and a range of other companies -- has adopted a bring-your-own-device policy as a way to match the workstyles of individual end users with the devices that enable them to be the most productive.
The freedom to choose your own device, one that suits both your work and personal style, will have as much impact on "PC" design during the next 10 years as Moore's Law -- or, even more important, Metcalfe's .
What those devices look like depends on the needs of the people using them. The same is true of winter coats: Computing devices and coats both supply a non-negotiable set of key functions. But there are a million variations on the basic coat to meet the needs of people in climates with snow, sleet, cold sunshine and warm rain; it would be impossible to identify only one garment as the definitive "coat."
In 10 years, a "personal computer" will be the totality of an end user's computing environment -- including the applications he or she uses both at work and at home, personal data and work data, network access, security and identity management. But it won't be a box sitting on a desk.
As our knowledge and access increase, the value of being more connected than the next guy declines, and the value of being able to do more with the information we all have becomes greater.
That may drive the next big wave of IT development, in artificial intelligence or quantum computing or other disciplines that promise to deliver not just data, but knowledge.
The look of the machines we use to access that knowledge will likely vary from user to user. Most people will probably call them phones.
Kevin Fogarty writes about enterprise IT. Follow him on Twitter ( @KevinFogarty ).
Read more about data center in Computerworld's Data Center Topic Center.