The first article I ever published in InfoWorld began, “Does anyone else find the Gartner Group annoying?”

While no longer a Group, the annoying part is alive and well, as evidenced by its recent prediction, reported in Network World, that, “Cloud computing will become so pervasive that by 2012, one out of five businesses will own no IT assets at all.”

To be fair, Network World’s headline mis-states Gartner’s prediction, as does Gartner’s lead paragraph, both of which emphasize virtualization and The Cloud.

What follows heads in a very different direction. Here’s what Gartner says:

The need for computing hardware, either in a data center or on an employee’s desk, will not go away. However, if the ownership of hardware shifts to third parties, then there will be major shifts throughout every facet of the IT hardware industry.

For example, enterprise IT budgets will either be shrunk or reallocated to more-strategic projects; enterprise IT staff will either be reduced or reskilled to meet new requirements, and/or hardware distribution will have to change radically to meet the requirements of the new IT hardware buying points.

In addition to having nothing at all to do with either virtualization or The Cloud, this is simply wrong. If ownership of the hardware shifts to third parties, nothing important will change, other than the company name on the asset tags and employee ID badges.

Unless that is, you think that leasing your car instead of owning it would change your driving habits and choices of destination.

Opinion: We should simply wave off this prediction as a childish need for attention.

But just for fun, let’s see if we can find a way it could happen in a meaningful way – for a fifth of all companies to own no IT assets by 2012 in a way that’s more interesting than that they lease them instead. Here’s what it would probably take:

Redefinition: “IT asset” can no longer include PCs, smartphones, or any other end-user technology, nor can it include on-premises communications gear such as PBXs, routers, and firewalls.

And “company” will have to exclude sole-proprietorships, whose entire IT infrastructure consists of laptops and smartphones, using lots of locally installed applications.

Creative destruction. Capitalism, like nature, is red in tooth and claw. Every year some companies go out of business while other new companies launch. If, by the start of 2012, enough creative destruction has happened, one in five companies could be less than two years old … young enough to have started life using nothing but Cloud technology and small enough to stay that way.

Point of diminishing returns: 20% of all companies decide they have everything they need in the way of information technology – no project backlog, no enhancement requests in the queue, nothing to keep developers occupied.

If redefinition and creative destruction don’t get us to the magic number, this would have to be the case. Even if “migrating existing applications to the cloud” means nothing more than re-hosting them using an infrastructure-as-a-service provider, it’s still a conversion – more so if it means redeveloping them using a platform-as-a-service provider or replacing them altogether through the magic of software as a service.

And as anyone knows who has participated in a conversion, it’s a non-trivial exercise … enough hard work, I’d bet, to fully absorb the entire available development capability of most organizations of any size.

For Windows or Linux-based applications the conversions might be merely very complicated. But how about companies that use … oh, I don’t know … perhaps IBM’s z/OS (the operating system formerly known as MVS), running COBOL/CICS applications? Just recreating the environment will take serious engineering, and that assumes you can find a reliable Cloud provider that sells z/OS in virtual form. Otherwise it’s conversion city.

Nor are the promised savings a sure thing.

Financially, The Cloud is like outsourcing. It converts CapEx spending (the cool way to say “capital”) into OpEx spending (operating costs, that is).

Somebody still has to buy the equipment, though. Cloud providers, like outsourcers, need discounts big enough to make a profit and still save customers money – likely when the customer is small and has little buying power; less likely for bigger companies that buy in enough volume to negotiate good prices for themselves.

My advice: Don’t bet the server farm.

Still, I’m confident enough to make two predictions of my own. The first: When 2012 arrives, Gartner will find a way to count that makes its prediction come true.

My second prediction is that the industry press will obediently cover the story.