We all know Microsoft as the tech giant responsible for the Windows operating system, Office productivity suite and Xbox gaming console, but what you might not know is that the company is also a research and development powerhouse.
The Redmond, Washington-based company filed an array of new US patent applications recently. Among them are inventions for sensor-based access to the internet and a design for a wind-powered datacentre.
The former, officially "sensor-based authentication to a computer network-based service", stipulates that a person's mobile device — smartphone, tablet computer, you name it — can access a web-based service without requiring the user to type in a conventional textual password of letters and numbers.
Instead, "physiological biometric traits of a user" are used, such as a person's voice (via a microphone), face (using a camera), fingerprint (camera again) or even movement — eg, if a user moves his or her device in a certain way to gain access, through the use of accelerometers.
The idea: make passwords more human, and leave the random numbers and letters to the computer.
Meanwhile, Microsoft's wind-powered datacentre patent describes various techniques for powering a datacentre through the use of a wind-powered generator.
That sounds relatively pedestrian until you read the document, which suggests that servers be installed in the hollow tower of the wind turbine itself.
From there, it functions as you'd expect: wind blows, the turbine blades rotate, power is generated, network-connected servers are powered. Excess power is offloaded to alternate power sources and, when the wind doesn't blow, the servers are throttled or powered down.
It's an interesting approach. Datacentres are big, loud, sprawling energy hogs. Is there a better way to integrate them? There might just be. One hiccup: do we really need datacentres where we need wind turbines? And are existing transmission lines robust enough to handle this concept?