Join Date: Oct 2005
Location: Chicagoland, once a year in Poland
Microsoft patents "pay-as-you-use" computer leasing
don't like the sound of this
Microsoft received a Christmas present from the USPTO this year in the form of a new patent on what the company describes as a type of pay-to-play computing. Although the specifics of the approved application are somewhat different from the ad-supported "free" PC business models of the late 1990s and early 2000s, a consideration of the merits of the patent identifies many of the same potential problems.
The patent abstract describes
a computer with scalable performance level components and selectable software and service options has a user interface that allows individual performance levels to be selected... Software and services may include word processing, email, browsing, database access, etc. To support a pay-per-use business model, each selectable item may have a cost associated with it, allowing a user to pay for the services actually selected and that presumably correspond to the task or tasks being performed. An administrator may use a similar user interface to set performance levels for each computer in a network, allowing performance and cost to be set according to a user's requirements.
We've seen pay-as-you-go systems before, particularly from certain big iron vendors, but Microsoft's now-patented nickel-and-dime system may run into many of the same snags that affected the "free PC" business. Skip back ten years, and the idea of handing over a no-cost PC system was an idea that drew considerable interest from press and consumers alike.
The concept worked as follows: instead of paying a large up-front cost, customers would sign a two-year ISP agreement at a rate of, say, $24.95 per month. In exchange for the system, lucky owners agreed to watch advertisements, receive special offers, or turn over their browsing histories. The terms varied, but at a time when users were rushing to get online and computers were still in the $700-$800 range, this was one heck of a deal—on paper.
In reality, the model proved unsustainable. The PCs in question were often slow, carried minimal amounts of RAM, and ISP service could be extremely spotty. As hungry as the US was for Internet access back then, the limitations and requirements of the systems turned off a number of potential buyers.
Microsoft's patent description makes a series of supportable claims, but the proposed solution is unattractive for a number of reasons. The problem, according to Redmond, is that consumers and businesses are forced to buy systems that can match their maximum potential performance needs, even if day-to-day usage requires only a fraction of that potential power. Similarly, expensive software packages, while necessary, may only be used intermittently or at certain times of the year. The company's goal is to provide an alternate solution to the inherent inefficiency of buying a system based on occasional need.
As the patent puts it:
A computer may have individually metered hardware and software components that a user can select and activate based on current need. Beyond simple activation, the user may be able to select a level of performance related to processor, memory, graphics power, etc. that is driven not by a lifetime maximum requirement, but rather by the need of the moment. When the need is browsing, a low level of performance may be used and when network-based interactive gaming is the need of the moment, the highest available performance may be made available to the user.
Microsoft presents the following example of how such a system might be priced. "The Office bundle may be $1.00 per hour, the Gaming bundle may be $1.25 per hour, and the Browsing bundle may be $0.80 per hour. Alternatively, a bundle may incur a one-time charge that is operable until changed or for a fixed usage period. Other pricing techniques are apparent."
The rest of the patent is devoted to describing the various ways in which fees could be implemented, adjusted, and metered. Redmond admits that the final cost of PC ownership could end up being higher than that of a one-time purchase, but notes that "the payments can be deferred and the user can extend the useful life of the computer beyond that of the one-time purchase machine."
Maybe I'm just too old-fashioned, but if there's an upside to this, it seems quite limited. This type of approach to computer billing might theoretically allow a company to directly tailor system capabilities to user needs, but it does so by requiring a huge amount of additional logistical complexity. Any company wanting to adopt this sort of system would first have to conduct what amounts to a desk-by-desk survey of exactly what company employees do and how they do it. Usage models must then be mapped and modified against pricing structures, which could potentially create absurd operating environments.
Higher-performing systems would still cost more, but it's safe to assume that an employee's ability to operate in peak performance mode would be closely monitored. Any pressure from the accounting department to cut costs will land squarely on the shoulders of professionals who legitimately require high-performance systems in order to do their jobs effectively. ("Bill, this is Greg from accounting. Any chance you can stop using your graphics card for a week to save us a bit of cash?")
The nickel-and-dime approach might have worked when even basic computers were $1,500+, but Dell has Vostro systems for as little as $289. Even if we assume a pricing plan whereupon a computer using Microsoft's proposed payment system costs just $1.00 per day for 350 days a year, the machine ends up costing more than a base PC in only months. Such a setup would allow gamers to get more horsepower than they could otherwise afford up front, but allowing such systems into homes opens the obvious risk of hacking and cracking the devices.
At this point, the market has created computers at almost every price point imaginable—it's hard to imagine how any vendor could simultaneously entice a company with this type of offer and make itself a profit. Startups with virtually no capital might be interested, but these probably aren't the sorts of companies an OEM wants to build a business segment on.
Windows 8 the next big failure, right after Windows ME